Keep it Simple: 1.e4: A Solid and Straightforward Chess Opening Repertoire for White 9056918052, 9789056918057

Citation preview

Keep It Simple: 1.e4

1

Keep it Simple

Christ of Sielecki

Keep It Simple: 1.e4 A Solid and Straightforward Chess Opening Repertoire for White

New In Chess 2018

2

Copyright Page

© November 2018 New In Chess Published by New In Chess, Alkmaar, The Netherlands www.newinchess.com Previously published (March 2018) as an online, interactive book on www.chessable.com All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission from the publisher. Cover design: Nilanga Fernando & Volken Beck Supervision: Peter Boel Proofreading: Chris Tilling, Sandra Keetman Editing and typesetting: Peter Boel Production: Anton Schermer Have you found any errors in this book? Please send your remarks to [email protected]. We will collect all relevant corrections on the Errata page of our website www.newinchess.com and implement them in a possible next edition. ISBN: 978-90-5691-805-7

3

Keep it Simple

Contents Preface Part I Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Chapter 8

Open Games: 1.e4 e5 Introduction 2.♘f3: rare lines Philidor’s Defence: 2.♘f3 d6 Petro Defence: 2.♘f3 ♘f6 Three Knights Game: 2.♘f3 ♘c6 3.♘c3 Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 4…exd4 5.♘xd4 4…exd4 5.♘xd4 ♗b4 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3 Sample games

Part II

Sicilian Defence: 1.e4 c5 Introduction Chapter 9 Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines Chapter 10 Rossolimo with 3…g6 Chapter 11 Rossolimo with 3…e6 Chapter 12 Rossolimo with 3…d6 Chapter 13 Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7 Chapter 14 Moscow Variation with 3…♗d7 Chapter 15 2.♘f3 e6 3.c3 Chapter 16 2.♘f3: other lines Sample games Part III

Caro-Kann Defence: 1.e4 c6 Introduction Chapter 17 2.♘c3: sidelines Chapter 18 Two Knights: 2.♘c3 d5 3.♘f3 dxe4 Chapter 19 Two Knights with 3…♗g4 Chapter 20 Two Knights with 3…♘f6 4

Contents

Sample games Part IV

French Defence: 1.e4 e6 Introduction Chapter 21 Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines Chapter 22 Exchange Variation with 4…♗d6 Chapter 23 Exchange Variation with 4…♘f6 Chapter 24 Exchange Variation with 4…♗g4/4…♘c6 Sample games Part V

1.e4: other moves Introduction Chapter 25 Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6 Chapter 26 Modern Defence: 1…g6 Chapter 27 Philidor and rare lines after 1…d6 Chapter 28 Scandinavian: 1…d5 2.exd5, rare lines Chapter 29 Scandinavian: 2…♕xd5 3.♘c3 ♕d8 Chapter 30 Scandinavian: 2…♕xd5 3.♘c3 ♕a5 Chapter 31 Alekhine’s Defence: 1…♘f6 Chapter 32 1.e4: rare lines Sample games Index of players Explanation of symbols Bibliography Author’s biography

5

Preface

level correspondence chess speaks for itself.

Preface Don’t you hate it when you are at a chess tournament, right before the rst round, and some local important person is going on about all kinds of things? You are eager to start playing, moving the pieces, getting into a struggle over the board – and someone is stopping you! I have to confess that I feel similarly with book prefaces. Usually I just skip them and dive right into the moves of the chapters I am most interested in, hungry for ‘real’ information – not some chit-chat. Therefore this won’t be a long preface. The main point of it is to tell you the idea behind the ‘Keep It Simple’ approach. In fact, ‘Keep It Simple’ has been my motto in chess and in particular as a chess instructor for a while, so writing and naming a book after it is a very logical thing to do.

My main ‘KIS’ guidelines are: • The chosen lines are simple to learn • It must be possible to nd your way if you forget your lines • Choose lines that may not be most critical, but uncomfortable for the opponent.

So what is this repertoire concept all about? Most opening books nowadays have reached enormous complexity, often spanning several volumes, totaling 1000+ pages. This depth of analysis is useful for very strong players, but not so much for amateur players. I felt it must be possible to ‘Keep It Simple’ instead, being more practical than scienti c in the choices. Against the main-line defences White does not get an advantage anyway – for example the drawing percentage in top

Compiling the repertoire according to these guidelines has led to a very classical, sound and reliable repertoire. You will get good endgames frequently, or positions with long-term assets like the pair of bishops. You won’t get unsound gambits or tricky lines that can be refuted – all lines are playable up to a very high level; in fact most of the repertoire has been played by top players, at least occasionally or in quicker time controls. The ‘KIS’ approach is of course applicable to many openings, so why is this a 1.e4 based White repertoire in particular? The reason is mainly my job as a chess teacher. Over the years I met many students or junior players that I coached at tournaments – and lots of them play 1.e4, of course. Whenever they asked for general advice or needed a quick preparation session before a game I was less of a help because I’ve played non-1.e4- openings all my life with white. This book project helped to x this. While researching for the book I got an excellent general overview on 1.e4 as a whole, trying to nd good lines that tted the ‘KIS’ approach.

7

8

Preface

I don’t know if you already play 1.e4, or if you try to use this book to add 1.e4 to your opening repertoire. Before this book project I had not played 1.e4 seriously in tournament chess for ages. Since starting with it I have played 1.e4 quite succesfully in all time controls. I feel the ‘KIS’ approach makes it a bit easier to adopt 1.e4, compared to more complex repertoires.

Christof Sielecki Dinslaken, Germany September 2018

The nal important point I need to mention is that ‘Keep It Simple: 1.e4’ was developed rst as an online, interactive book on the platform www.chessable.com. On the website you can learn the moves based on text and videos. This product was released in March 2018. The paper book you are now reading has the same content, only adjusted to t the format better. Based on the feedback of Chessable users I have added some lines that were played right after the release or that I simply forgot initially – having hundreds of readers before the paper book goes to print is excellent quality assurance! The only substantial di erence between the Chessable version and this paper book release is the add-on of 30 instructional games for this book, illustrating typical middlegames and ideas for the selected lines. Somehow I still managed to make this intro too long, despite all my good intentions at the beginning! So just one more thing. Always remember: ‘Keep It Simple for you, make it di cult for your opponents!’ 9

10

Introduction

Let’s turn to the more popular and better moves.

PART I

Open Games: 1.e4 e5

Chapter 2: 2…d6

Introduction

Chapter 1: 2.♘f3 – rare lines The classical reply to 1.e4, 1…e5, is popular on all levels, from junior events up to World Championship matches. You will face it very frequently, and being equipped with a good repertoire against it is going to translate into many points scored. The following chapters will provide you with all you need to know to face 1…e5 with con dence. 2.♘f3 After 2.♘f3 Black only has three respectable moves available. These are 2…♘c6, 2…♘f6 (the Petro or Russian Defence) and 2…d6 (Philidor’s Defence). Other moves are mostly unsound gambits or they violate some other opening principle. We should get at least a solid edge against those rare lines. Black’s rare lines are examined in Chapter 1. 11

This is Philidor’s Defence. It’s not very popular, but actually not that bad if Black knows it well. There are many bad lines as well, and these are quite often seen at the amateur level. It’s worth checking them as you have chances to score quickly against these inferior lines. Our choice against the Philidor is 3.d4, putting more pressure on Black’s centre. For details please see Chapter 2. Chapter 3: 2…♘f6

12

Introduction

The Petro has a deservedly solid reputation. Black is content with a slightly passive, but tough-to-crack position. I recommend to play 3.♘c3. This is a very convenient choice as we have the Scotch Four Knights in our repertoire. Now Black’s most popular move is actually 3…♘c6, transposing to the Four Knights immediately. We will examine the Petro in Chapter 3. Chapter 4: 2…♘c6

My repertoire suggestion now is 3.♘c3, aiming for the Scotch Four Knights Opening examined in Chapters 5-8. However, Black sometimes avoids 3… ♘f6 and goes for an o beat option. These are all quite welcome, because White will get an edge with good play. Chapter 4 examines Black’s alternatives on move 3.

Here we go 4.d4. This is the Scotch Four Knights Opening. This line enjoys a very good reputation as it features quick development for White and a very sound pawn structure. It is easier to learn and to understand than the main alternatives, the Italian (Giuoco Piano) and Spanish (Ruy Lopez) Opening. Black’s most popular and obvious reply is to take on d4, but there are alternatives, mostly 4…♗b4, that are examined in Chapter 5. Chapter 6: 2…♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4

Chapter 5: 2…♘c6 3.♘c3 ♘f6

Here the overwhelming main line for Black is 5…♗b4 (Chapters 7 and 8), but

13

14

Introduction

there are alternatives that we will check in Chapter 6. Chapter 7: 2…♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4 ♗b4

The move that has brought new attention to the Scotch Four Knights. It was basically unknown until Kramnik played it against Aronian in 2012. After that game, hundreds of games have been played, and the conclusion is that White has a low-risk game with chances to play for a small edge. Chapter 8 discusses the possibilities after 10.h3.

Black’s pinning move is the best and most popular option. Most of the time you will reach the main line starting position of Chapter 8, but of course there are some possible deviations along the way. All these are found in Chapter 7. Chapter 8: 2…♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4 ♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

15

16

Chapter 1: 2.♘f3: rare lines

CHAPTER 1

2.♘f3: rare lines 1.e4 e5 2.♘f3

After 2.♘f3 the most common move is 2…♘c6, but Black may avoid it with the respectable choices of 2…♘f6, the Petro or Russian Defence, or 2…d6, the Philidor Defence. Other moves are mostly unsound gambits, or they violate some other opening principle. We should at least get a solid edge against those rare lines. Let’s check the rare moves one by one.

The Latvian Gambit basically loses for Black, but it is notoriously murky, so you still need to be precise, especially in rapid/blitz. 3.♘xe5 ♕f6 3…♘c6 4.d4!. This is good to remember. I had this 3…♘c6 once in a simul game and was not then aware of 4.d4. Other moves look promising, like 4.♕h5+, but they don’t net that much – surprisingly! After 4.d4 Black lacks a decent move. 4.d4 4.♘c4 is also much better for White, but our choice is simpler to play. 4…d6 5.♘c4 fxe4 6.♘c3 ♕g6

A) 2…f5?

7.f3! 17

18

Chapter 1: 2.♘f3: rare lines

This is important. The best approach against the Latvian, or any other shady gambit, is to focus on your own attack rather than grabbing material. Here White opens up everything while Black is underdeveloped. 7…exf3 7…♘f6 8.fxe4 ♘xe4 9.♗d3 and Black loses material. 8.♕xf3 And White has all the fun with the better development and the open lines to use it.

development and f6 sticks out) 5.d3 dxe4 6.dxe4 ♕xe4+ 7.♗e2.

B) 2…f6?

In this wide open position White’s development advantage and the black weaknesses will mean a tough defence for Black. 4.♕h5+ ♔e7 4…g6 5.♕xe5+ and we win the rook in the corner. 5.♕xe5+ ♔f7 6.♗c4+ d5 This makes it a bit more di cult to win. After 6…♔g6 the computer already sees a forced mate: 7.♕f5+ ♔h6 8.d4+ g5 9.h4.

This is often called the ‘Damiano Defence’, but not because he thought it was any good. Damiano in fact analysed its defects already back in the 16th century. 3.♘xe5! Just refuting it right away. 3…fxe5 3…♕e7 avoids the quick disaster of the capture on e5, but Black’s position is already severely damaged, of course: 4.♘f3 d5 (instead 4…♕xe4 leads to a similar position type. Black has no

19

And that’s enough to know. If you don’t manage mate in 9 like the comp and take longer… who cares?! 20

Chapter 1: 2.♘f3: rare lines

7.♗xd5+ ♔g6 8.h4 This is the most convincing. White threatens to take on b7 now, but also has ♕g3+ available. 8…h6 8…h5 9.♗xb7 is basically the same as 8…h6.

9.♗xb7! ♗d6 10.♕a5 And we win the rook as Black cannot take the bishop. C) 2…d5?

The Elephant Gambit. The only good thing about it is the funny name! It just doesn’t work. 3.exd5 e4

21

After 3…♗d6 4.d4 e4 5.♘e5 White is a pawn up for nothing. 4.♕e2! This is important to remember. 4…♘f6 5.d3 ♕xd5

6.♘bd2! Avoiding a pin after 6.♘c3?! ♗b4!. 6…♘c6 7.dxe4 ♕h5 8.♕b5 This is a ‘Keep It Simple’ approach – White may try to punish Black harder, but 8.♕b5 secures a clean advantage. 8…♗c5 9.♘b3 ♘xe4 10.♗e3 ♗b4+ 11.c3 ♕xb5 12.♗xb5 ♗d6 13.♘fd4

And Black’s pawn structure gets destroyed. White is better and Black has no active play. 22

Chapter 1: 2.♘f3: rare lines

The rare lines in this chapter are rare for a reason: they are mostly bad! I have provided some easy-to-learn antidotes against the gambits and oddities that people might surprise you with. Amongst the lines here, 2…♕e7 is actually not as bad as it looks. You’ll get a better version of a Closed Philidor, which is favourable of course, but you are not winning straight away.

D) 2…♕e7?!

Well, this can’t be great, but it is not completely terrible either. Black intends a later d6/c6 set-up, much like in the closed Philidor examined in Chapter 27. He loses some time with the queen obviously. 3.♘c3 c6 4.d4 d6 5.♗c4 h6 6.0-0 ♘f6 7.a4

We just play like we do against the Closed Philidor. It promises a slight advantage for White. Please see this chapter for more details and make sure to study the sample games as well. Conclusion

23

24

Chapter 2: Philidor’s Defence: 2.♘f3 d6

CHAPTER 2

Philidor’s Defence: 2.♘f3 d6 1.e4 e5 2.♘f3 d6

3.d4 The best move, attacking Black’s centre. Now there is only one fully sound move for Black – to take on d4. However the bad alternatives are still played frequently, so studying them makes a lot of sense.

This is very popular at club level, but quite bad. 4.♗c4! Very important! Instead, 4.♘c3 ♘gf6 would transpose to the Closed Variation of the Philidor, which nowadays arises from 1.e4 d6 2.d4 ♘f6 3.♘c3 e5 4.♘f3 ♘bd7. This is, of course, still a bit better for White and part of our repertoire (see Chapter 27) via this move order, but we should exploit Black’s bad move order via 1… e5. A1) 4…♗e7? This is a direct blunder. 5.dxe5! ♘xe5 5…dxe5? 6.♕d5. That escalated quickly! 6.♘xe5 dxe5

A) 3…♘d7?!

7.♕h5 And that’s it for the important e5pawn. 7…g6 8.♕xe5 ♘f6 9.♗h6 And Black is a pawn down and his king is stuck in the centre. A2) 4…h6? This also loses immediately. 25

26

Chapter 2: Philidor’s Defence: 2.♘f3 d6

5.dxe5 dxe5 White also wins after 5…♘xe5 6.♘xe5 dxe5 7.♗xf7+.

hold e5, but it is the only move to avoid immediate trouble. 5.♕xd4 Now the queen is safe on d4, with the knight already on d7. 5…♘gf6 6.♘c3 ♗e7 7.♗f4 And White is better, with both castling options being possible. A4) 4…♘gf6?

6.♗xf7+! This will be a fun game now… 6…♔xf7 7.♘xe5+ ♔f6 Or 7…♔e7 8.♘g6+. The move 4…h6 weakened an important square… 8.♘c3! Instead, 8.♕d4 should win as well, but 8.♘c3 is completely killing. 8…♗b4 Some other ways to end it are: A) 8…♘e7 9.♕f3+ ♔xe5 10.♘b5 (the most forcing move) 10…♘f6 11.♘xc7 and that’s that!; B) 8…♔xe5 9.♕d5+ ♔f6 10.♕f5+ ♔e7 11.♘d5+ ♔d6 12.♗f4+ ♔c6 13.♕e6+ ♗d6 14.♘b4+ ♔b6 15.♗xd6 and again Black loses quickly. 9.♕d4 ♗xc3+ 10.bxc3 ♕e8 11.♘xd7+ White is getting the material back with interest – and the attack remains. A3) 4…exd4 This does not t with 3…♘d7 at all, which signalled Black’s intention to

27

5.dxe5! 5.♘g5 looks attractive as well, but is not as good due to 5…d5 and Black is worse, but still posing problems. 5…♘xe5 Obviously even worse is 5…dxe5? 6.♘g5. We played 4.♗c4 for a reason… 6.♘xe5 dxe5 7.♗xf7+ ♔xf7 8.♕xd8 ♗b4+ 9.♕d2 ♗xd2+ 10.♘xd2 With a clean extra pawn and a weak black e5-pawn. A5) 4…c6 Most popular, but following this Black still struggles. 5.0-0 ♗e7 5…exd4 6.♕xd4 is simply better for White due to the quick pressure on d6 (♗f4, ♖d1). 28

Chapter 2: Philidor’s Defence: 2.♘f3 d6

6.dxe5! Usually, such a trade helps Black, as d4 is a better pawn than d6, but here there is a concrete idea. 6…dxe5 A common motif emerges after 6… ♘xe5 7.♘xe5 dxe5 8.♕h5 and we win the centre pawn, just like after 4…♗e7.

7.♘g5! This is the important point to remember! A51) 7…♘h6?

9…gxh6? even leads to a mate after 10.♕h5+ ♔f8 11.♗xe6 ♕e8 12.♕xh6#. 10.♕h5+ ♔f8 Now two pieces are hanging. Has White gone wrong? Note that 10…g6 11.♕e2 is a disaster as well. Black’s structure is terrible and the king remains unsafe in the centre. 11.f4! And this attack wins quickly. A52) 7…♗xg5 8.♕h5 With a double attack on e5 and f7, thus regaining the piece. 8…♕e7 9.♕xg5! This is a lot better than 9.♗xg5, as then Black gets 9…♘gf6 with a tempo on the queen and can regroup with …♘f8 quickly. 9…♕xg5 10.♗xg5 ♘gf6 11.f3 With a great position for White. The pair of bishops and the weak dark squares are great assets. He will play moves like ♗e3, ♗e2, ♘d2, a2-a4, ♘c4 etc. Sample game 1, Vescovi-Pelikian, Sao Paulo 2001, shows how White will exploit those advantages. B) 3…♘f6?! This is played to reach the Closed Philidor after 4.♘c3 ♘bd7, but White has something better. 4.dxe5 ♘xe4

8.♘e6! Easy to miss – now Black is lost. 8…fxe6 9.♗xh6 ♘b6

29

30

Chapter 2: Philidor’s Defence: 2.♘f3 d6

5.♕d5! This is quite annoying for Black. 5…♘c5 6.♗g5 ♗e7 It’s no improvement to play 6…♕d7 7.exd6 ♗xd6 8.♘c3 0-0 9.0-0-0, leading to a wide open position where White is better mobilized and coordinated. Black is struggling here. 7.exd6 ♕xd6 8.♘c3 c6 9.♕xd6 ♗xd6 10.0-0-0 White is much better developed, with chances to attack the uncastled king, which is possible even with the queens already traded.

4.♗c4! White is more or less winning already. 4…exd4 1) 4…fxe4 5.♘xe5! dxe5 6.♕h5+ ♔d7 7.♕f5+ ♔c6 8.♕xe5 and White wins easily; 2) 4…♘c6 5.♘g5 ♘h6 6.d5! (this is a bit counter-intuitive, as the bishop’s diagonal is closed – but it is very strong) 6…♘e7 7.♘c3 g6 8.h4 and Black is under heavy re without any counterplay. 5.♘g5! ♘h6 6.0-0 White will get a strong attack with ♖e1 etc. that should win. No need to go further as this line is extremely rare and you will nd good moves over the board. D) 3…♗g4?!

C) 3…f5?

This strike was Philidor’s original idea. It does not work though. 31

Well known from Morphy’s Opera game… and bad. 4.dxe5 ♗xf3 With 4…♘c6 Black gives up the pawn, but it is not enough: 5.exd6 ♗xd6 6.h3 ♗h5 7.♗b5 and White is a pawn up for nothing much. 5.♕xf3 dxe5 6.♗c4 ♕f6

32

Chapter 2: Philidor’s Defence: 2.♘f3 d6

Morphy’s famous ‘Opera Game’ against the Duke of Brunswick and Count Isouard (Paris 1858) went 6…♘f6!? 7.♕b3! and White was already winning material. 7.♕b3 b6 8.♘c3 And White is close to winning already. 8…c6

This is the only respectable move for Black. He temporarily gives up the centre in order to quickly castle with … ♘f6/…♗e7/…0-0 and then try to strike with …d6-d5 at White’s centre, in most cases. 4.♘xd4 ♘f6 Instead, 4…g6 will almost always transpose to 4…♘f6 5.♘c3 g6, as omitting …♘f6 makes little sense: 5.♘c3 ♗g7 6.♗e3 ♘e7 (6…♘f6 is better, and transposes, as mentioned) 7.♕d2 ♘bc6 8.0-0-0 0-0 9.h4 with a simple and strong game for White. 5.♘c3 F1) 5…g6

9.♘d5! cxd5 10.♗b5+ ♔e7 11.♕xd5 And White has a decisive attack. E) 3…♘c6

4.♘c3 This we examine via 2…♘c6 3.♘c3 d6 4.d4 – White is comfortably better. F) 3…exd4! 33

This was a pet line of Danish legend Bent Larsen, but it has a somewhat shady reputation today. White’s attacking formation, reminiscent of the Sicilian Dragon, is very easy to play. 6.♗e3 ♗g7 7.♕d2 0-0 8.0-0-0 Compared to the Sicilian Dragon, Black lacks the open c- le for counterplay. It’s also not easy to get the queenside pawns involved. Moving the c-pawn, for instance, would weaken d6 substantially. 34

Chapter 2: Philidor’s Defence: 2.♘f3 d6

8…♘c6 Black may also try 8…♖e8 9.f3 a6 for some play with the pawns on the queenside: 10.g4 (if Black does not do anything soon, he will be steamrolled on the kingside with a standard pawn storm) 10…b5 11.♗h6 b4 12.♗xg7 ♔xg7 13.♘ce2 with h2-h4/♘g3 to come. Black is in terrible shape. 9.f3 ♘xd4 10.♗xd4 ♗e6 11.g4 c5 12.♗e3 ♕a5

13.♗h6 White is much better here. He is better in the centre and has attacking prospects on the kingside. What does Black have? F2) 5…♘c6 6.♗b5 This we cover via the Four Knights move order (Chapter 6). F3) 5…♗e7

6.g3 It is surprisingly di cult to prove something against 5…♗e7. In fact, some natural tries lead to absolutely nothing; for example, 6.♗e3 0-0 7.♕d2 d5! gives Black a good game. A closer inspection hinted that 6.g3 might be the most promising. I was a bit reluctant to recommend this, to be honest. We almost never anchetto in our repertoire, so it seems a bit inconsistent – but it’s the best move, it seems. White immediately tries to increase his control over d5 – this is very logical as Black’s play is based on a quick …d6-d5 in most cases. I’d like to mention that 6.♗f4!? is a very sharp alternative, but it had too many crazy lines for the purpose of this repertoire. It’s worth checking out if you like to analyse sharp complications. F31) 6…0-0

35

36

Chapter 2: Philidor’s Defence: 2.♘f3 d6

This solid line is probably best for Black, but natural moves give us a slight pull. 7.♗g2 ♖e8 7…♘c6 was played in our sample game 2, J.Polgar-Bu, Khanty-Mansiysk 2010. Please examine this game for a good example of White’s slow buildup in these lines. 8.0-0 ♗f8 9.h3 Our set-up controls d5 reliably, so this freeing advance is not available to Black without due preparation. 9.h3 is a ne move that controls g4 and prepares to expand on the kingside. 9…c6 10.g4 This may be the start of an attack with g4-g5 and f2-f4 to come, but can also be used to give the bishop on c1 a good post on g3. This is the option I prefer, if Black continues to wait. 10…d5 Otherwise ♗f4/♕d2/♖ad1 are easy moves to follow up with, putting more pressure on the centre. 11.exd5 ♘xd5 12.♘xd5 cxd5

13.♗f4 With a very comfortable edge. Black has a huge development disadvantage, and ♘b5 is a threat.

37

38

F32) 6…d5 This is the theoretically most challenging and most principled move. It’s the one option where we need to know some lines, while slow set-ups, starting with castling, will be slightly better for us without much e ort. 7.e5

7…♘g4 1) 7…♘e4 is inferior as the pawn emerging on e4 will be rather weak: 8.♘xe4 dxe4 9.♗g2 ♕d5 10.0-0 (just going to attack e4 with ♖e1 next) 10…

Chapter 2: Philidor’s Defence: 2.♘f3 d6

♘c6 11.♘xc6 ♕xc6 12.♖e1 ♗f5 13.c3!, and with ♕c2 or ♕e2 to come the e4-pawn will fall. Note that 13.♕e2 is not as convincing due to 13… 0-0-0! and now you can’t take e4 as Black has …♖d1+ at the end of the trade sequence; 2) 7…♘fd7? 8.♘f5! is close to lost already for Black. 8.♗g2 Here 8.e6 looks very tempting, but it is not 100% clear. Our choice is simpler to handle and still dangerous for Black. 8…0-0 1) 8…♘xe5 9.♕e2 (winning a tempo to speed up development) 9…♘bc6 10.♘xc6 ♘xc6 11.♘xd5 (a wide open centre position favours the side with the more active pieces. Here White is much better due to the active knight and the bishop on g2) 11…0-0 12.♗e3 ♖e8 13.0-0-0 and Black is under serious pressure; 2) 8…c6?! is dubious because the knight on g4 is lacking squares after White covers e5: 9.♗f4. I don’t see a good reply for Black now. 9…g5

Making it worse, but it is still important to check this. 10.e6! is strong, and absolutely necessary to refute Black’s play: 10…♘xf2 11.exf7+ ♔f8 12.♔xf2 gxf4 13.♖f1 with a great attack for White. 9.0-0 ♘xe5 After 9…c6, 10.♖e1 is extremely awkward to meet, 11.h3 being the threat. Black does not have a good reply that I (or the engine) can see. 10.♖e1 ♗d6 11.♘xd5

His more active pieces give White a comfortable edge. Conclusion The Philidor is not a bad opening if Black plays 3.d4 exd4!. Against this we have good chances for a small advantage, but nothing special. There are, however, many lines that are bad or downright unplayable for Black that still happen quite frequently below master level (and sometimes even there!). It’s useful to know the refutations and best replies to get an immediate headstart out of the opening.

39

40

Chapter 3: Petro Defence: 2.♘f3 ♘f6

CHAPTER 3

Petroff Defence: 2.♘f3 ♘f6 1.e4 e5 2.♘f3 ♘f6

This is the only good option that keeps the game within the realms of the Petro . All other sensible moves transpose somewhere else or blunder e5. With 3…♗b4 Black basically enters a Ruy Lopez Berlin Variation a tempo down. Alternatively, there is 3…d6. This is rare, but of course possible. Play now transposes into a Philidor: 4.d4 ♘bd7 (4…exd4 5.♘xd4 leads to Chapter 2) 5.♗c4 and this is the Closed Philidor that we usually get via 1.e4 d6 2.d4 ♘f6 3.♘c3 e5 4.♘f3 ♘bd7 5.♗c4, examined in Chapter 27.

The Petro has a deservedly solid reputation. Black is content with an at times slightly passive, but tough-tocrack position. 3.♘c3 This is a very convenient choice as we have the Scotch Four Knights in our repertoire. Now Black’s most popular move is actually 3…♘c6, transposing to the Four Knights immediately. An additional option that ts well with our repertoire is 3.♘xe5 d6 4.♘f3 ♘xe4 5.d3 ♘f6 6.d4 d5 7.♗d3, and surprisingly we have transposed to the Exchange French! This was played by World Champion Magnus Carlsen multiple times, which inspired me to look at the Exchange French in the rst place. 3…♗b4

41

4.♘xe5 This is the most consistent move. We grab e5 and Black needs to give up his pair of bishops to get material equality. The resulting positions are extremely solid for Black, but also purely defensive in nature. White will either make his bishop pair count and gain an advantage, or draw. Black has very few winning chances. 4…0-0 This is the most popular move as Black needs to castle anyway, so he keeps things exible. 42

Chapter 3: Petro Defence: 2.♘f3 ♘f6

1) 4…♕e7 5.♘f3 (here 5.♘d3 is also ne, of course, but 5.♘f3 is even better) 5…♗xc3 6.dxc3 ♕xe4+ 7.♗e2 and White will gain even more time on the black queen; 2) 4…d6 5.♘d3 ♗xc3 6.dxc3 ♘xe4 7.♗e2 leads to a position very similar to the main line. 5.♘d3 ♗xc3 6.dxc3 ♘xe4 7.♗e2

10.♗e3 White will play ♗f2, ♖e1 and ♗f1 and then try to slowly make progress. Preparing c3-c4 is a good plan. Black has little active ideas on his own. An excellent display of White’s strategy is sample game 3, Andriasyan-Gregoryan, Lake Sevan 2013.

Black now has a crucial choice. He may play with …d7-d5 or with …d7-d6. White always retains a slight edge due to the bishops. A) 7…♖e8 This delays the decision about the dpawn for one move. 8.0-0 And Black will now go …d7-d5 or …d7d6, transposing to lines B or C.

C) 7…d6 This seems the most logical to me: Black has a light-squared bishop, so he puts his pawns on dark squares. 8.0-0 C1) 8…♘c6 This is the old main line. Black just develops the knight to the most natural square. 9.♘f4

B) 7…d5 8.0-0 c6 Black decides on a light-square set-up. I now like the following engine suggestion, which is rarely played in practice: 9.f3 ♘f6

Heading towards d5 and getting out of the way. d3 is not an ideal square in the long run. Here White enjoys the longterm assets of the bishop pair and more space. The usual set-up involves ♘d5, ♗e3 and c3-c4 in the next few moves. f2-f3, to kick the black knight away from the centre, also happens

43

44

Chapter 3: Petro Defence: 2.♘f3 ♘f6

frequently. Let’s have a look at a sample line that features an additional idea. 9…♖e8 10.c4 ♗f5 11.f3 ♘c5 12.♖f2!? a5

13.♗f1 Followed by ♗c1-d2-c3. Note that the regrouping of ♖f2/♗f1 also prepares g2-g4 at a later stage. White generally wants to expand slowly, supported by the bishop pair. C2) 8…♘d7 This set-up is currently favoured by the experts on the Black side of this line. One idea is to trade a knight on c5. 9.♘f4 The best square – where else to go? Moving the knight now is also a precaution against …♘c5 by Black, when you need to move the knight anyway if you want to avoid trades. In general we like to keep pieces on due to the extra space. So it makes sense to preserve the knight here.

45

9…♘df6 The most popular move, and consistent with 8…♘d7. It is quite di cult to analyse this position move by move, but in fact it is not that essential to learn moves by heart here. White has the pair of bishops and will be better if he manages to gain more space and push the black pieces back, in particular the knights. White’s ideal scenario is to push the pawns on both sides in an endgame, and su ocate Black slowly. I will show the most important line to know from here, without any alternatives for Black. Other moves don’t change much what White will do: play f2-f3, c3-c4, develop the c1-bishop on the long diagonal if possible. 10.c4 ♖e8 11.f3 ♘c5 12.b3 A logical sequence. White uses the pawn to control central squares. The only task left is to activate the bishop on c1. Thus, b2-b3 makes a lot of sense. 12…♕e7 Black prevents 13.♗b2 due to 13… ♕e3+, winning the knight on f4. Also note that …g7-g5 could be a problem,

46

Chapter 3: Petro Defence: 2.♘f3 ♘f6

as the knight needs to protect the bishop on e2. 13.♖e1 White just needs to consolidate to be better as he has the bishop pair forever. Note that the dark-squared bishop would be fantastic without a counterpart on the long diagonal, like on b2 or c3. The ♖e1 covers the bishop, taking care of the …g7-g5 idea. 13…♕e5 The point of 18.♔g1!. White is slightly better here. Black has no particular compensation for White’s bishop pair.

14.♗d2! This is quite a strong idea. White intends to play c2-c3 and ♗f1 to harass the black queen on e5. 14…♕d4+ Otherwise, the mentioned plan of c2c3 and ♗f1 will be played, with a comfortable game for White. 15.♔h1 ♗d7 16.c3 ♕f2 17.♖f1 ♕h4 18.♔g1! ♘h5?! 19.♘d5

47

Conclusion Black sometimes goes 3…♗b4, avoiding the transposition into the Four Knights. This line is not to be feared as it leads to a rather static game that favours White. We have more space and the bishop pair, and a comfortable small advantage. It is a build-up game, expanding slowly to su ocate Black in the long run.

48

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

CHAPTER 4

Three Knights Game: 2.♘f3 ♘c6 3.♘c3

my favourite 4….♗b4+. Our move order to the Scotch Four Knights is easier to learn as it cuts down options for both sides. A) 3…♗b4

1.e4 e5 2.♘f3 ♘c6 3.♘c3

We are aiming for the Scotch Four Knights Opening, which is discussed in the following chapters 5-8. In this chapter we deal with Black avoiding the natural development of the fourth knight. One important point worth mentioning: the Scotch Four Knights may also arise after the immediate 3.d4, after which Black’s most popular continuation is indeed 3…exd4 4.♘xd4 ♘f6 5.♘c3, which transposes to our repertoire. Once you have mastered the Scotch Four Knights via our 3.♘c3 move order, it makes a lot of sense to add the Scotch move order with 3.d4 to your repertoire. This way you get similar structures but some additional options like 3…exd4 4.♘xd4 ♘f6 5.♘xc6 – of course Black also has some extra choices, most notably 4…♗c5 or 49

A premature bishop move. 4.♘d5 White now either gains time or the bishop pair. A1) 4…♘f6 5.♘xb4 ♘xb4 6.d4 exd4 7.e5 ♘e4 After 7…♕e7 8.♗b5! White gets a strong attack after he castles next move. 8.♘xd4 0-0 9.c3 And White is better with the bishops in a wide-open position. A2) 4…♗a5 5.c3 Preparing d2-d4. 5…♘f6 6.d4!

50

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

A21) 6…♘xe4? 7.dxe5! Now the threats 8.♗d3 and 8.♕c2 are very tough to meet. Note that a knight on c5 runs into a b2-b4 fork. 7…♗b6 This will lose immediately, but it needs to be checked as f2 is attacked. 8.♕e2 ♗xf2+ After 8…♘xf2 9.♗g5 White wins the queen – not bad after 9 moves. 9.♔d1 And Black loses a piece.

The favourite move of the rst o cial World Champion, William Steinitz. White gets an edge in a complicated position. 4.d4 exd4 5.♘xd4 Some time before this book was even planned I tried the tricky 5.♘d5 in a classical time control game against GM Dambacher. Unfortunately, after 5… ♗g7 6.♗g5 ♘ce7! Black is ne, so we should opt for the simple solution. 5…♗g7 6.♗e3

A22) 6…exd4 7.b4 ♗b6 8.♗g5!

And Black is in grave trouble already.

This is best here. White will only take on c6 after Black has committed more. B1) 6…d6

B) 3…g6

51

52

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

This is quite passive. White should now go for a quick 0-0-0 and attack. 7.♕d2 ♘f6 With 7…♘xd4 8.♗xd4 ♗xd4 9.♕xd4 ♕f6 Black tries to trade o as many pieces as possible. We should not agree to too many simpli cations as we have more space and more ideas for our pieces to improve: 10.♕e3, followed by 0-0-0 and further play in the centre and on the kingside (♗c4, f2-f4 etc.). 8.f3 Preventing the annoying 8…♘g4. 8…0-0 9.0-0-0

7.♘xc6 bxc6 8.e5 And Black must now go back. 8…♘g8 9.f4 ♘e7 10.♕f3 White is better after 0-0-0 and ♗c4, with more space and better pieces. Black’s king is not safe on the kingside with h2-h4 on the cards.

And White is better with g2-g4, and h2-h4 is coming.

C1) 4…exd4 5.♘xd4 ♘f6 Here 5…g6 6.♗e3 ♗g7 7.♕d2 transposes to 3…g6 (Line B above). 6.♗b5 And it transposes to Chapter 5, which reaches the position via 3…♘f6 4.d4 d6 5.♗b5 exd4 6.♘xd4.

B2) 6…♘ge7 Again a passive choice, so… 7.♕d2 0-0 8.0-0-0 And White is better, with h2-h4 being the simple plan. B3) 6…♘f6 After this move, queenside castling is less promising as Black would have more counterplay with …♖e8 and … d6-d5. We have a stronger alternative instead: 53

C) 3…d6 Of course, we go d2-d4 against this passive move. Most of the time we will transpose to lines examined elsewhere in the book. 4.d4

C2) 4…♘f6 5.♗b5 Again, this is examined in Chapter 5. C3) 4…♗g4 This does not transpose elsewhere… but it isn’t great either. 5.♗b5 exd4 54

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

After 5…♗xf3 6.♕xf3 exd4 7.♘e2 we will regain the pawn with a nice position, for example 7…a6 8.♗xc6+ bxc6 9.♘xd4. Black’s structure is weak and he lacks development. 6.♕xd4 a6 7.♗xc6+ bxc6 8.♗e3

D1) 4…♗xf2+?! This makes it even worse for Black. 5.♔xf2 ♘xe5 6.d4 White is a lot better here. He has the bishop pair and control of the centre. The king position will often be resolved with ♗e2/♖f1/♔g1, castling by hand. 6…♕f6+ 7.♔g1 ♘g4 After 7…♘e7, just avoid taking the knight due to …♕b6+! with mate to follow! 8.♗e3! is a good reply.

White is much better with a lead in development and more space. Black’s weakened pawns are also a factor. D) 3…♗c5 Black threatens mate on f2 and the knight cannot be taken due to 8… ♕xd4+ with mate coming! Quite a scary situation, but we have a refutation. 8.♕d2! Everything is covered now and Black has a miserable position.

This is surprisingly dubious due to the standard pseudo-sac on e5: 4.♘xe5! An important motif you should remember.

D2) 4…♘xe5 5.d4 Regaining the piece immediately. 5…♗d6 6.dxe5 ♗xe5 7.♗d3 d6 After castling with 7…♘f6 8.0-0 0-0, 9.♗g5! is strong now, as …h7-h6 and …g7-g5 is unattractive for Black. White is comfortably better here. 8.0-0 ♘f6

55

56

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

prepared against ambitious black players.

9.♘e2! A good move, threatening 10.f4 to win the bishop on e5. Note that here 9.♗g5?! is not great due to 9…h6 10.♗h4 g5 when Black is ne. 9…♘d7 10.c3 This is a useful move that frees c2 for our bishop if Black play …♘c5 later. 10…0-0 11.f4 ♗f6 12.♘g3 And White is better due to his centre control and some attacking chances on the kingside. Conclusion Black’s tries to avoid the Four Knights are not fully sound. In all cases White is better, but the game is complicated. Note that quite frequently these lines are played by players that want to keep more tension, ghting for a win. They fear that the Black side of the Four Knights gives very few winning chances, so they deviate. In fact I have looked at the lines presented here quite a bit from the Black side, because I had the same intention. At the end I never played any of them, as I felt it is too risky against good preparation. So take a good look at this chapter to be better 57

58

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

CHAPTER 5

Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 1.e4 e5 2.♘f3 ♘c6 3.♘c3 ♘f6 This is by far the most popular move, completing the development of the four knights. The alternatives are less reliable for Black and give White an edge without any particular problems. This is examined in Chapter 4. 4.d4

The Scotch Four Knights Opening. This line enjoys a very good reputation as it features quick development for White and a very sound pawn structure. It is easier to learn and to understand than the main alternatives, the Italian (Giuoco Piano) and Spanish (Ruy Lopez) Opening. Now, taking on d4 is the most natural and popular move, but Black has alternatives we need to check in this chapter.

A rare move at professional levels, but not that infrequent at club level. 5.♗b5 I like this best, even though 5.dxe5 is also promising. Now 6.♗xc6 and 7.dxe5 is threatened, winning the important e5-pawn. We have transposed to the ancient Ruy Lopez Steinitz Defence, so our repertoire has some Spanish avour as well! A1) 5…♗d7 6.♗xc6 Why give up the bishop?! It is a forced line, so hang on. 6…♗xc6 If 6…bxc6? 7.dxe5 wins the e-pawn. 7.♕d3 Covering e4, threatening the e5-pawn, and preparing queenside castling. 7…exd4 8.♘xd4 ♗d7 After 8…♗e7, 9.♘f5 is annoying. 9.♗g5 ♗e7 10.0-0-0

A) 4…d6

59

60

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

Now Black has three moves. Two of them are quite similar in nature and lead to a white edge. The third option is even worse for Black, but extremely complicated to play. I am now going to cheat a bit with the structure of lines: the mentioned rst two options are under B1 and B2, the third one should technically be B3, but I led it under C to better cope with the depth of lines. And in most cases we will play the familiar f2-f3/g2-g4 attacking scheme.

B1) 5…♕e7

A2) 5…exd4 6.♘xd4 This position may also arise from 4… exd4 5.♘xd4 d6 6.♗b5. 6…♗d7 7.♗xc6 bxc6 We have already covered 7…♗xc6 8.♕d3 via Line A1 above. 8.♕f3 And ♘f5 will give White good play. B) 4…♗b4

This move does not enjoy a fully sound reputation, but it’s tricky if you don’t know it. 5.♘xe5!

6.♕d3! The same move as after 5…0-0. 6…♘xe5 The spectacular 6…♘xd4 is playable for Black, but it does not equalize. 7.♕xd4 ♗c5 8.♕d3 ♕xe5 9.♗e3 ♗xe3 10.♕xe3 0-0 11.0-0-0 and White is slightly better due to his advantage in space and chances for a kingside attack. 7.dxe5 ♕xe5 8.♗d2 0-0 8…♗xc3? 9.♗xc3 ♕xe4+ 10.♕xe4+ ♘xe4 11.♗xg7 is a terrible ending for Black. 9.0-0-0

61

62

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

B11) 9…♗xc3 Starting forcing play, but it is not entirely convincing. 10.♗xc3 B111) 10…♕xe4?

11.♕g3! White is close to winning already. 11…♕g6 1) 11…♕g4 12.♗xf6 ♕xg3 13.hxg3 gxf6 14.♗d3 d5 15.♖xh7 and White is winning with ♖dh1, f2-f4 etc. to come; 2) 11…♕c6 12.♖d4! (a brutally strong computer move! White swings the rook to h4 with devastating e ect) 12…d5 13.♗d3 ♖d8 14.♖h4 g6 15.♕g5 and White wins. 12.♗d3 ♕h6+ 13.♗d2 ♕h5 14.♕xc7 With a fantastic position for White, as seen in Spielmann-Bogoljubow back in 1919.

14.gxf3 gxf6 15.♖g1+ ♔h8 16.♖xd5 ♗e6 17.♖h5 ♖g8 18.♖g3 and White is better in this ending. Black is now basically forced to take on g3, as ♗d3 is very annoying to meet; 3) 11…♘xe4.

This loses outright to a powerful response: 12.♕d4!, threatening to give mate on g7 and to play 13.f3 to win the knight at the same time. White wins a piece at least! 12.f3 d5 13.g3 ♕h6 14.♗g2 And White is slightly better due to the bishops. B12) 9…d6 10.f3 And White will attack with the standard g2-g4, h2-h4, etc. B2) 5…0-0

B112) 10…♕f4+ 11.♖d2 ♖e8 1) 11…♕xe4 12.♕g3 gives White a strong position again. Note that you can also regain c7 easily; 2) 11…d5 12.♕f3! (a new move suggested by Stock sh; it is simple and strong) 12…♕xe4 13.♗xf6 ♕xf3 63

64

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

C) 5…♘xe4?

6.♕d3! Important to remember. White is covering the centre and prepares 0-0-0. 6…♖e8 7.♗d2 Just returning the pawn for good play. 7…♘xe5 7…♘xd4?! looks nice, but just leads to a worse version of 7…♘xe5: 8.♕xd4 c5 9.♕d3 ♖xe5 10.0-0-0 and White is much better. Note that Black cannot get rid of the backward pawn directly: 10…d5 11.♘xd5 ♘xd5 12.♗f4! ♖e8 13.exd5 and it’s just a fat extra pawn for White. 8.dxe5 ♖xe5 9.0-0-0

And White is better, f2-f3/g2-g4 etc. being the main plan. 65

This is the bad but tricky line. White is better afterwards, but with exactly one reply only. Of the whole repertoire, this is about the only variation where you absolutely need to play super-precisely to get a huge advantage. If you don’t manage to do that you will very likely end up even being worse as the price of every move is extremely high. 6.♕g4! ♘xc3 7.♕xg7 ♖f8 7…♘d5+ 8.c3 doesn’t work for Black either: 8…♕f6 9.♕xf6 ♘xf6 10.♘xc6 ♗xc3+ 11.bxc3 dxc6 12.♗d3 and White is much better with his superior structure and the bishop pair. 8.a3 A crazy position, but things will settle down quite quickly. 8…♘xd4 This is the critical move that sets the board on re! 8…♗a5 is relatively best, but still unsatisfactory for Black: 9.♘xc6 dxc6 10.♕e5+ ♕e7 11.♕xe7+ ♔xe7 12.♗d2. The point! White gets the piece back with an advantage. 9.axb4 66

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

This forces the next moves. 9…♘xc2+ 10.♔d2 ♘xa1 11.♔xc3

Attacking f7 and preparing ♖e1 at the same time. Wow, quite a sight! But it is just winning for White! The following complications are very attractive to analyse. You should set some time aside and go through the lines shown here and add some of your own analysis, supported by a computer. This will prove to be excellent attacking training and it will help you to understand this line better. It might never appear in one of your games, but it still is instructive to study and will improve your overall abilities. Back to the actual position. White is threatening both 12.♗h6 and 12.♗c4, crashing through. The only moves to pose us some problems are connected to Black playing ….a7-a5 and/or …♕e7, immediately distracting us due to the king’s position on c3. C1) 11…a5 This move does not save Black, but we need to be very precise to prove it. 12.♗c4

67

C11) 12…axb4+ 13.♔d2 Instead 13.♔d3 wins as well, but our move is simpler as the king covers e1, making ♖e1 easier to achieve.

C111) 13…♘b3+ 14.♔d1! You can also take the knight, but this is a direct win. 14…♕e7 1) 14…♘xc1 already leads to a forced mate: 15.♗xf7+ ♔e7 16.♕g5+ ♔d6 17.♘c4+ ♔c6 18.♕d5#; 2) 14…d5 15.♗b5+ c6 16.♖e1 and again Black gets crushed on the e- le. 15.♖e1 ♖a1 16.♗xb3 And White has a winning attack. 68

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

C112) 13…♕e7 14.♖e1 ♕d6+ 15.♔e2 ♕e7 16.♔d1

We need to check 14…c6 15.♘xf7+ (there are some other wins, but this is very straightforward) 15…♕xe1 16.♘d6+ ♔d8 17.♕xf8+ ♔c7 18.♘e8+ ♔b6 19.♗e3+ and it’s over. 15.♗xf7+ ♔e7 16.♕g5+ ♔d6 17.♘c4+ ♔c6 18.♕d5#

This funny king dance wins easily now. Black collapses on the e- le. C12) 12…♕e7 A familiar mating pattern in this line. C122) 13…axb4+ 14.♔d2 This position we have already checked via 12…axb4+ (line C11). White wins on the e- le.

The most important defensive try. 13.♖e1! Very important to remember! Other moves do not win, but this does the job!

C123) 13…d5 Black’s nal bullet, but it fails. 14.♗b5+ c6 14…♔d8 allows an attractive mate: 15.♕xf8+ ♕xf8 16.♘xf7+ ♕xf7 17.♗g5+ ♕e7 18.♗xe7#.

C121) 13…♕xb4+ 14.♔d3 And now Black is out of tricks! There are no sensible checks anymore and White’s attack is winning the game. 14…♕xe1 This leads to a forced mate. 69

70

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

Black’s best shot. 12…a5 13.♖e1! transposes to the line 11…a5 12.♗c4 ♕e7 13.♖e1! (line C12). 13.♗b5+ c6

15.♘xc6! It pays o to know this beforehand… not so easy to nd all these shots over the board. 15…♕xe1+ 16.♗d2 An amazing position! White has given up lots of material, but the black queen is attacked and the king is caught in a cross re. 16…♕e4 Here 16…bxc6 returns all the material to get a lost ending: 17.♗xc6+ ♗d7 18.♗xd7+ ♔xd7 19.♗xe1 and White wins. The extra pawn, better minor piece and, most importantly, bad king safety on the black side decides the game. 17.♘b8+! A very nice shot, checking on the starting square of the sorry piece on a1! 17…♔d8 18.♕xf8+ ♔c7 19.♕c5+ ♔xb8 20.♕d6+ ♔a7 21.♗e3+ And the black queen is nally forced to take the piece, decisively losing material.

14.♗e3! Very nice and important to remember – other moves don’t win! 15.♗c5 is a decisive threat now. C21) 14…a5 15.♘xc6 This wins, but it is not so obvious as White’s king is also in the open. Let’s check: C211) 15…axb4+ 16.♘xb4+ ♗d7 17.♗xd7+ ♔xd7 It’s easier against 17…♕xd7 18.♖e1! 00-0 19.♖xa1 and things have settled down to a winning advantage without any particular problem. 18.♘xd5

C2) 11…♕e7 12.♗c4 Putting pressure on f7 and preparing ♖e1. 12…d5 71

72

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

king is a huge asset. The g- and f-pawns might create a passed pawn quickly. C212) 15…bxc6 16.♗xc6+ ♗d7 17.♗c5 axb4+ 18.♗xb4 d4+ 19.♕xd4 ♕e6 20.♗xd7+ ♕xd7 21.♕e4+ And here Black will probably resign.

Our attack is very strong obviously, but still we have a king on c3. 18…♖ac8+ 19.♔d3 ♕d6 20.♖d1 Yet another great move! 20…♖c2 20…♕xd5+ 21.♔e2 ♖c2+ 22.♔e1 and nally our king is safe and Black falls apart. 21.♔e4! ♕g6+ Against 21…♖e8+ 22.♔f3 there are no more checks! 23.♘f6+ or 23.♘b6+ will end the game now. 22.♕xg6 fxg6 23.♖xa1 ♖xb2 24.♖d1

And we have nally reached a winning endgame. Note that White’s pieces coordinate perfectly and the active

73

C22) 14…f6 15.♕xe7+ ♔xe7 16.♗c5+ ♔e6 17.♘xc6! bxc6 18.♗xc6 d4+ 19.♗xd4 ♖b8 20.♗xa7 ♗b7 21.♗xb8 ♗xc6 22.♗a7

And White will regain the knight on a1 with a winning endgame. Conclusion I hope you had your seatbelt on while being on that wild ride! This line leads to crazy positions, but we need to be ready for complications if the result is that excellent. I was trying hard to avoid such long, sharp lines for economical reasons – you don’t get this often at all and learning so much seems excessive. However I think it still makes sense, because this is not just ‘some opening line’, but actually excellent attacking play! It’s a good exercise in calculation if you use the 74

Chapter 6: 4…exd4 5.♘xd4

material presented here to start thinking yourself about positions and add some own analysis in cases where you are unsure. So just treat it as general chess education rather than just learning some moves by heart. If you actually get it on the board in a real game… all the better!

CHAPTER 6

4…exd4 5.♘xd4 1.e4 e5 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 exd4 Most natural and best. The alternatives are passive or slightly unsound, as examined in the previous chapter. 5.♘xd4

A) 5…♘xd4 This move is played quite often at the club and junior level. White is happy to get the queen in the centre. Despite being in the middle of the board, it is actually in a quite safe position. It also hinders Black’s development. 6.♕xd4 d6 That’s the issue – Black can no longer conveniently move his bishop out. 7.f3 One of many good moves. We head for queenside castling and a pawn storm by White. This is often the best thing to do against passive play. 7…♗e7 8.♗e3 0-0 9.0-0-0

75

76

Chapter 6: 4…exd4 5.♘xd4

8.♗b5 ♗d7 9.♗xc6 bxc6 10.0-0 dxe4 11.♖e1 0-0-0 The only alternative, 11…f5 12.fxe4 fxe4 13.♕e2 ♕e5 14.♘f3 ♗c5+ 15.♔h1 ♕f5 16.♕xe4+ ♕xe4 17.♖xe4+ ♔d8 18.♗f4, leads to a white advantage as well, again due to Black’s weak pawns. 12.♖xe4 ♕b4 13.♘b3

White is already much better. The position is also very simple to play with g2-g4/h2-h4 etc. on the agenda. B) 5…d6 6.♗b5 This transposes to a line we examine via 4…d6 5.♗b5 exd4 6.♘xd4 (Chapter 5). C) 5…♘xe4 And White is better due to Black’s unsafe king and bad pawn structure. He’s got the bishop pair, but it’s tough to play with the defects mentioned. D) 5…♗c5

What?! Yes, this is actually a possible move. It is not great, but it has enormous shock value. Let’s have a look. 6.♘xe4 ♕e7 This is the point. 7.f3 d5 The only move to get the piece back. 77

This is Black’s main alternative to 5… ♗b4, often chosen by ambitious black 78

Chapter 6: 4…exd4 5.♘xd4

players to avoid the main lines. 6.♗e3 Taking on c6 instead is less testing. We will see that after 5…♗b4, the main line, Black sometimes returns to c5 with his bishop. Compared to these lines we would be a tempo down. That’s not the end of the world, but keeping the knight on d4 is tougher for Black to meet. D1) 6…♘xd4?! This trade is almost always bene cial for White. Black loses time and ends up activating White’s pieces. 7.♗xd4 ♗xd4 7…♕e7 8.♗e2 is also better for White with his space advantage in the centre. 8.♕xd4 d6 9.0-0-0 White is already threatening e4-e5; Black is in trouble. D2) 6…d6? Allowing a terrible pawn structure… 7.♘xc6 bxc6 8.♗xc5 dxc5 9.♕xd8+ ♔xd8 10.0-0-0+

It’s quite strange to move the bishop again, but it’s not so bad. White is a full tempo (♗e3) ahead compared to 5… ♗b4, but it does not seem to matter as much as you might rst think. White only gets a small advantage. 7.♘xc6 bxc6 Black can try to spoil the pawn structure by going 7…♗xc3+ 8.bxc3 bxc6 9.♗d3 d6 10.0-0 0-0 11.♗g5, but White is slightly better with his bishops and the annoying pin on the black knight.

8.♕d4! ♕e7 9.0-0-0 0-0 Being greedy with 9…♗xc3? 10.♕xc3 ♕xe4? 11.♗d3 is suicidal for Black. 10.♗g5 ♖e8 11.♗xf6 gxf6 12.♗c4 And White is better due to his pawn structure. D4) 6…♗b6

White has a very enjoyable ending. D3) 6…♗b4!? 79

80

Chapter 6: 4…exd4 5.♘xd4

Covering the bishop by retreating it. This is by far the most popular move. 7.♕d2 This is best. White aims for quick queenside castling. D41) 7…♘g4 This forces the action. We need to take on c6 now: 8.♘xc6 bxc6 It is very comfortable for White to face 8…dxc6 9.♕xd8+ ♔xd8 10.♗xb6 axb6 11.f3. Here White is slightly better, due to the mobile 4 vs 3 pawn majority on the kingside while the black queenside majority is hampered by the doubled pawns. 9.♗d4 0-0 10.f3 ♘f6 11.0-0-0 And White is comfortable in this position. We go g2-g4/h2-h4 etc.

A very logical sequence after 6…♗b6. Now Black either goes for forcing play with 9…d5 or will end up permanently cramped. 9…d5 The only critical reply, but White is better. The passive 9…d6 10.g4 is no fun for Black. White’s play is too easy – just pawnstorm the kingside. 10.exd5 ♘xd5 11.♗g5

D42) 7…0-0 8.0-0-0 ♖e8 9.f3 This is how Kramnik played in the stem game of this line against Aronian. 11…♘ce7 1) 11…♘xc3?! was Aronian’s creative, but not fully correct reply. Black does not get quite enough for the queen: 12.♗xd8 ♘xd1 13.♗h4 (Kramnik took 81

82

Chapter 6: 4…exd4 5.♘xd4

on c7, but 13.♗h4 is even stronger) 13…♘xd4 14.♕xd1 ♘f5 15.♗g5 h6 16.♗d2 ♘e3 17.♗xe3 ♗xe3+ 18.♔b1 ♗f5 19.♗d3.

Black can only try to set up a fortress that probably won’t hold. It’s just a two-results game with White either winning or drawing; 2) 11…♘f6 12.♘xc6 ♕xd2+ 13.♗xd2 bxc6 14.♗d3 and White is eternally better with Black su ering due to his weak queenside. 12.♗b5 Very strong and direct. Black is already in trouble. 12…c6 13.♘xc6 bxc6 14.♗xc6 ♘xc3 15.♕xc3 ♘xc6 16.♗xd8 ♘xd8 17.♖he1

83

And White is better here, as Black’s minor pieces don’t coordinate all that well. If White manages to trade rooks, the queenside pawns will easily win the day. Conclusion The line starting with 5…♗c5 is the main part of this chapter. It leads to interesting play, but it looks somewhat dubious for Black. Still, the complications are not easy to master. You should look closely at the mainline position after 11.♗g5. It is the most critical, and the toughest to play for both sides. Some lines lead to unfamiliar material distributions. It is very interesting to use those positions in training games or against engines to explore further. This is also helpful for your general chess education, not just your opening knowledge.

84

Chapter 7: 4…exd4 5.♘xd4 ♗b4

CHAPTER 7

4…exd4 5.♘xd4 ♗b4 1.e4 e5 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4 ♗b4

The established main line for the last 500 years… well, maybe not that long, but the oldest game I have with this position is from 1873. 6.♘xc6 The only move. There is no other way to make ♗d3 possible, desirable because the ♗d3 will comfortably cover the threatened e4-pawn. 6…bxc6 Very natural, and the main move by a mile. Two playable alternatives need a quick check though: 1) 6…dxc6 is rare and not great. White gets a good ending without any problems: 7.♕xd8+ ♔xd8 8.♗d2.

85

White is very comfortable here as his kingside majority is healthy, while Black has a hampering doubled pawn on the queenside. It’s like a Ruy Lopez Exchange variation without any drawbacks for White who still has the pair of bishops; 2) 6…♗xc3+ is unpopular, but not that bad. Black gives up the bishop pair in order to spoil White’s structure: 7.bxc3 bxc6 8.♗d3 0-0 (8…h6? 9.e5 is terrible for Black) 9.0-0 (it is worth noticing that the immediate pin with 9.♗g5 is not a great idea here, since Black has pin motifs on the e- le when White has not castled, for example: 9… d5! 10.e5 ♕e8, or 10.♕f3 dxe4 11.♗xe4 ♖e8) 9…d6.

86

Chapter 7: 4…exd4 5.♘xd4 ♗b4

10.♗g5. This pin is important to keep some pressure. Black could have prevented it with an earlier …h7-h6, but that would have been punished with e4-e5, kicking the knight away. Now f2-f4 is on the agenda, trying to attack in the centre and kingside. White is slightly better. 7.♗d3 We develop the bishop and cover e4. Now most of the time Black will go for …d7-d5, either immediately or after castling rst. This is the most respectable way for Black to play, according to the motto that Black is OK in the open games if he gets in …d7-d5 unpunished. The absolute main line of the Scotch Four Knights features this approach and is covered in Chapter 8. This chapter features all the possible deviations before this main-line position is reached. Black has two main ideas to avoid the main line: 1. not playing …d7-d5 quickly, but rather delaying it or going for a …d7d6 set-up. 2. playing …d7-d5, but trying to use move order tricks or concrete tactics.

This check is not seen all that frequently in high-level play. In the coming endgames, White has at least a slight pull. 9.♕e2 A11) 9…cxd5 This leads to the black king being exposed in the centre. He has to be precise to avoid trouble. 10.♕xe7+ ♔xe7 Or 10…♗xe7 11.♘b5 ♔d7 12.0-0.

A) 7…d5 8.exd5 Black has two ways now to not transpose to the main line. A1) 8…♕e7+ This is seriously bad for Black already. Moves like ♖d1 and c2-c4 are coming. The pawns on c7 and d5 are a liability, and the unsafe king does not help either. 11.0-0 87

88

Chapter 7: 4…exd4 5.♘xd4 ♗b4

Instead, 11.a3 is a good alternative. Castling is very logical as well, unpinning the knight and preparing to connect the rooks soon. 11…c6 12.♘a4 Looking at the weak square c5 and preparing c2-c4 to further weaken the pawns. 12…♖e8 13.h3 A good move, avoiding being harassed after 13.♗e3 ♘g4. 13…♔f8 14.♗e3 ♔g8 15.♖ac1 ♗d6 16.c4

And White enjoys the better endgame. He has the better structure (count the pawn islands) and well-coordinated pieces. We’ll see more of this endgame type in the main line.

More principled is 12…♗a5, but 13.b4 ♗xb4 14.axb4 ♘d5 15.0-0 gives excellent play for the pawn. Black’s extra pawn is not great to begin with and we have the bishops and play against the king on top. 13.♗xc3

This position I had on the board against IM Mikhail Zaitsev in February 2018. White is a lot better here with play against the pawn structure and the black king. I spoiled an almost winning position later to a draw. A13) 9…♕xe2+ 10.♔xe2 cxd5 10…♘xd5 leads to further trades, but White retains a small pull by 11.♘xd5 cxd5 12.♗f4 c6 13.c4 ♗e6 14.♖hc1. 11.♘b5 ♔d8 12.♖d1

A12) 9…♘xd5 This leads to a worse endgame for Black if we are precise. 10.♕xe7+ ♔xe7 11.♗d2 ♘xc3 Logical, as 12.♘xd5+ was the threat. 12.a3 Very important and strong! If we just recapture our pawn structure is no better than Black’s. 12…♗d6 89

90

Chapter 7: 4…exd4 5.♘xd4 ♗b4

And White is better. The black king is unsafe and the pawns are still weak. A2) 8…0-0 Sometimes Black tries to trick us with this move order. Black hopes that White will grab on c6, but that is far too risky. It is best to just play 9.0-0! to transpose to the main line (line C) examined via 7…0-0 here. A3) 8…cxd5 This is by far the most common, and transposes to the main line after 9.0-0 0-0 10.h3 (Chapter 8). Black has an entirely di erent approach in: B) 7…d6

91

This is an interesting option for Black. He intends to play …♗c5 and possibly …♘g4 to start quick counterplay. In many cases he will not castle kingside at all. 8.♗g5 This is our normal reply to …d7-d6, pinning the knight. The special case here is that Black has not castled yet, so a possible …h7-h6 and …g7-g5 is more likely to happen. 8…h6 9.♗h4 ♗e6 A good waiting move, hoping for a commitment by White. 1) 9…g5 is too hasty, because White has not castled yet either: 10.♗g3 h5 11.h4 and Black will not nd a safe place for the king anymore; 2) 9…0-0 10.0-0 – this we have checked via 7…0-0. White is slightly better due to the unpleasant pin. 10.a3!? This is a yet untested move that avoids 10 0-0 g5 11 ♗h4 h5, which is what Black wanted to get on the board. 10…♗xc3+ 1) 10…♗a5?! 11.b4 ♗b6 12.♗a6! g5 13.♗g3 ♗c8 14.♗xc8 ♕xc8 15.h4 –

92

Chapter 7: 4…exd4 5.♘xd4 ♗b4

White is better here as the black king will never be completely safe; 2) 10…♗c5?! leads to serious problems: 11.♗a6! and White threatens both 12.♗b7 and b2-b4 & a3a4, trying to trap the bishop. 11.bxc3 g5 12.♗g3 ♘d7 13.♕d2

I prefer White here as Black will struggle long-term with his king position. f2-f4 is coming soon. C) 7…0-0 8.0-0 And we have the nal split in this chapter. Black may delay …d7-d5 even longer, or omit it (C1 and C2), or play it right away (C3). C1) 8…♖e8

93

This was played a number of times by Aronian. White only gets a small edge against it. 9.♗g5 h6 10.♗h4 C11) 10…♖b8 After slow moves like this, White has f2-f4 available. 11.f4 ♗e7 12.b3 d5 12…♘d5 is a yet untried move suggested by the engines: 13.♗xe7 ♕xe7 14.♘xd5 cxd5 15.exd5 and White is better, even if Black manages to regain the pawn. 13.e5 ♘d7 14.♗f2 And I prefer White with his space advantage in the centre. C12) 10…g5 The most forcing option, but also risky for Black. 11.♗g3 ♗xc3 12.bxc3 ♘xe4 13.♕h5 Black has grabbed a pawn, but now needs to be very precise. 13…d6 This is the only good move according to the computer. White has a risk-free game, though, as we will see in the coming lines. 1) 13…♔g7? has been the most popular move, but it just loses against best play: 14.♗xe4 ♖xe4 15.f4 g4 (the only move to delay defeat) 16.♗f2! and the bishop coming to d4 will be fatal for Black; 2) 13…♕f6 allows a dangerous initiative for White as well: 14.♖ae1 d5 15.f3 ♘d6 16.♖xe8+ ♘xe8 17.f4 and White keeps the pressure up. 14.♕xh6 ♗f5 15.f4

94

Chapter 7: 4…exd4 5.♘xd4 ♗b4

Most interesting, but if you are happy with a simpler position without risk you may also play 15.♗xe4 ♖xe4 16.♖ae1. 15…♖e6 16.♕h5

And White has an easy game, with both ♕f3 coming and possibly ♘a4, depending on Black’s play.

Black needs to defend carefully here. He is OK, but it is not easy to do in practical play. C2) 8…d6 If Black aims for this set-up he should probably go for it without castling rst. Now the pin is annoying. 9.♗g5 h6 10.♗h4 And it is tough for Black to avoid …g7g5, which weakens the kingside considerably. 10…♕e7 Black does not want to play …g7-g5, but now White gets an attractive position by preventing it aggressively. 11.f4

95

C3) 8…d5 9.exd5 ♗g4 This is the nal possibility to avoid the line from Chapter 8 with the natural 9…cxd5 10.h3.

This is a very recently introduced new development in the Scotch Four Knights. As far as I can tell, the rst game featuring this move was Wei YiVidit, Liaocheng 2018. I recommend to play: 10.♘e2 This move has not yet been tested in a practical game. The position is not very

96

Chapter 7: 4…exd4 5.♘xd4 ♗b4

forcing, so I will just show some lines to illustrate what might happen. It is more obvious to play 10.f3, but the following concrete lines lead to simpli cations and don’t seem to promise any tangible advantage. One key line is 10…♗h5 11.dxc6 ♕d4+ 12.♔h1 ♗xc3 13.bxc3 ♕xc3 14.♖b1 ♗g6 15.♗b2 ♕xc6 16.♗xf6 ♕xf6 17.♗xg6 hxg6. This was played in the mentioned stem game. It’s a good line if you badly need a draw, but it poses no problems for Black at all. 10…cxd5 11.h3 ♗h5 12.c4 Moving the knight to e2 did not look very active, but it allows this desirable pawn lever. 12…c6 Or 12…dxc4 13.♗xc4 ♗d6 14.♖e1, when White is slightly better due to the pawn structure. 13.♕c2

White is avoiding any attempts to mass-simplify to a draw. We’ll just get a ‘play chess’ position and go from there. Conclusion This chapter contains ways for Black to avoid the main line of Chapter 8. It’s a mixed bag of lines – sometimes Black tries to complicate the play, sometimes rather to simplify the game. The most important lines to study are C1 (Aronian’s 9…♖e8) and the entirely new idea of 9…♗g4, examined above in line C3. This line is not yet wellknown, but looks like a very viable option for Black.

With the idea to play ♘f4 to chase the bishop. It is fascinating that it took such a long time for a move like 9… ♗g4 to be introduced into practice. There is no experience yet with my suggestion of 10.♘e2, but it looks like 97

98

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

CHAPTER 8

5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3 1.e4 e5 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4 ♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

h4 after a later …h7-h6. With 10.h3 White will almost always go for ♗f4 soon. Note that White has a completely sound pawn structure and no weaknesses at all. This is the reason why this line is considered to be so solid for White: you have no structural risks to consider and easy development of your pieces, making it an ideal choice for us. Now the most popular move is 10…c6, but Black may try to delay this move or go for a di erent central pawn structure.

10.h3 is the move that has brought new attention to the Scotch Four Knights. It was basically unknown until Kramnik played it against Aronian in 2012. After that game hundreds of games have been played, and the conclusion is that White has a low-risk game with chances to play for a small edge. Black may equalize, but it is not so easy. What does 10.h3 do, by the way? It prepares ♕f3 and keeps options open. It will be useful in almost all cases, making luft for the king and taking g4 under control. This allows a set-up with ♕f3 and ♖d1, for instance. The traditional main line was 10.♗g5, but it is not clear whether White will want to take on f6 or drop the bishop back to 99

A) 10…h6 Black prevents 11.♗g5 (a move that we rarely go for anyway) and does not commit to …c7-c6 yet. 11.♕f3 After this, 11…c6 is by far the most common move. The c6-based lines are discussed via the 10…c6 move order. A1) 11…♖b8 This usually signals Black’s intention to play …♗b7 and omit …c7-c6, in order to use the bishop on the long diagonal. 12.♗f4 A11) 12…♗d6 13.b3 ♖b4 A very common idea to force a commitment of the f4-bishop. 14.♗xd6 ♕xd6 15.a3 ♖b8 After 15…♖f4 16.♕g3 the rook is rather vulnerable on f4.

100

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

13…♖xe1+ 14.♖xe1 White has improved with the last pair of moves, but Black is still only slightly worse. 14…d4?

16.♕g3 This is a typical idea in this line. The endgame after the queen exchange is just a little bit more comfortable for White. 16…♕xg3 17.fxg3 ♗e6 Pushing 17…c5 18.♘a4 leaves the pawns very weak. 18.b4 ♘d7 19.♘a4 White is a little bit better here. His pieces are more active and a future c2c4 break opens up the c- le.

This does not work, but it’s a line we should know beforehand. 15.♕g3! ♘h5 16.♗xc7 ♘xg3 17.♗xd8 dxc3 18.♖e8+ ♗f8 19.♗e7 And White is winning.

A2) 11…♖e8 12.♗f4 ♖b8 Instead, 12…c6 is the main and best move, leading to positions we will discuss after 10…c6. 13.♖fe1 This simple move is quite annoying as White threatens to take on e8, winning a pawn.

B) 10…♖e8 Black does not commit to …c7-c6 just yet. If he plays this move later (the majority of cases) we will transpose to 10….c6 lines. 11.♕f3 ♖b8 12.♗f4 ♗d6 12…♗a5. 13.b3 ♖b4 Black has also tried 13…d4 here, assessed as equal by Martin Lokander in his book opening repertoire: the Open Games with Black. After 14.♗xd6 ♕xd6 we should play 15.♘b5 to try for something (15.♘e4 is too stale). Now after 15…♕c5 16.b4 ♕xb4 17.♘xc7 Black still has some problems to solve. 14.♗xd6 ♕xd6 15.a3 ♖b8 16.♕g3 ♕xg3 17.fxg3

101

102

A12) 12…♗b7 This allows a forced sequence. 13.♘b5 d4 14.♗xc7 ♗xf3 15.♗xd8 ♖fxd8 16.gxf3 Here the engine sees enough compensation for Black. This is very likely true, but still Black has to prove it in an actual game.

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

A well-known endgame structure that allows White slight pressure without a risk. Ideas are ♘a4/b3-b4 to x the c5square and/or c2-c4 to attack the d5pawn. C) 10…♖b8 Black prepares to play the bishop to b7. By protecting it Black enables a future …d5-d4 push against the white queen on f3. 11.♕f3 ♗b7 Instead, 11….♗e6 is mentioned in Martin Lokander’s book. The following moves I agree with: 12.a3 ♗e7 13.b4 c5. Now I would play 14.♗f4 ♖b7 15.bxc5 ♗xc5 16.♖e1, still claiming White’s position is easier to play. You have ideas like ♘e2, ♗f4-e5-d4, playing on the weakened dark squares. 12.♗g5 We play the bishop to f4 almost all of the time, but here is the exception. After 12.♗f4 d4 we lack a good reply.

C1) 12…h6 Black encourages the trade on f6. The arising structure, with the doubled pawns, is seen quite often in this opening and is usually OK for Black. However, here we have a good version for White because the bishop is misplaced on b7 as it does not control the f5-square. 13.♗xf6 ♕xf6 14.♕xf6 gxf6 15.♘b5 ♗c6 16.♘d4 ♗d7 17.♘f5

White is slightly better here, though admittedly Black has good drawing chances. You may also wait with ♘f5 and play 17.c3 or a similar move, which keeps a slight edge as well. C2) 12…d4

103

104

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

This leads to simpli ed positions that are only very slightly better for White. 13.♘e4 ♗xe4 13…♗e7?? fails to a well-known idea: 14.♘xf6+ ♗xf6 15.♕f5 and White wins. 14.♗xe4 ♗e7 14…h6 15.♗xf6 ♕xf6 16.♕xf6 gxf6 17.♖fd1 is no fun for Black. His structure is weak and he has no active plan, while White can target various pawns. 15.♗xf6 ♗xf6 16.b3 This is just a bit more comfortable for White. The bishop will come to c4 and will be more active than its counterpart. These positions are almost impossible for White to lose. C3) 12…♗e7 13.♖fe1 h6 14.♗h4 d4 This is the principled move, but it does not fully equalize. Let’s check a logical sequence of moves. 15.♘e4 ♘xe4 16.♗xe7 ♕xe7 17.♗xe4 ♕b4 18.♗xb7 ♕xb7 19.♕xb7 ♖xb7 20.b3 c5 21.♖e5 ♖c8 22.♖ae1 ♔f8 23.♖d5

This is annoying to play for Black. The pawns are weak and White has easy ideas, like walking the king to the queenside. Black needs to be precise not to drift into serious trouble. D) 10…c6 This is the most common. Black supports his pawn on d5 and prepares a future …♗d6 without having ♘b5 to worry about. 11.♕f3

Our set-up involves the moves ♕f3 and ♗f4, almost no matter what Black does. D1) 11…h6

105

106

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

Stopping ♗g5, but we prefer the bishop on f4 anyway. 12.♗f4 ♗d6 13.♖fe1 Very natural, but White also has moves like 13.♘e2 or 13.♖ad1. The current position is not a very forcing one, where just one line of play is possible and others fail tactically. 13…♖b8 It also looks natural to play 13…♗xf4 14.♕xf4. And White has the sunny side of the torture, as Yasser Seirawan once said about a similar type of position! D2) 11…♖e8 12.♗f4

This was played in the well-known game Wang Hao-Anand, Wijk aan Zee 2013. This game is our sample game 6. It is well worth studying as it shows how easily Black can drift into a worse position. 14.b3 ♖b4 A very typical way of playing, but in this precise position there is a small detail that works in White’s favour. Let’s see: 15.♗xd6 ♕xd6 16.♕e3! ♖b7 The queen was attacking a7. Note that 16…d4 17.♘e4 is very nice for White. 17.♘a4 ♗e6 18.♕c5 ♕xc5 19.♘xc5 ♖e7 20.♗f5 ♖fe8 21.♗xe6 fxe6 22.♖e5

107

12…♗d6 Nikolaos Ntirlis advocates 12…♕e7 in his book Playing 1.e4 e5. This is indeed a good alternative. I suggest to play 13.a3 ♗d6 and now 14.♖fe1!?, a new move. It looks a bit odd, as often the two rooks are better than the queen, but after 14…♕xe1+ 15.♖xe1 ♖xe1+ 16.♔h2 White is at least slightly better. Black is not fully coordinated yet, which becomes very apparent after trading on f4. ♕c7 will be very annoying in that case. Best is 16…♗f8, 108

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

but after 17.♘e2 White is for choice. Note that ♗f4-d2-c3 is a good improving idea. It is probably better for Black to altogether avoid taking on e1 and instead respond with 14…♗e6, which is not an easy move to make unless you know it. Now a normal move like 15.♘e2 leads to a typical middlegame similar to 12…♗d6, the move I have as the main line. 13.♖fe1 ♗e6 14.♘e2 Again we prepare the knight transfer to d4/f4 and free up the c-pawn to attack d5 later with c2-c4. 14…♕b8 15.b3 ♗d7 16.♗xd6 The normal-looking move 16.c4?? is a blunder due to 16…♖xe2!. 16…♕xd6 17.c4

Now Black has a decision to make about the d5-pawn. Taking on c4 leaves the c6-pawn isolated and activates the c4-bishop, so that looks bad. If he allows White to take on d5, he will again get an isolated pawn that White can target. This position was reached in sample game 6, Rublevsky-Najer, Eilat 2012. Please study this game to learn 109

more about the endgame structures that are common in this line. D3) 11…♗d6 12.♗f4 ♖b8 With 12…♕c7 Black forces the trade on d6: 13.♗xd6 ♕xd6 14.b3 (this is usually played only after being provoked by …♖b8, but it is useful anyway, to prepare the c2-c4 lever later) 14…♗d7 15.♕g3 (not the only option, but I like these endgames) 15… ♕xg3 16.fxg3 and now: 1) 16…♖fe8 17.♘a4 ♗e6 18.♖ac1 ♘d7 19.c4 and I like White here. It’s not much, but it’s easy to play; 2) 16…c5

17.♖xf6!. This is not a sacri ce really… White has enough material and easy play: 17…gxf6 18.♘xd5 ♔g7 19.♖f1 and White has an easy game, attacking Black’s pawns. 13.b3 ♖b4 Or 13…♖b7 14.♗xd6 ♕xd6 15.♕g3 ♕xg3 16.fxg3, which is yet another version of this endgame. There are some potential chances here for White, due to c2-c4. In general, engines tend to evaluate those endgames as equal, but I don’t think it is that easy for Black 110

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

to play. I play 1.e4 e5 as Black and I can assure you that it is not a particularly fun experience. 14.♗xd6 ♕xd6 15.♕g3 ♕xg3 16.fxg3

I prefer White here, even though it can’t be much objectively. As mentioned, it is a risk-free way to play. Win some, draw some, lose none. I think going for these endgames is good practical play. There is still enough imbalance in the position to outplay weaker opponents (if they get that far in the rst place) and you are very solid against stronger opposition. An excellent example for this type of endgame is sample game 5, Klein-Van Wessel, Netherlands tt 2015/16. 16…♗e6 Black avoids 16…c5?! 17.♖xf6! gxf6 18.♘xd5 with good play for the exchange, like in similar cases. 17.a3 ♖bb8 18.b4 This set-up is a good alternative to the c2-c4 plan. White tries to x the pawns on light squares and ideally blockade them with ♘a4 and ♔f2-e3 etc. 18…♘d7 19.♘a4 ♖fc8 20.♖fb1 ♖c7 21.♔f2

Conclusion As usual the ‘main line’ is the main line for a reason. After 4.d4 Black can more or less force this line as there is no option to deviate for White. Black is solidly placed here and very close to equality for sure. However, in particular in many of the endgames I think White’s position is easier to play and Black has almost zero chances to ever win, while we can try for small things. My experience with the Scotch Four Knights in general is: there is still enough going on to beat players that are weaker than you – and against stronger opposition you are very safe.

111

112

Sample games

 

Sample games Game 1 KP 3.11 (C41) Giovanni Vescovi 2572 Jefferson Pelikian 2415 Sao Paulo 2001 (5)

1.e4 e5 2.♘f3 d6 3.d4 ♘d7 As discussed in the theoretical section, this move order to reach the Closed Philidor (sometimes also called Hanham or Black Lion set-up) is imprecise. The modern way to enter the Philidor is 1.e4 d6 2.d4 ♘f6 3.♘c3 e5 and now 4.♘f3 ♘bd7 5.♗c4 ♗e7 etc. 4.♗c4! Now Black won’t manage to complete his set-up with …♗e7, …c7-c6 etc. Please see the theoretical section for details. 4…c6 5.0-0 ♗e7 6.dxe5 dxe5 7.♘g5 ♗xg5 8.♕h5 ♕e7 9.♕xg5 ♕xg5 10.♗xg5

This unattractive ending for Black is still one of the better scenarios he may reach. At least there is no immediate disaster, but playing against the bishop pair and with weak dark squares is no fun. I selected this game to annotate because it shows very nicely how to exploit these long-term advantages. 10…♘gf6 Instead 10…♘c5 was played in AnandIvanchuk, Monaco rapid 1996. Sometimes even top players mess up move orders and stumble into bad lines: 11.f3 ♗e6 12.♗e2 ♘f6 13.♘d2 ♘fd7 14.♘c4 f6 15.♗e3 ♗xc4 16.♗xc4. White is better, but Black managed to hold a draw. 11.f3 It is important to understand that 11.♘c3?! is dubious here as the knight has no useful role on c3. The d5- and b5-squares are safely guarded and thus the knight is misplaced.

11…b5?! This makes things even worse for Black, as Vescovi nicely demonstrates in the game.

113

114

Sample games

11…♘c5 was making it more di cult for White: 12.♘d2. Now Black should just wait with a move like 12…♘e6. If he plays 12…♗e6?!, 13.b4! is a nice idea to take note of: 13…b5 14.♗xb5 cxb5 15.bxc5 ♘d7 16.♗e3 ♖c8 17.♖ab1 a6 18.c4 and White will be up a pawn in this endgame. 12.♗e2 ♘b6 13.♘d2 ♗e6

the position might lead to situations where Black sets up a fortress-like defence – worse, but tough to crack down. 17…cxb5 18.♖c6 ♖b8 19.b3 0-0 It is a bit surprising that Black still castles here, but keeping the king in the centre also had drawbacks. After 19…♔e7, 20.♖c7 is very annoying, as the natural 20…♖hc8?? blunders a piece due to 21.♗xb6. 20.a4 bxa4 21.bxa4 a5 22.♗b5

14.♖fc1 This is the key concept to remember. The pawn push c2-c4 will open up the queenside to White’s advantage, making full use of the bishop pair. By the way, this position had been reached before and the later World Championship challenger Veselin Topalov went for a similar approach: 14.b3 ♘fd7 15.♖fc1 f6 16.♗e3 ♔e7 17.c4 bxc4 18.♘xc4 ♘xc4 19.♗xc4 ♗xc4 20.♖xc4 with a clear advantage in Topalov-Franco Ocampos, Pamplona 1994. 14…h6 15.♗e3 ♘fd7 16.c4 a6 17.cxb5 This is the most logical play, opening the position. A line like 17.c5 ♘c8 18.a4 is of course also very good for White, but closing

This is the position White had in mind when he traded on b5. The bishops dominate the board and the a5-pawn is a lot easier to attack than its colleague on a4. 22…♖fc8 23.♖ac1 ♔f8 24.f4 There was nothing wrong with further improvement using small moves like 24.♔f2, but playing concretely also has its points. ‘Strike while the iron is hot’ is just as good an advice as ‘Don’t rush!’ 24…exf4 25.♗xf4 ♖xc6 26.♖xc6 ♖b7 27.♗d6+ ♔g8 Instead 27…♔e8 28.♘f3 ♔d8 29.♘d4 is close to a zugzwang already. 28.♔f2 ♗g4 29.♖c1

115

116

Sample games

Black’s last move had an idea! 29.♔e3? ♗d1 and White has spoiled it. 29…♗e6 30.♔e3

30…f5 After 30…f6 31.♔d4 ♔f7 White needs to decide if he likes to grind on or trade down to the following ending. 32.♖c7 ♖xc7 33.♗xc7 ♔e7 34.♗xd7 ♘xd7 35.♗xa5 is winning with precise play, but the opposite- coloured bishops are somewhat worrisome. 31.e5 ♗d5 32.g3 ♔f7 33.♔d4 This is not optimal, but it does not spoil the huge advantage. A prep move like 33.♖c3 was better, followed by ♔d4 after all. 33…♔e6 33…♘f8 was tougher, but it won’t save the game. 34.♔c3 ♘e6 35.♘b3 ♗xb3 36.♔xb3 still wins in the long run. 34.♖c3 g5 35.♘b3 f4 36.gxf4 gxf4 37.♖h3 ♗xb3 38.♖xb3

As a true bishop pair fan I need to put a diagram here! 38…♖a7 39.♗c6 ♘c8 40.♗d5+ ♔f5 41.♗e4+ ♔g4 42.♗a3 ♘db6 43.e6 ♘xa4 44.♔e5 White now has mating ideas, starting with ♗f5+. 44…♘ab6 45.♔f6 The engine sees a ‘mate in 13’. I don’t, but it’s easy to see that Black is lost. 45…♔h4 46.♗c5 ♖a6 That allows a quicker nish. 47.♗f2+ 1-0

47…♔g4 48.♗f5+ ♔h5 49.♖h3# is something even a human can calculate. The line starting with 3…♘d7 is basically extinct on the professional 117

118

Sample games

level, but still occurs quite frequently below that. This game is an excellent example how to press in this very favourable endgame. Game 2 KP 4.4 (C41) Judit Polgar 2682 Bu Xiangzhi 2695 Khanty-Mansiysk ol 2010 (4)

1.e4 e5 2.♘f3 d6 3.d4 exd4 The only fully respectable choice for Black at this point. It pays o to learn the dubious/bad moves like 3…♘d7 as they are still played frequently on the non-professional level. 4.♘xd4 ♘f6 5.♘c3 ♗e7 6.g3 This seems to be the most testing for Black. White goes for a central grip on the d5-square. 6…0-0 6…d5 must be the critical move. White seems to be slightly better here as well, as demonstrated in the theoretical section. 7.♗g2

From here Black will have a hard time playing …d6-d5 under satisfactory conditions. White’s main objective is to prevent the central push or make it unattractive. At the same time he should strive for further expansion, in particular on the kingside. 7…♘c6 8.♘xc6 Keeping it simple! White could also have kept the tension with castling, for instance, but the capture makes sure we know what to play for: pawn structure. Note that Black gets something as well with the d5-square now being covered. 8…bxc6 9.0-0 ♘d7

This set-up is sensible for Black. He is worried that White will play e4-e5 soon, so he increases his control over this key square. Judit Polgar now goes for White’s most promising piece development scheme. 10.b3 Against the half-open b- le the anchetto is often a good recipe, protecting the queenside. In this particular case b2 is a good square for White’s bishop, countering Black’s piece on f6. 119

120

Sample games

10…♗f6 11.♗b2 ♖b8 12.♕d2 ♖e8 13.♖ad1

A very harmonious set-up for White. But what to do next? White has some ideas to choose from. A logical trade would be that of the dark-squared bishops, as Black’s bishop on f6 is his most active piece. In terms of pawn structure changes, f2-f4 and possibly e4-e5 comes to mind. Whether e4-e5 is possible is, of course, very much dependent on the concrete tactics. One important point: Black needs to make moves as well… and oftentimes White’s plan will shape up in reaction to what Black does. 13…c5 A clear case of the nal point just mentioned. Black changes the structure in order to operate with …c5c4 and possibly place the bishop on b7. 14.♗c1 This is not the most obvious move, and probably also not the best one available. I like 14.♘a4 best in this position. As mentioned, it makes sense to trade the bishops. In addition, the queen has the occasional idea to go to a5, targeting 121

the queenside: 14…♗xb2 15.♘xb2 ♘b6 (in order to avoid ♘c4, but now White switches to f2-f4/e4-e5 ideas) 16.f4 ♗b7 17.e5.

analysis diagram

And White maintains a nice pull in the centre. 14…♘b6 15.♖fe1 ♗g4 16.♘e2 Black has certainly improved his situation and now spots a good opportunity to possibly repair the pawn structure. 16…a5 17.h3 17.♕xa5 ♖a8 18.♕d2 ♖xa2 is obviously ne for Black. 17…♗xe2 The trade on e2 is mandatory. Instead, 17…♗d7 18.a4 is awkward for Black. 18.♖xe2 a4 So Black gave up the pair of bishops to improve his pawn formation. I think White is slightly better, but Black is quite solid. White wants to open the position now, improving the longrange bishops. 19.♔h2 ♕c8 20.f4 axb3 21.cxb3 ♕a6 Here 21…♗d4 is the engine suggestion. 22.♕c2 ♕a6 23.♗e3 ♗xe3 122

Sample games

24.♖xe3 is a likely line to follow, transposing to the game.

analysis diagram

22.♕c2 Surprisingly Judit Polgar misses 22.e5!. This is not easy to calculate, but she was one of the best attacking players of all time – fully capable of doing it. Let’s check what she might have overlooked: 22…dxe5 (after a move like 22…♖bd8 White will coordinate her pieces with 23.♕c2 and White is much better without the need to nd anything special) 23.fxe5 and now: A) 23…♗g5 24.♕e1 ♗xc1 25.♖xc1 – Black is in terrible shape here. The cpawn is doomed and e5-e6 is on the cards, starting an attack; B) After 23…♗xe5…

123

I suspect the following blow was missed by Polgar: 24.♖xe5! ♖xe5 25.♗b2 and Black is completely busted. He cannot retreat the rook due to 26.♕g5… and is out of options: 25… ♕e2 26.♗xe5 ♕xe5 27.♕d8+ ♕e8 28.♕xc7 and White wins easily. A good position to show a clearly superior bishop against an unlucky knight. 22…♗d4 23.♗e3 Trade the opponent’s best piece! Simple, but e ective. 23…♗xe3 24.♖xe3 Black now faces the e4-e5 push. It is impossible to avoid completely. 24…♖e7

124

Sample games

25.e5 c4 Black tries to at least weaken White’s structure a bit, but it does not help much. 26.bxc4 dxe5 27.c5 ♘d7 28.c6 ♘f8 29.fxe5 White has won a pawn and is still rmly in control. 29…♕a7 30.♖e2 ♖be8 31.♖d5 ♕a6 32.a4 ♘g6 39.♖d7! I think it is quite amusing that the rook is given up for the bad knight in order to show how strong the bishop is. 39…♘xd7 40.cxd7 ♖d8 41.♗c6

33.e6! A very nice and active solution. White gives back the rather unimportant pawn and now convincingly converts. 33…fxe6 33…♖xe6 34.♖xe6 fxe6 35.♖b5 is also crushing. 34.♖d7 ♘f8 35.♖xe7 ♖xe7 36.♖f2 ♕a5 37.♕d2 ♕xd2 37…♕xa4 38.♕d8 ♖f7 39.♖xf7 ♔xf7 40.♕xc7+ is even easier for White. 38.♖xd2 ♖e8

125

The pawn is unstoppable, ultimately leading to White being a piece up. 41…♔f7 42.a5 ♔e7 43.a6 ♔d6 44.♗b7 ♔xd7 45.a7 ♔d6 46.a8♕ ♖xa8 47.♗xa8 ♔e5 48.♔g2 ♔d4 49.♔f3 c5 50.♗e4 h6 51.♗g6 c4 52.♔f4 ♔c3 53.♗f7 g5+ 54.♔e3 ♔b2 55.♗xe6 c3 56.♗f5 c2 57.♗xc2! ♔xc2 58.g4 1-0 A very nice display of White’s chances in the main line of Philidor’s Defence. After 6.g3 White enjoys good central control and can play for a slow buildup quite often. In the game Judit Polgar 126

Sample games

got excellent attacking chances, but missed a strong pawn push on move 22. It was still more comfortable for White and tough for Black to defend, a task that ultimately proved to be too di cult. Game 3 RG 7.4 (C42) Zaven Andriasyan 2631 Karen Grigoryan 2563 Lake Sevan 2013 (8)

1.e4 e5 2.♘f3 ♘f6 3.♘c3 ♗b4 4.♘xe5 0-0 5.♘d3 ♗xc3 6.dxc3 ♘xe4 7.♗e2 d5 8.0-0 c6 Black has the principled choice of a light- or a dark-squared set-up. In my mind, putting the pawns on dark squares is more in line with having a light-squared bishop, but tastes may di er. In any case, the line starting with 3…♗b4 feels comfortable for White. 9.♗e3 I suggest to start with 9.f3, but the game quickly enters a similar path. 9…♘d7 10.f3 ♘d6

Black’s main issue in this line is the lack of good, safe outposts for the knights. The square d6 is not great, but f6 does not look any better with the white pawn being on f3. White’s general strategy is about slowly expanding to exploit the pair of bishops in the long run. The best situation for the bishops is a position that can slowly be opened up without giving the enemy knights dynamic possibilities. 11.♖e1 b6 With the black knight being directed towards f5, this bishop deployment makes sense. 11…♖e8 is an alternative way to develop: 12.♗f2 ♘f8 13.b3 ♗f5 14.♕d2 and White enjoys the standard long-term pull with his bishop pair. The key to success is a slow building of the position, expanding slowly. Pushing g2-g4 on the kingside is an idea that should be carefully prepared. 12.♗f1 ♗b7 13.b3 ♘f5 14.♗f2 This is the target set-up to aim for. Black’s knights are limited; now White tries to gain space slowly. 14…♕f6 15.♕d2

15…g6 127

128

Sample games

This is an unfortunate decision, probably underestimating White’s reply. It was better to try 15…♖fe8 to avoid the weakening of the dark squares. White is still comfortably better, of course: 16.♕f4 (not as e ective as after 15…g6, but still strong) 16…c5 17.g4 ♘e7 18.♕xf6 ♘xf6 19.a4. This is a sensible way to continue. White has gained space and a4-a5 is a threat to Black’s queenside. This is a tough position to defend for Black. 16.♕f4 c5 A) 16…♕d8 avoids the queen trade, but still is rather sad: 17.a4 a5 18.♖ad1 and c3-c4 is the next mini plan for White, who is clearly better; B) 16…♕xc3? loses immediately; 17.g4 ♘g7 18.♗d4 ♕a5 19.♕c7 is one way to end the game. 17.g4 ♘d6 18.♗g3 This keeps a large advantage, but it was even stronger to play 18.♕xf6 ♘xf6 19.♗h4!.

analysis diagram

This is the important point. Now both knight moves to e8 either lose the 129

exchange or allow 20.♖e7, so …♔g7 is provoked: 19…♔g7 20.♗g3! and ♗e5 will now involve a deadly pin. On top of that, 21.♖e7 is the second threat that Black has to face. This position is not defensible. 18…♘e8 19.♕xf6 ♘exf6 20.♔f2 Compared to 18.♕xf6, Black’s position is slightly better organized, so he is not in danger of losing quickly. The danger of losing slowly however is very substantial, as White has a very promising position. 20…c4 Black was probably quite desperate, but this worsens the situation. Giving up on even more dark squares (d4!) helps White a lot. 21.♘e5 ♘c5 22.♗h4 Strong and concrete play. Black has no good reply to this attack on the knight. 22…♘fd7 If 22…♔g7, 23.b4 already wins decisive material. 23.♘xd7 ♘xd7

24.♖e7 Now d7 and c4 are attacked. Black’s position is falling apart. 24…♗c6 25.bxc4 ♘c5 130

Sample games

Or 25…dxc4 26.♖d1 (even stronger than taking on c4 immediately) 26… ♘c5 27.♗xc4 and White is winning. 26.♖d1 ♘a4 Instead, 26…dxc4 27.♗xc4 leads to the line mentioned above. 27.♗f6 This is enough, but the more forceful 27.cxd5 ♘xc3 28.dxc6 ♘xd1+ 29.♔g3 ♖ac8 30.c7

analysis diagram

would have shortened the game. 27…dxc4 28.♖d6 28.♗xc4 ♘b2 29.♖d6 was a good alternative, also keeping a winning advantage. 28…♗b5 29.♖d5 a6 30.g5 ♖fe8 31.♖xe8+ ♖xe8 32.♖d6 ♘c5 Black gives up a pawn in order to get the knight back into the game and trade o the monster on f6. The alternative 32…♖b8 keeps the pawn, but the position of White’s f6bishop is just too dominant. For example, 33.♔e3 ♘c5 34.f4 ♖e8+ 35.♗e5 ♘d7 36.♔d4 is a possible line to continue. 33.♖xb6 ♘d7 34.♖b7 ♘xf6 131

34…h6 35.♗h3 ♘xf6 36.gxf6 ♖d8 37.♖b6 is also winning for White, but was a bit tougher. 35.gxf6 h6 35…♖d8 36.a4 ♗xa4 37.♗xc4 ♗e8 38.♗xa6 and the c-pawns will win, just like in the actual game. 36.a4 ♗c6 37.♖a7 ♗xa4 38.♗xc4 ♖f8 39.♗d3 ♖c8 40.♖xa6

The extra c-pawns win quite easily. White is well organized and Black has no counterplay at all. 40…♗d7 41.c4 ♗e6 42.♖a4 ♗d7 43.♖b4 ♖c6 44.♗e4 ♖xf6 This pawn has done its job, now it’s all about the c-pawn. 45.c5 ♖a6 46.♖b6 ♖a5 47.c6 ♗e6 48.c7 White has calculated this until the end of the game, foreseeing that the second c-pawn plays a signi cant role. If you spot a forcing line that is completely straightforward, you should continue to calculate and take it as deep as possible or necessary. If there is just one reply for your opponent in a certain position, you need to include that reply and nd your move in advance. Visualizing a deep line can be 132

Sample games

tough, but is often the key to win, especially in an endgame. 48…♖c5 49.♖c6 ♖xc6 50.♗xc6 ♗c8 51.♔e3 ♔f8 52.♔d4 ♔e7 53.♔c5 g5 54.♔b6 f5 55.♗b7 ♔d7 56.c4 1-0

Black is helpless against c4-c5-c6. This game nicely shows the long-term assets White has in this line. The pair of bishops is very annoying to face for Black. White enjoys a risk-free pull and there is little counterplay. That being said, the pawn structure is symmetrical, and breaking down Black’s defences requires patience and precision. Game 4 SO 3.2.5 (C45) Wang Hao 2752 Viswanathan Anand 2772 Wijk aan Zee 2013 (13)

you have gained some experience with the Scotch Four Knights. 3…exd4 4.♘xd4 ♘f6 The main line, but 4…♗c5 enjoys a good reputation as well. Against this move White has a wide range of options, including the solid 5.♘xc6 and the sharper idea of 5.♘b3, intending to play ♘c3/♕e2/♗e3 soon to castle queenside. 5.♘c3 This transposes to our repertoire. Here White has the additional option to play 5.♘xc6 bxc6. Now White has a choice between the sharp and heavilyanalysed 6 e5, the tricky move 6.♕e2, and 6.♗d3, which is quite similar in nature to the Scotch Four Knights. 5…♗b4 6.♘xc6 bxc6 7.♗d3 0-0 8.0-0 d5 9.exd5 cxd5 10.h3 c6 This is the main line by far. It is very logical to protect the d5-pawn and thus add more exibility to the pieces. Covering the b5-square is also very useful as Black will most likely improve his bishop with …♗d6 later and wants to avoid ♘b5 in response. 11.♕f3 ♗d6 12.♖e1 I recommend to play 12.♗f4, but the di erence is minor. In fact, we will very soon transpose to a position which was examined in the theoretical section. 12…h6 13.♗f4

1.e4 e5 2.♘f3 ♘c6 3.d4 I recommend to reach the Scotch Four Knights via 3.♘c3, but the ‘regular’ Scotch move order is an excellent option to expand your repertoire once

133

134

Sample games

14.♕xf4 ♗e6

This position is part of our repertoire via 11…h6 12.♗f4 ♗d6 13.♖e1. White is happy to trade the bishops in these structures. There are multiple reasons for that. On a very general note, White’s pawn structure is favourable, and trading minor pieces should emphasize that (similar to trading in IQP and hanging pawns scenarios). More concretely, White may be able to x the pawns on light squares and use the c5-square for a knight – this is only imaginable without the bishop on d6. Another common scenario is White pushing c2-c4 and taking on d5 later to give Black the IQP. Again this is more favourable without the bishops, with better control over the key squares d4, c5 and e5. 13…♗xf4 The main continuation is 13…♖b8, as examined in the theoretical section. I think this is more precise than Anand’s move, which gives White more control over the dark squares. It is not a serious mistake, of course, but little things tend to add up to real problems for Black in the main-line Scotch Four Knights. 135

15.♘e2 The knight is clearly misplaced on c3, so regrouping it is a standard manoeuvre. By removing the knight, the pawn lever c2-c4 also comes into play. Stock sh suggests 15.b4 instead, aiming for a blockade on the dark squares. This is also going to be eternally better for White, but only very, very slightly. Wang Hao’s move is more exible regarding the pawn structure, keeping the option c2-c4. Nevertheless, let’s look at 15 b4 some more: 15…♕b8 16.♕d4 ♕d6 17.♘a4 a5 18.a3 axb4 19.axb4.

136

Sample games

16…♖e8 was keeping the tension and is more precise: 17.♕xb8 (not forced, obviously; you may also play with ♕d2, ♘f4 and later aim for c2-c4) 17… ♖axb8 18.♘f4. Compared to the game Black has gained time. It is still not clear if he has equalized completely though. 17.♘xf4 ♖fe8 analysis diagram

This is the type of position White will get after 15 b4. Black is limited to just sitting in a slightly passive position, but it is di cult for White to improve with a straightforward plan. Still, this position type is extremely safe for White, and excellent if you want to play with the draw in hand. 15…♕b8 15…c5 is a principled move, but Black’s pawn structure remains vulnerable: 16.c3 (this is not the only option, of course, but I like the regrouping illustrated by the following line) 16… ♖e8 17.♕d2 a5 (it is not clear that this move is necessary, but Black has to watch out for b2-b4 and often prefers to make this harder to play for White) 18.♘f4 (this was White’s intention. The knight looks at d5 and e6 now) 18…♗d7 19.♖xe8+ ♗xe8 20.♗e2. This bishop is directed to f3, putting more pressure on the centre. White is still exerting some pressure while running no risks at all. 16.b3 ♕xf4 This feels a bit rushed.

137

Against 17…♗d7 the typical pawn lever 18.c4 will improve White’s structural edge even more. Black will get a second isolated pawn, either on d5 or c6. This coming endgame is very one-sided and gives excellent chances to press for a win. 18.♘xe6! Keep it simple! After this trade the remaining minor pieces favour White – the bishop is just a lot stronger than the knight. 18…fxe6 The alternative 18…♖xe6 does not improve Black’s task: 19.c4. Again this move to attack the pawn on d5. White’s bishop is a lot stronger here, with its ability to support play on both sides of the board. An additional bonus is the 138

Sample games

protection of the possible entry squares c2 and e2 – not so relevant at the moment, but it might matter later. 19.♗g6 An inviting post for the bishop. 19…♖e7 20.c4 White plays this anyway, even without the white bishop being able to recapture on c4. 20…♘d7 20…dxc4 is a clear concession that leads to an excellent position for White: 21.bxc4 and White has ideas like ♖a1-d1-d6 and ♖e5 to improve and increase the pressure on Black’s pawns. 21.♖ac1 The c- le will open up in short order. 21…♘f8 22.♗d3 ♔f7 23.cxd5 cxd5

Moving the majority makes a lot of sense, of course, but I think 24.f4 would have been good as well. It helps to x the black centre and enables ♔f2e3 etc. 24…♘d7 25.♖c6 ♘f6? This natural-looking move is Black’s rst serious mistake. He was under some nagging pressure, but it was not yet a serious issue. 25…♘b6! was a better defence, stopping the a-pawn for the time being. White is still for choice, but at least there is no obvious way forward. White should improve with moves like 26.♔f1 and be patient. 26.a4 Now White’s play is a lot quicker on the queenside while Black has no good counterplay. 26…e5?! It is very understandable that Anand does not want to wait forever and goes for an active-looking move. Still it seems that not moving the pawns in the centre was more tenacious, for example with 26…♖b8. 27.f3!

A good moment to asses the current situation. White has some pluses that add up. Better pawn structure (2 pawn islands vs 3) with the prospect to create an outside passed pawn, better minor piece and control of the only open le. Black should still be able to defend the position, but it is not fun to play at all. 24.b4

139

140

Sample games

This allows ♔f2 and makes …e5-e4 less attractive for Black. 27…♖d8 28.♔f2 Here 28.a5 was even stronger. After 28…e4 29.fxe4 dxe4 30.♗c4+ ♔f8 31.b5 White’s advantage is nearing decisive proportions. 28…e4 29.fxe4 dxe4 Or 29…♘xe4+ 30.♔g1. This will be very hard to defend for Black – knights are notoriously bad at ghting outside passers… and this scenario will occur. 30.♗c4+ ♔g6 30…♔e8! was a better try. After 31.♔e3, now 31…♘d5+ is possible, forcing the minor pieces o . This certainly helps Black, but it is still a long way to draw after e.g. 32.♗xd5 ♖xd5 33.♖c8+ ♔f7 34.♖f1+ ♔g6 35.♖c6+ ♔h7 36.♖c3 and White’s more active king is a serious plus. 31.♔e3

31.a5 was a better choice, but Black’s possible counterplay was tough to anticipate. After 31 a5 White should win comfortably. 31…♔h7?

141

The engine suggestion 31…♔g5! is surprisingly strong. Black’s idea is to simply invade with …♔g5-h4-g3 ! Not the most natural plan ever, and di cult to nd in what was likely a time-trouble scenario: 32.♖c5+ (32.a5 ♘d5+ 33.♗xd5 ♖xd5 should also draw for Black) 32…♔h4 33.♔f2 ♖d2+ 34.♗e2 e3+ and again Black’s sudden activity should secure the draw. 32.♖c5 In order to avoid the check on d5. White is back in control. 32…♖ed7 32…♖b7 was a bit tougher, but I don’t think Black will hold in the end. It is just too di cult to defend: 33.b5 ♘d7 34.♖d5 ♖c8 35.♖d4 ♘f6 36.♖b1. This is a possible continuation after 32…♖b7. White is still clearly better. 33.a5 ♖d2 34.♖e2 ♖d1 35.a6 Fixing the target on a7. 35…♘d7 36.♖c7 ♘e5 37.♗e6 ♔g6 38.♖xa7

Finally White has transformed his strategic advantage into material – now the a-pawn will decide the game quite easily for White. 142

Sample games

38…♖1d6 39.♗a2 ♘c6 40.♖b7 ♖d3+ 41.♔f2 ♘d4 42.♗f7+ ♔f5 43.♖e1 ♖d2+ 44.♔g1 ♔f4 45.♗c4 g6 46.a7 ♔g3 47.♗f1 ♘f3+ 48.gxf3 exf3 49.♖f7 1-0 At rst glance you might think the endgame arising after move 17 is nothing to write home about, but in fact the small advantages add up. Vishy Anand enjoys a reputation as being an excellent defender, but he could not cope with the job in this game. One issue is the lack of a clear set-up or goal to play for as Black. You are just going to be slightly worse forever. Game 5 SO 3.2.5 (C47) David Klein 2509

In the theoretical section I recommended to play 17.a3. With a2a3 and b2-b4 my goal is to x the pawns on the light squares. However this ending allows for more than one approach, and Klein does not mind allowing Black to play …c6-c5. While I prefer my recommendation, I chose to present this game anyway because White demonstrated excellent play against the black pawns, which is helpful to understand the general strategy of this line. 17…♖bb8 The rook on b4 was not doing that much, so Black decides to connect it again with its colleague on f8. 18.♖f4

Rudy van Wessel 2398 Netherlands tt 2015/16 (3)

1.e4 e5 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4 exd4 5.♘xd4 ♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3 c6 11.♕f3 ♗d6 12.♗f4 ♖b8 13.b3 ♖b4 14.♗xd6 ♕xd6 15.♕g3 ♕xg3 16.fxg3 ♗e6

This is a strong move that demands a precise answer. This position, with Black to move, would be an excellent prophylactic exercise. 18…c5? Black needed to play 18…a5! 19.♘a4 (19.♖a4 ♖a8 leads nowhere for White) 19…♘d7 20.♗a6 ♖fe8 21.♖e3. This introduces the idea to play ♖c3, attacking the c-pawn. I’d still take White in this position, but Black is not

17.♖ae1 143

144

Sample games

under as serious pressure as he was in the actual game. 19.♖a4 ♖b7 20.♗a6

Before White relocates the bishop, he forces the black rook to a less favourable square. Note that now the a4-rook is always able to go to a5, where it is stable and exerts pressure on both the central pawns and on a7. This is a lot better for White than in the line after 17…a5!. 20…♖c7 21.g4 A useful insertion. Black shouldn’t allow g4-g5, so… 21…h6 22.♖e5 22.♗f1 was better, and is actually played two moves later. 22…♘d7 23.♖e1 23.♘xd5? ♗xd5 24.♖xd5 ♘b6 would have been a sad end to the ‘active rooks’. 23…♘f6 I don’t think Black had anything substantially better than trying to repeat. 24.♗f1 Now White is back on track. 24…♖b8 145

25.♖a5 It was also very promising to go 25.♘b5 ♖cb7 26.♖a5. The idea is to play a2-a4 and then take on a7. Simple, but how to meet this idea? 26…♗d7 keeps the pawn, but it is still nagging pressure after 27.a4 ♗xb5 28.axb5. In the game, Klein follows a di erent path, which is also very much worth noting. He intends to increase the pressure on d5 by playing ♖d1 and placing the bishop on f3. 25…♔f8 26.♖d1 ♖bb7 27.♗e2 ♖b8 The lesser evil was 27…c4 28.♘b5 ♖c5 29.bxc4 dxc4 30.a4. This is not fun to play for Black, but better than the game. 28.♗f3 d4

146

Sample games

By now 28…c4 29.♘xd5 ♗xd5 30.♗xd5 is just a clean pawn down. Maybe this was still o ering better saving chances than the game. 29.♘b5 ♖d7 30.♘xa7 ♗d5 31.♘c6 ♗xc6 Not a happy move, but it was debatable if the alternative puts up more resistance: 31…♗xf3 32.gxf3 ♖c8 33.♘e5 ♖e7 34.♘c4 seems very comfortable for White as well, with no apparent obstacles on the way to convert this advantage to a full point. 32.♗xc6 ♖c7 33.♗f3

41.♖xc4 sees the rook ideally placed, with c2-c3 coming soon. 41.♖d2 ♖d6 42.♔f2 ♖d8 43.♔e2 ♖h8 44.♔d1 White is playing it slowly, not rushing anything. The ‘Don’t hurry!’ motto in one-sided endgames is very e ective, more often than not because the opponent will hit the self-destruct button in his passive and unattractive situation. 44…h5?

Having a passed a-pawn with a bishop on f3 on an open diagonal is one of the endgame situations where the di erence between bishop and knight is demonstrated very cleanly. Without rooks, it would be extremely easy to win. In the game it’s a bit more work, but Black has no serious counterplay. 33…♘d7 34.♔f2 ♘e5 35.♗e2 f6 36.♗f1 ♔f7 37.♔g3 ♖d8 38.♖a4 ♖d6 39.♗b5 ♖b6 40.♗c4+ ♔e7 On general grounds Black would love to take the bishop, but 40…♘xc4

Black had to continue to wait, for instance with 44…♖b8. A logical continuation of White’s play would now be 45.♔c1 ♖b6 46.♔b2 ♖b8 47.c3, to move forward. 45.g5! Now Black gets additional weaknesses on the kingside. This is too much to bear. 45…h4 45…fxg5 leads to an immediate disaster: 46.♖e2 ♔d6 47.♖a6+ ♘c6 48.♖e6+ ♔d7 49.♖g6. 46.gxf6+ gxf6 47.♖e2 ♖h5 48.♖e4 ♔d6 49.♖f4 The white rooks really do all the heavy lifting in this game.

147

148

Sample games

49…♖h6 50.♗b5 ♔e7 51.♖a8 ♔f7 52.a4 ♔g6 53.♖g8+ ♔f7 54.♖b8 ♔g6 55.♖g8+ ♔f7 56.♖d8 ♔g6 57.♖f2 ♔g5 58.b4 Finally making some progress. 58…cxb4 59.♖xd4 ♖h8 60.♖xb4 ♖d8+ 61.♖d2 ♖dc8 62.♗a6 ♖a8 63.♗b5 ♖c3 64.♔c1 ♖ac8 65.♔b2 ♖3c5 66.♗a6 ♖8c6 67.♗f1 ♖c7 68.♖f2 f5

69.♖b5 It is useful to trade one rook. After the trade, the king can be used more actively to support the pawns as there is less danger with just one rook left. 69…f4 70.♖xc5 ♖xc5 71.♔b3 f3 72.gxf3 ♔f4 73.♖g2 ♘xf3 74.♗d3 ♔e3 75.♔b4 ♖c8 76.a5 ♘e5 77.♖e2+ ♔d4 78.♖e4+ ♔d5 79.♖xh4 1-0 The endgames after the trade on g3 occur quite frequently in the main line of the Scotch Four Knights. Klein’s play in the examined game is showing nicely why White has the better end of a position that is often called ‘equal’. Black is just sitting with little ideas while we can still try for little things. I have played 1.e4 e5 on the black side for a long time and never enjoyed the

149

prospect of getting into those endgames – it’s just no fun to play at all. Game 6 SO 3.2.5 (C45) Sergey Rublevsky 2693 Evgeny Najer 2664 Eilat EU-Cup 2012 (3)

1.e4 e5 2.♘f3 ♘c6 3.d4 exd4 4.♘xd4 Sergey Rublevsky has relied on the Scotch Game his whole career, playing almost all major lines to vary his play while still sticking with the typical open game type of the Scotch. His games are a great source of ideas and are well worth studying. 4…♘f6 5.♘c3 ♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 0-0 9.0-0 cxd5 10.h3 Kramnik brought this move back onto the map in early 2012 – already some months later it became trendy and was played quite frequently. 10…c6 11.♕f3 ♖e8 12.♗f4 ♗e6 13.♘e2 ♗d6 14.♖fe1 ♕b8 15.b3

Very standard moves by both sides. If you have forgotten preparation in the main line Scotch Four Knights, just go 150

Sample games

with the standard h3/♕f3/♗f4/♘e2 set-up. Quite frequently this will be the best set-up, and it will always be a reliable solution even if you missed a chance to do something slightly better. 15…♗d7 Attention: with this innocent- looking move Black threatens to take on e2! 16.♗xd6 ♕xd6 17.c4

22.♖c1 ♖c8 23.♖xc8+ ♗xc8 24.♘e2 ♔f8 25.♘d4 and Black still needs to be precise to draw; 2) 19…♕a3 looks a bit more active than the trade, after which Black basically gives up any hope of ever winning the game: 20.♖ed1. Now Black needs to avoid the natural 20… a4?

White has accomplished the standard pawn lever against Black’s centre, trying to weaken it even more. 17…d4 This is the most ambitious and best move in my opinion. Engines also like to keep the tension and evaluate the resulting position as roughly equal. As usual I think it might be very close to equal, but White’s position is easier to play for sure. We need to check Black’s alternative 17…a5. I suggest to play 18.cxd5 cxd5 19.♕f4 and now: 1) 19…♕xf4 20.♘xf4. Black should draw this endgame, but White still has some ideas. You have the potential outside passer and d5 remains a liability. Let’s have a closer look at a possible line: 20…♖xe1+ 21.♖xe1 ♖e8

analysis diagram

due to 21.b4 and Black is in trouble, as 21…♖ab8 22.♖ab1 ♕xa2 23.b5 is very awkward. The queen is short of squares and the b-pawn is dangerous. 18.♕f4 Going for the endgame. White players also have tried a forced solution with 18.c5 ♕xc5 19.♖ed1, intending to get the pawn back with a ♘xd4 tactic. However, after 19…♖ad8 this attempt leads nowhere; for example, 20.♖ac1 ♕b6 21.♘xd4 ♗xh3 22.♕xc6 ♗g4 holds for Black. 18…♕xf4 19.♘xf4 c5 20.♗f1 The knight is best placed on d3, in front of the pawn and at the same time attacking c5. 20…♗c6 21.♘d3 ♘d7

151

152

Sample games

This whole sequence was very natural after the queen trade.

White is not better in this position, but he can still try to pose problems. We will see that Najer in fact very quickly goes wrong. 22.g4 A useful space gainer that prepares ♗g2 in addition. 22…f5? This is a surprising mistake by Najer, who is a very strong GM. Black should have played 22…a5.

b3-b4 to undermine the black pawn chain. It is a bit di cult to analyse such a position move-by-move. My suggestion is to practice it by playing against an engine with both sides. It will be interesting to see that very drawish-looking positions still contain ideas. 23.gxf5 ♔f7 Maybe he intended 23…♗e4, but missed 24.♖xe4 ♖xe4 25.♗g2 with a white advantage. 24.b4? I wonder what Rublevsky disliked about 24.♗e2. As far as I (and Stock sh!) can see Black is just worse with simple lines, e.g. 24…♔f6 25.♗h5

analysis diagram

and Black is in trouble. 24…cxb4 25.♘xb4 ♗a4 26.♗g2 ♖xe1+ 27.♖xe1 ♖e8?

analysis diagram

This is an equal endgame, but that does not mean it is impossible to win. There are still ideas left, like trying to play for 153

154

Sample games

The best choice was to go 27…♖b8. Now 28.♘c6 ♗xc6 29.♗xc6 ♘c5 should hold for Black. White now has a forced line to reach a knight ending. 28.♖xe8 ♔xe8 29.♗c6 ♗xc6 30.♘xc6 This is very critical for Black. I am not sure if perfect play would save the draw for Black, but in a practical game White has excellent chances to win. 30…d3? It seems the best defence was 30… ♘b6. After 31.♔f1 ♔d7 32.♘xd4 ♘xc4 33.♔e2

with best play. After the game move I believe it is lost for Black. 31.♔f1 a6 31…♘b6 32.♘e5 ♔e7 33.c5 looks like a win for White as well. 32.♔e1 ♘c5 33.f3 ♔d7 34.♘e5+ ♔d6 35.f4 h5 36.♔d2 ♘e4+ 37.♔xd3 ♘f2+ 38.♔e3 ♘xh3 39.♘f3 ♔e7 40.♘g5 And Black resigned. As it happens quite often in the Scotch Four Knights, we reached an endgame quite quickly. This game shows that even very strong GMs can easily misplay these positions, while we are mostly playing for two results. If Black plays perfectly, it will be a draw – but that applies to chess in general.

analysis diagram

it is still tough for Black, but engine analysis suggests that he should draw 155

156

Introduction

PART II

Sicilian Defence: 1.e4 c5 Introduction

The Sicilian Defence is popular on all levels and is Black’s most respected reply to 1.e4, on par with the classical 1…e5. The non-symmetrical nature of the position often leads to a tense, complicated game. The most principled way to ght the Sicilian is to play the Open Sicilian, basically going 2.♘f3 and 3.d4, almost regardless of Black’s second move. The main practical issue with this approach is: Black has a very wide range of subvariations that you need to prepare for and most of them require very precise move-by-move knowledge. One key idea of the ‘Keep It Simple’ approach is that it should be possible for you to nd good moves over the board in case you have forgotten your concrete lines. In the Open Sicilian however there are many lines that are almost impossible to gure out over 157

the board. It can easily happen that one inaccurate move leads to severe trouble. This is the reason why the Open Sicilian is not featured in this book. Instead I decided to suggest an antiSicilian repertoire, based on ♗b5(+) and a delayed c2-c3 (Alapin), depending on Black’s 2nd move choice. One practical advantage of this repertoire is that it starts with 2.♘f3. That means that once you have mastered the lines presented here you can easily, at a later stage, expand your repertoire with an Open Sicilian against a particular black 2nd move, while still not needing to learn all possible Black systems. For example, you may decide at some point to play 3.d4 against 2…e6. That is an Open Sicilian, but Black cannot play lines like the Najdorf, Dragon or Sveshnikov anymore. Most likely you would get a Taimanov, Kan, or Scheveningen. That’s far more managable than playing the Open Sicilian all the time. This way the suggested repertoire is easily modi able – you may use all of it, or replace a particular line with something else later if you so desire. Let’s have an overview on the presented lines. Chapter 9: 2.♘f3 ♘c6 3.♗b5

158

Introduction

starting position of the Rossolimo Black has tried many di erent rare moves. We’ll have a look at all these minor options in Chapter 9. Chapter 10: 2.♘f3 ♘c6 3.♗b5 g6

This is the Rossolimo Variation, our choice against 2…♘c6. Nowadays this is not a sideline anymore, but possibly White’s main try for an advantage, which avoids entering the Open Sicilian with 3.d4. At the professional level, the main reason for avoiding the Open Sicilian after 2…♘c6 is the Sveshnikov Variation after 3.d4 cxd4 4.♘xd4 ♘f6 5.♘c3 e5. White struggles to nd anything against it, so many Pros have switched to 3.♗b5. For our purposes the Rossolimo is a very logical choice as it is based on very sound principles, including quick development and good pawn structures. The positions are usually more strategic in nature, and it is possible to nd them over the board in case you have forgotten what you looked at before the game. Black has a plethora of options against the Rossolimo. But don’t feel too intimidated, as none of them require extremely detailed knowledge to avoid trouble. You will be OK with commonsense moves and you will frequently get good positions using the ideas presented in this repertoire. In the 159

This is the most popular reply, and examined in Chapter 10. Black prepares the anchetto and tries to increase his control over d4 this way. It also emphasizes kingside development more than the other moves. Note that White will castle on move 4, while Black’s king will be in the centre for a while. This is not a big issue yet, as the position is closed, but further delaying it might be a cause for later concerns. Chapter 11: 2.♘f3 ♘c6 3.♗b5 e6

160

Introduction

Chapter 13: 2.♘f3 d6 3.♗b5+ ♗d7

The second most popular move for Black, only surpassed by 3…g6. Black’s main idea is to go …♘ge7, to allow a recapture on c6 with a knight. The other point is to support a later …d7d5. This line is discussed in Chapter 11. Chapter 12: 2.♘f3 ♘c6 3.♗b5 d6

Black intends to play …♗d7 followed by …a7-a6 to question the bishop. This plan is quite logical, but at the same time rather slow. I think that White has good prospects to get an edge against it. It is important to note that we might also get this position via 2… d6 3 ♗b5+ ♘c6. Please refer to Chapter 12 for this line. 161

3.♗b5+ is called ‘Moscow Variation’. I brie y tried to nd out why this name is used, but came up empty. This line always had a somewhat lame reputation, but this has changed lately. As White got more and more frustrated with battling the Najdorf, this move gained more popularity and is now used regularly by almost all elite players, including World Champion Magnus Carlsen. 3…♗d7 is the most solid reply for Black as it does not clog up his development that much. We will trade on d7 and later set up a Maroczy Bind in most cases. The Moscow with 3… ♗d7 is the subject of Chapter 13. Chapter 14: 2.♘f3 d6 3.♗b5+ ♘d7

162

Introduction

This is the usual choice of ambitious players. They avoid the immediate bishop trade, but Black’s development is delayed. This variation usually leads to complex middlegames with oftentimes all pieces and pawns on the board. Please see Chapter 14 for this combative line. Chapter 15: 2.♘f3 e6 3.c3

is that you can avoid any 3.♗b5 lines with this move order. Our repertoire features ♗b5 against 2…d6 and 2… ♘c6, but against 2…e6 we need a di erent approach. My recommendation is to go 3.c3, in the style of the Alapin Sicilian. Compared to c2-c3 on move 2 Black is committed to an early …e7-e6. This of course is not a huge issue for Black, but it greatly reduces his choice of lines. This makes it rather easy for White to learn this system. It is examined in Chapter 15. Chapter 16: Rare lines on move 2 After 2.♘f3, Black’s most popular moves are 2…♘c6, 2…d6 and 2…e6. Besides these main lines, Black has some rare moves available that we should know good answers to – they can be tricky to handle if you have nothing prepared. Chapter 16 is the ‘odds and ends’ of the Sicilian and shows how to reply to the o beat lines.

Going 2…e6 is Black’s third most popular move, right after 2…d6 and 2…♘c6. In recent years, the lines connected with it, in particular the Taimanov after 3.d4 cxd4 4.♘xd4 ♘c6, have become very popular. The Taimanov is a good system for Black, but an additional factor in favour of it 163

164

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

CHAPTER 9

Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

the Nimzo-Indian, but with White being two tempi up. The alternative 5…d6 is worth checking.

1.e4 c5 2.♘f3 ♘c6 3.♗b5

A) 3…a6 To be honest, I always thought that this move makes little sense, but closer study revealed that it is not as bad as I thought. Against the main lines 3… e6/3…g6/3…d6 White often takes on c6 completely unprovoked on move 4, so why encourage the trade? In addition having the pawn on a6 is rather, a negative compared to the pawn position on a7, as b6 is weakened. As mentioned previously, things are not as simple as you might think – White is better, but needs to be precise. 4.♗xc6 A1) 4…bxc6 5.0-0 d5 This weakens c5 enormously. It looks a lot like a reversed Sämisch Variation of

165

6.e5!. This is a very strong move! White is also better after other sensible moves, like 6.c3, but this is very powerful. 6…d5. Otherwise the position, and especially the e- le, will open even more quickly. After giving up the bishop pair White will usually need compensation in the form of a structural weakness (exploitable doubled pawns) or a lead in development. 7.h3. White has scored 100% in my database from here. A later b2-b3, ♗a3, ♘b1-c3-a4 plan will often just win the c5-pawn. 6.d3 ♗g4 7.h3 ♗h5 8.c4 Of course this is not the only way to a nice edge, but xing the doubled pawns makes a lot of sense. 8…dxe4 Trying to solve the problems with trades does not ultimately help. 9.dxe4 ♕xd1 10.♖xd1 ♗xf3 11.gxf3 e6 12.♘c3 ♘f6 13.♘a4 ♗e7 14.♗e3

166

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

A typical set-up to put pressure on c5. 14…♘d7 15.f4 White is much better here with simple means. Doubling on the d- le and a potential ♖d3-b3 comes to mind. A2) 4…dxc6

I think this is better than 4…bxc6 as it at least opens up the c8-bishop. 5.0-0 A21) 5…♗g4 6.h3 ♗h5 6…♗xf3 7.♕xf3 gives White an easy game. We will play with d2-d3 and a later f2-f4 plan after ♕e2/♕g3. 7.d3 ♕c7

167

This makes the most sense. Black goes for …e7-e5 and prevents a possible g2g4/♘e5 plan. 8.♘bd2 There is nothing wrong with 8.♗e3 either, but quickly going to c4 with the knight seems very sensible to me. 8…e5 9.♘c4 ♗d6 10.a4 White now threatens to play a4-a5, xing the black weaknesses on the queenside. 10…a5 11.g4 ♗g6 12.♘h4 A future ♘f5 and/or f2-f4 will give White a very comfortable game. A22) 5…♕c7 This is the main move and this position is often reached via a di erent move order: 3…♕c7 4.0-0 a6 5.♗xc6 dxc6!?. Black intends to play …e7-e5 and clamp down on the d4-square. 6.a4 I like this direct approach, but there is nothing fundamentally wrong with 6.d3 either. 6…a5

168

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

5.♗xc6 ♕xc6 5…dxc6 6.a4 transposes to 3…a6 (line A). 6.♘c3 White’s advantage in development has reached scary proportions, but the position is not yet open. We need to play a quick d2-d4 now and accelerate the play. 6…e6 7.d4 cxd4 8.♘xd4 ♕c7 9.♖e1 The most standard reply, preventing the clamping move a4-a5. 7.d4 An interesting idea. White trades o one of the weak black pawns, but in return opens up the position, which makes sense as White is ahead in development. Note that the black queenside squares are still weak, and now actually exploitable. In a more closed position it is tough to get anywhere on the queenside. 7…cxd4 8.♕xd4 ♗g4 9.♘bd2 ♖d8 10.♕e5 ♕xe5 11.♘xe5 ♗e6 12.♘ec4 And Black has severe problems with his queenside. White is much better. Note that b2-b3, followed by ♗c1-b2-c3, is a simple way to increase the pressure. B) 3…♕c7 This move is mostly played to allow a recapture on c6 with the queen. There is one substantial drawback, however – Black further delays kingside development and castling. 4.0-0 a6 The most popular and principled move. On other moves we simply go ♖e1/c2c3/d2-d4, opening up the position while ahead in development. 169

This is best here. It prevents the natural move 9…♘f6 due to e4-e5. White has excellent attacking chances here. Note that White has ideas like ♕g4, preparing ♗f4. An additional option is ♗c1-g5-h4-g3 to put more pressure on d6. Black’s position is tough to defend. C) 3…♕b6 This move is a relative of 3…♕c7, but a bit more challenging as we are provoked to play an early ♘c3, which limits our options. Still, Black has problems equalizing, again, due to his sluggish development. 4.♘c3 e6

170

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

5.♗xc6 This is important, and a bit surprising at rst, but you will see how dangerous it is to take now. If we wait until … ♘ge7 is played it will be much more di cult to come up with an aggressive way to open up the position. 5…♕xc6 5…bxc6 is not the most logical looking move, but after seeing the problems following the queen recapture, it is understandable that it has been tried in some games: 6.0-0 d5 7.d3 and a later b2-b3/♘a4/♗a3 plan will be strong. Note how ine ective Black’s c8-bishop is. 6.d4 cxd4 7.♘xd4 ♕c7 8.0-0 a6 This transposes to a line usually arising from 3…♕c7. Going 8…a6 looks very slow, but Black can hardly allow ♘db5 ideas. 9.♖e1 This position is identical to a line starting with 3…♕c7. Black is under serious pressure. D) 3…♘d4 This move is somewhat premature. The knight jump to d4 is quite common in positions where White 171

already has a knight on c3. In these cases, taking on d4 loses time as c3 is immediately attacked. Here White has not committed to anything yet, so he can adapt perfectly to the pawn structure change that Black initiates. 4.♘xd4 cxd4 5.0-0 White has more than one decent plan now. One idea is to go for a d2-d3/f2-f4 plan and play on the kingside. The bishop often drops back to b3 then. The alternative option is to play c2-c3, trying to open the position. 5…g6 6.c3 As mentioned, there is also d2-d3 etc. 6…♗g7 7.cxd4 ♗xd4 8.♕a4 ♕b6 9.♘a3

And White is a lot better. Black’s piece placement is shaky and tough to handle. E) 3…♘a5 A knight on the rim is dim. Well, sometimes this saying is actually true! In the given case this funny knight jump is not completely terrible, but certainly not great. 4.c3

172

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

This probably ts best with our repertoire. Just going for d2-d4 in Open Sicilian style is also ne though (4.0-0 and then d2-d4 without c2-c3 rst). 4…a6 5.♗e2 ♘f6 On other slower moves we will just go d2-d4, of course. 6.e5 ♘d5 7.0-0 e6 8.d4 cxd4 9.♕xd4 White is slightly better here. Note that ♕g4 is a good way to continue and put pressure on Black’s kingside.

6…cxd4 7.cxd4 exd4 8.♘xd4 a6 9.♗c4 0-0

F) 3…e5 10.♘f3! This is an important move. White avoids trades as Black is still somewhat ill-coordinated. Now normal developing moves should conserve White’s edge. G) 3…♘f6

This move has gotten some modest popularity in recent years, due to the original development plan that we will see soon. 4.0-0 ♗d6 This probably would have gotten some bad press 100 years ago, but it is not that bad. The position is quite closed, so Black can a ord some liberties. I still think it does not equalize though. 5.c3 ♘ge7 6.d4 Most straightforward, but you may also go with the d2-d3-based plan here. This is slower and less challenging, but also OK if you prefer a manoeuvring game.

This is a sideline, but it is somewhat underestimated in my opinion. It is not easy to prove anything against it. Usually 4.♘c3 is viewed as critical, but this might get very complicated, which I feel is not warranted for a rather rare line.

173

174

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

4.e5 This will lead to positions that are very similar to the line starting with 3….g6 4.0-0 ♗g7 5.♖e1 ♘f6 6.e5. This was my main reason to go with 4.e5 here. It is easier to play and heads to middlegames that are part of our repertoire anyway. G1) 4…♘g4

Ambitious, but probably not that great. 5.♗xc6 The e5-pawn was exposed, so there is no choice. We gain lots of time though, which gives ample compensation for the loss of the bishop pair. 5…dxc6 5…bxc6 is very slow and should give White a strong initiative. We need to open the position now, as usual after … bxc6: 6.0-0. This is useful anyway. It is not easy to suggest moves now for Black. Extreme care is needed to avoid a quick loss. 6…♕c7 7.♖e1 e6 8.d4. Just one example. White is attacking in the centre and might exploit the weak dark squares in Black’s camp. 6.0-0

175

This is a familiar pawn structure in our repertoire. Here the key issue is the g4knight that will drop back to h6, probably going to f5 next. Now a major di erence to other lines is that White has not yet played the b1-knight to c3. Let’s look at a sample line to illustrate this. 6…g6 No other way to develop the kingside. 7.d3 ♗g7 8.♕e2 ♘h6 9.h3 0-0 10.♗f4 ♘f5 11.c3

And after due prep White will threaten g2-g4. Black is under some pressure, and White’s position is much easier to play. G2) 4…♘d5 5.0-0 ♘c7 A frequently seen alternative is 5…g6 6.♖e1 ♗g7 7.♘c3, transposing to Chapter 10.

176

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

6.♗xc6 dxc6 6…bxc6 is risky as it delays development even further: 7.♖e1 (this is aiming for a transposition again. Black has no good way to avoid it) 7… g6 (what else to develop?!) 8.♘c3 ♗g7 9.♘a4!. This position is usually reached via 3…g6 4.0-0 ♗g7 5.♖e1 ♘f6 6.e5 ♘d5 7.♘c3 ♘c7 8.♗xc6 bxc6?! 9.♘a4!. White is better, please see Chapter 10 for further information on it. 7.h3 Avoiding any …♗g4 ideas. 7…g6 8.♖e1 ♗g7 9.d3 ♘e6 10.♘bd2 Instead, 10.♘c3 is also possible, but this would be a slightly better version of the 3…g6 main line as White has played h2-h3, which he does not need to do in the 3…g6 line. I like the idea to play ♘bd2 here. 10…0-0 11.a4 Allowing a4-a5 is a bit awkward for Black, so he will most likely stop that. 11…a5 12.♘f1 ♕c7 13.♗d2

White’s idea is to play ♕c1 and ♗h6. The f1-knight might be activated with ♘f1-h2-g4. I like White here, but Black is quite solid, of course.

177

178

Conclusion As usual, sidelines are sidelines for a reason. These replies are less reliable than the main branches, but given the rather slow nature of the play Black is not in terrible shape. We will get comfortable, easy-to-play positions with at least a small advantage. In this chapter 3…♘f6 is probably Black’s best try to equalize. The resulting positions however should be familiar to us as we play in a similar manner against 3…g6.

Chapter 10: Rossolimo with 3…g6

CHAPTER 10

Rossolimo with 3…g6 1.e4 c5 2.♘f3 ♘c6 3.♗b5 g6

This move is almost always useful. One point is that it allows ♗f1. In this particular case we will actually take on c6 quite frequently, so this is not as important as in other lines.

4.0-0 What could be more natural? White’s main alternative is to play 4.♗xc6, leading to a strategically very complicated battle. This is my recommendation if you like to expand your repertoire. Note that against 3… e6 we also go 4.0-0, while 4.♗xc6 is the other basic option there, as well. The Rossolimo is a rather versatile line, making it an ideal choice for any repertoire. You can start with one line and later add more variety, without needing to learn something entirely new. 4…♗g7 5.♖e1

A) 5…a6 A rare move, but it is useful to see how it is handled. 6.♗xc6 bxc6 After 6…dxc6 7.d3 Black will probably play either …♘f6 or …e7-e5, both of which lead to slightly worse versions of the main lines. It’s hard to suggest something fundamentally di erent for Black.

7.c3!

179

180

Chapter 10: Rossolimo with 3…g6

It is important to understand that White needs to play c2-c3 and d2-d4 here, exploiting his lead in development. If you play passively, Black might get in …d7-d6/…e7-e5 and …f7-f5 and gain a space advantage. 7…♘f6 8.d4 cxd4 9.cxd4 0-0 10.♘c3 d5 After 10…d6 11.h3 White will develop with ♗f4/♕d2/♖c1. Trading with ♗h6 would be useful, of course. It is important to note that it is hard for Black to get rid of his potentially weak c-pawn. 11.e5 ♘e4 12.♘d2! This is a lot better than taking the pawn on e4, which would have given Black some activity on the light squares as good compensation. 12…♘xc3 13.bxc3 c5 14.♘b3 c4 15.♘c5

The most straightforward way. Instead 6.c3 is good as well. 6…bxc6 On 6…dxc6, 7.e5! is very strong. White gets excellent control over the dark squares and the c8-bishop remains a passive spectator. White’s next moves will be d2-d3/♘bd2-c4 or e4. Note that f6 and d6 are tough for Black to cover.

7.c3! Aiming for a full centre. This is very much comparable to the scenario we examined after 5…a6 6.♗xc6 bxc6. 7…♘e7 8.d4 cxd4 9.cxd4 0-0 10.♘c3 d6 11.♗e3

This is very nice for White. Black’s bishops play no important role, while the knight is quite annoying on c5. B) 5…e6 Mixing …e7-e6 and …g7-g6 does not work all that well here. White gets a solid edge now. 6.♗xc6 181

White is better here, in the long term. Black’s c6-pawn is not going away and 182

Chapter 10: Rossolimo with 3…g6

the black bishops don’t impress. C) 5…e5

The only fully respectable alternative to 5…♘f6. Black clamps down on d4 and intends to develop with …♘ge7 next. 6.♗xc6 This is one of the few cases where I recommend taking on c6, when still relatively unprovoked. Here I believe that White has chances for a small edge, while the alternatives are either toothless or quite complicated, but not leading to an advantage. If you like to explore something entirely di erent, have a look at 6.b4!?. This gives enough compensation and interesting play. As stated above – worth a try as an additional option, but I didn’t like it as a main recommendation.

This recapture leads to a comfortable game for White. 7.c3! We have seen on multiple occasions already that this approach is best after …bxc6. White opens up the centre while being better developed, and ultimately aims to target central weaknesses. Contrary to general wisdom, it is quite often an idea to open the position against the bishop pair – it’s a matter of timing and often very much dependent on concrete factors. 7…♘e7 8.d4 cxd4 9.cxd4 exd4 10.♘xd4 0-0 11.♘c3

C1) 6…bxc6

This sequence is basically forced after 7.c3. Black’s main problem is that the 183

184

Chapter 10: Rossolimo with 3…g6

pawns on d7 and c6 have problems moving to set up a reliable central structure. There will always be some square or pawn weakness that White might play against. 11…♖b8 This is the most often played move, and maybe the one that poses the hardest challenge for White. 1) 11…h6 prevents ♗g5, but it is very slow: 12.♗f4! d6 13.♕d2 g5 14.♗g3 ♖e8 15.♖ad1 looks terrible for Black. d6 is just too weak; 2) 11…d5? 12.exd5 cxd5 13.♗g5 and White is already close to winning. 12.♘b3 White threatens ♗c1-f4-d6 now, with a deadly bind. 12…d5 13.♗f4 This very natural move is a novelty, suggested by Stock sh. 13…♖b4 14.♕d2 Introducing ♘xd5 into the position. 14…a5 15.♗g5 d4 16.♖ad1 This is still complicated, but ne for White. C2) 6…dxc6 This is more reliable than 6…bxc6. 7.d3 C21) 7…♘e7 This looks natural, but I think White’s task is easier here, compared to 7… ♕e7. 8.♗e3 b6

185

9.a3! b2-b4 is the essential plan here. White opens up the queenside to make it easier to attack the weaknesses. 9…a5 Just allowing the push with 9…0-0 10.b4 cxb4 11.axb4 ♕c7 12.♘bd2 f5 13.♘c4 is better for White. He has some weaknesses to attack and is very harmoniously placed. 10.♘bd2 ♕c7 11.♘c4 Note that Black has problems playing …b6-b5, as c5 would be fatally weakened. White’s threat is 12.b4 now. 11…a4 12.b3 And White opens up the queenside favourably. Note how ine ective the ♘e7 and ♗g7 are. C22) 7…♕e7

186

Chapter 10: Rossolimo with 3…g6

This is played to cover e5 and thus prepare …♘f6, which is the rst step for a possible later …♘f6-d7-f8-e6-d4. 8.a3 White has quite a huge choice here. I like the direct 8.a3, but you may also go for 8.h3/8.♘bd2 etc – it’s a slow position. 8…♘f6 8…a5 9.♗e3 ♘f6 10.a4 freezes the queenside. We won’t manage b2-b4 anymore, so we need a new plan. 10… 0-0 11.♘a3 b6 12.♘c4 ♘d7 13.b3. White’s idea is to play f2-f4 in the very long run. Before this is possible, we need to either trade the dark-squared bishops or provoke …f7-f6 by Black. Note that the trade of bishops makes sense, as after a later f2-f4 break the dark squares in Black’s camp will be weak. The bishop on g7 looks weaker than the ♗e3, but Black’s bishop is an important defender.

9.b4 Here we go. White immediately opens up the position. Of course, a slower course of action is also possible. 9…0-0 Instead going 9…cxb4 is a concession. Black cannot grab b4 anyway, so it makes little sense to take now: 10.axb4 0-0 (10…♕xb4? 11.♗a3 ♕b6 12.♘bd2 and White’s attack is very strong) 11.h3 and White is slightly better here. We have some targets on the queenside, in the long run, and Black’s counterplay is less obvious. 10.bxc5 ♘d7 Not the only move, but the structure will be the same after the alternatives. 11.a4 ♘xc5 12.♗a3 I prefer White. His position is very stable and we have some ideas on the queenside. The pin on the ♘c5 is also quite annoying.

187

188

D) 5…♘f6

Chapter 10: Rossolimo with 3…g6

6.e5 Grabbing space with tempo. 6.c3 is a good alternative if you wish to expand your repertoire. 6…♘d5 7.♘c3 ♘c7 7…♘xc3 is too accommodating. White now gets nice pressure along the d- le and Black’s development is lacking: 8.dxc3 0-0 9.♕d5 b6 10.♗f4. Black is under severe pressure. White may also go for a kingside attack with a later ♖ad1/♕d2/♗h6 set-up. The e5-pawn is huge asset that cuts the board in two.

8.♗xc6 D1) 8…bxc6 A rare move that neglects development. 189

9.♘a4 This is a novelty suggested by Stock sh. Usually White plays 9.♘e4 here, and is comfortable, but 9.♘a4 seems even better. 9…♘e6 10.b3 0-0 11.♗b2 d6 If Black does not commit with this move White has many improving moves at his disposal. A set-up with pawns on d3 and c4 comes to mind, as well as ♕d2-e3, ♗b2-a3 to play against c5 and provoke …d7-d6, after all. 12.exd6 This is one major point of ♘a4. Now the simpli cations lead to a slight, but very one-sided, advantage for White. 12…exd6 13.♗xg7 ♘xg7 14.d4 cxd4 15.♕xd4 This structure bears some resemblance with the Scotch Four Knights. Black su ers from central weaknesses. D2) 8…dxc6

9.♘e4 The nice post on e4 is one major motivation behind advancing the epawn. This knight may play a role for coming kingside play or will be used in

190

Chapter 10: Rossolimo with 3…g6

the centre after Black plays …f7-f6, in the future. First of all, c5 is attacked. 9…♘e6 This is the most popular and simplest reply, but it forfeits the idea of developing the bishop to g4. Black also may play 9…b6 (this is more ambitious than 9…♘e6, but also not an easy choice for Black as White now has 10.♘f6+, an irritating move that we will avoid, nevertheless) 10.h3 (as mentioned, there is 10.♘f6+, but after 10…♔f8 11.♘e4 ♗g4 or even 11…h6 Black seems to be OK. After our simple move, 10.h3, the game will very likely look at lot like the main line 9…♘e6, as …♘c7-e6-d4 is on Black’s agenda anyway) 10…0-0 11.d3 ♘e6 12.a4 with very similar play to the main line. 10.d3 0-0 11.♗e3 b6 12.♕d2

White aims for the bishop trade. We will try to make Black’s life uncomfortable on the kingside while allowing very little on the queenside. This position type does not look like a lot, but the score is very good as it is much simpler to handle for White.

191

Black is mostly just sitting there without a clear active plan. 12…♘d4 This is the most solid and straightforward choice. 1) 12…f5 (the e5-pawn is cramping, so Black immediately gets rid of it) 13.exf6 exf6 14.♗h6. White is slightly better here. We are quicker on the e- le and Black has some weaknesses, like the kingside and d6. A very instructive game that reached this position is Rozentalis-Carlsen, Malmö 2004. It is Game 7 among the selected sample games; 2) 12…♖e8 (Black prepares to keep his bishop pair) 13.♗h6 ♗h8 14.♘eg5 ♘d4 15.♘xd4 cxd4 16.♖e2 looks like a better version of the main line. Black is under some pressure as ♕f4-h4 is a simple plan. 13.♘xd4 cxd4 14.♗h6 c5 Black sometimes starts with 14…♕d5 instead, but …c6-c5 will be played soon anyway, so it’s not a big di erence. 15.♕f4 ♗b7

16.h4

192

Chapter 11: Rossolimo with 3…e6

The main point of White’s play is that the pawn on e5 allows this kingside initiative, and if Black wants to solve this completely he will very likely have to play …f7-f6 at some point. This has to be well prepared, though, as White might double on the e- le more quickly. This does not amount to all that much, but White’s position is easier to play here. This is also supported by the scores that favour White quite heavily. Note that Black has no active ideas at all, and this means White’s position is very safe. Conclusion Our repertoire choice against 3…g6 limits Black’s activity to a minimum. If Black follows the main path up to 16.h4 at the end of the chapter, it is a rather one-sided a air where Black might be equal with best play, but this is not easy to prove in practice – see the statistics. Many earlier deviations are downright bad for Black. An important concept is to play in the centre with c2c3 and d2-d4 if Black recaptures on c6 with the b-pawn. In those positions White’s lead in development may easily reach decisive proportions.

193

CHAPTER 11

Rossolimo with 3…e6 1.e4 c5 2.♘f3 ♘c6 3.♗b5 e6

4.0-0 Here White’s main alternative is 4.♗xc6, creating an imbalance in the position immediately. White gives up the pair of bishops, but creates weaknesses in Black’s camp. I prefer 4.0-0 for our repertoire, because it is easier to learn and understand. If you like to expand your repertoire at a later stage, I advise you to have a closer look at 4.♗xc6. 4…♘ge7 This is consistent with 3…e6. Black prepares …a7-a6 and/or …d7-d5. 1) Sometimes Black goes for 4…♘f6. This is rare and not particularly good: 5.e5 (there is nothing wrong with other sensible moves like 5.♖e1 or 5.♗xc6 either) 5…♘d5 6.♘c3 ♘xc3 7.dxc3. It is important to take in this way, as it gives White an even greater edge in development. A quick ♗f4 will 194

Chapter 11: Rossolimo with 3…e6

make it hard for Black to get the ♗c8 into play; 2) 4…a6 5.♗xc6. OK, if Black insists we will take it. This now transposes to 3…a6 (Chapter 9). 5.♖e1

7…a6 8.♗a4 b5 9.♗b3 ♘c6 10.♘d2 ♗b7 11.f4 ♕c7 12.♘f3

Here I prefer White slightly, but Black is not in bad shape either.f4-f5 is a key for White later. White has to commit regarding his bishop on b5. He may play 5.c3, signalling his intention to drop it to b3/c2 later. This has some similarities to the Ruy Lopez. My choice is 5.♖e1 in order to drop it back to f1 and then expand in the centre. Both moves are good and of roughly equal value theoretically. A) 5…♘d4 A rare move, but not bad. Usually Black goes for this when White has committed a knight to c3, so a capture on d4 loses further time. As there is no knight on c3, we should be able to get a small advantage here. 6.♘xd4 cxd4 7.d3 White will place the bishop on b3 and then try to start central/kingside play with f2-f4, as well as transfer a knight to f3. A sample line for this strategy goes: 195

B) 5…♘g6 Black prepares …♗e7 and …0-0. 6.c3

Our goal is to get a full centre, so this makes a lot of sense. B1) 6…d5 7.exd5 ♕xd5 A mix of the Rossolimo and Alapin Sicilians. It looks a bit odd to me, but it’s not so bad for Black. 8.d4 196

Chapter 11: Rossolimo with 3…e6

This is logical, trying to use the better development. 8…cxd4 9.♘xd4 ♗d7

10.♘xc6 was a serious threat. 10.♗d3 I suggest to play this new move, leading to sharp play in the centre. The idea is to play the bishop to e4. 10…♘xd4 11.♗e4 ♕d6 12.cxd4 ♗c6 13.♘c3 We are now threatening to push d4-d5; the black king is still in the middle of the board. 13…♗e7

A really cool engine suggestion that keeps the initiative. 14…♘xh4 This is the critical choice for Black. The bishop could not take the h-pawn due to d4-d5 and 14…♖d8 15.h5 ♘h4 16.♕g4 is also very dangerous for Black. 15.d5 exd5 16.♘xd5 ♘g6 17.♗e3 With a strong initiative for White. I could continue this line further, but you have to stop at some point. Black may equalize with precise play, but learning moves up to move 30 is too much. B2) 6…a6 7.♗f1 ♗e7 8.d4 cxd4 9.cxd4 d5 10.e5 We have reached a position more typical of the French Defence. I think White is ne here as Black still has his ♗c8 problem and a later …f7-f6 gives White a ready-made target on e6. 10…0-0 11.♘c3 f6 12.exf6 ♗xf6 13.♗e3

And the structure gives White a slight edge. This move is played to block the e- le. 14.h4!

197

C) 5…a6 6.♗f1

198

Chapter 11: Rossolimo with 3…e6

Preserving the bishop. 6 ♗xc6 is actually quite OK as well. After 6… ♘xc6 7 d4 we would, however, get a position more in the spirit of the Open Sicilian, which I want to avoid in this repertoire. 6…d5 1) 6…♘g6 7.c3 leads to the line starting with 5…♘g6, examined earlier; 2) 6…g6?!.

analysis diagram

The moves …e7-e6 and …g7-g6 usually don’t mix well if White is able to open the centre quickly. This is one of those cases. White needs to act quickly to exploit the weak d6-square: 7.d4! cxd4 8.♘xd4. This is a bad version of an Open Sicilian. Let’s have a look: 8…♗g7 9.♘xc6 ♘xc6 10.♘c3 0-0 11.♗f4. This is the problem. Black is stuck with the bishop on c8 and with his central weaknesses. 7.exd5

199

7…♘xd5 1) 7…♕xd5 is better than it looks. Black tries to keep control over d4, thus trying to prevent a further opening of the position: 8.♘a3! (important move – White eyes the b6 weakness) 8…♕d8 (this move is hard to avoid; ♘c4 would force it on the next move anyway) 9.♘c4 ♘f5 10.a4 b6 11.c3 and White has ♕b3 in store with some pressure; 2) 7…exd5? is pretty awful, but I am still surprised that it was not played a single time in my database: 8.d4 (Black is in deep trouble with his king in the centre and his development lagging) 8…cxd4 9.♘xd4 ♘xd4 10.♕xd4 and Black will probably lose the d5-pawn in the long run. 8.d4

200

Chapter 11: Rossolimo with 3…e6

White tries to open the centre as he is better developed and already castled. C1) 8…cxd4 This natural move is less precise. 9.♘xd4 ♗e7 Black accepts a pawn structure defect, but he needs to get castled. After 9… ♘xd4 10.♕xd4 Black has no good follow up to prepare kingside castling. Normal developing moves give White a good game here. 10.♘xc6 bxc6 11.♘d2

The knight heads towards c4. White is slightly better here due to the pawn structure. C2) 8…♗e7 201

White gets a comfortable edge now, as Black is not well prepared for the opening of the position. 9.dxc5 0-0 Following 9…♗xc5 10.c4 ♘f6 11.♕xd8+ ♘xd8 12.♘c3 White is in very good shape. Note that ♘a4/♗e3/ ♗f4 ideas are coming, to exploit the weak dark squares. There is also the simple a2-a3/b2-b4 plan to put more pressure on Black. 10.a3 ♗xc5 11.b4 ♗e7 12.c4 ♘f6 13.♕c2

White is much better already. Space, better central control and the much better bishops add up. ♗f4/c4-c5 will also create a potential outpost on d6. 202

Chapter 11: Rossolimo with 3…e6

C3) 8…♘f6!

This is the best move. I think it is quite remarkable that Black can a ord to further delay his development, but concrete analysis shows that Black is very solid here. 9.♗e3 cxd4 With 9…♗e7 Black delays the capture on d4. 10.dxc5 (if White goes 10.c4 instead, it will often transpose to the main line, however there is the irritating move 10…♘g4 that I want to avoid by taking on c5 now) 10…♕xd1 11.♖xd1 ♘g4 12.♗f4 ♗xc5 13.♗g3 h5!.

This is necessary. Black needs to initiate concrete play, otherwise h2-h3 203

will send the knight back and, with simple moves like c2-c4, ♘c3, a2-a3, b2-b4 etc., White will be better. 14.♖d2! (we need to cover f2 to defuse …h5-h4) 14…h4 15.♗c7. Instead 15.♘xh4 ♘xh2! leads nowhere. White now intends to play ♘c3-e4 if possible. It’s a complicated game with chances for both sides. 10.♘xd4 ♗d7 10…♘xd4 will often just transpose to the main line: 11.♗xd4 ♗e7 12.c4. This is the transposition to 10…♗d7. Instead, 12.a4 is also a ne idea, trying to keep c4 free for the knight at a later stage. 11.c4 ♘xd4 12.♗xd4 ♗c6 13.♘c3 ♗e7 14.a3 White tries to set his majority in motion. 14…a5 15.♕d3 0-0 16.♖ad1

White enjoys a small edge with his well-centralized pieces and chances to exploit the queenside majority. This position was reached in sample game 8, Carlsen-Gelfand, London Candidates 2013. It is well worth studying the

204

Chapter 12: Rossolimo with 3…d6

mentioned Carlsen game to get a better picture of this type of position. Conclusion 3…e6 is a solid response to the Rossolimo Variation. We quickly play 0-0 and ♖e1 for swift development and push in the centre later. If Black knows the main line and plays precisely, he will be very close to equal, but if anybody, White is asking all the questions. The sample game CarlsenGelfand is an excellent example for this type of play. One mistake by Black may lead into trouble, while White has easy moves to continue. One reason is his majority on the queenside, which is very useful to have.

CHAPTER 12

Rossolimo with 3…d6 1.e4 c5 2.♘f3 ♘c6 3.♗b5 d6

4.0-0 ♗d7 This is by far the most popular and best move. Others delay the development even more and are quite risky: 1) 4…♗g4 is quite ambitious, trying to set up a counter-pin on our knight: 5.c3 (d4-d5 would be nice to get in, so White prepares the central expansion) 5…♕b6 (continuing with ambitious play. Of course, Black may go with a slow move, but White has an easy game with moves like d2-d4, ♖e1 etc. Black is still some moves away from castling) 6.♗a4 ♘f6 7.d4 0-0-0 (this looks crazy, but it has been played by the strong GM Tiviakov. White is a lot better though) 8.♗xc6 ♕xc6 9.d5 ♕a6 10.♘bd2 and White will continue with c3-c4, maybe b2-b3 and ♗b2. It’s just a Benoni where Black’s king and queen are strangely placed. White might also go for a b2-b4 based plan instead; 205

206

Chapter 12: Rossolimo with 3…d6

2) 4…a6 5.♗xc6+ bxc6 6.c3 transposes to 3…a6 (Chapter 9). 5.♖e1 ♘f6 Black also plays 5…a6 here. Usually he goes 5…♘f6 rst, then …a7-a6. This early …a7-a6 will transpose most of the time: 6.♗f1 ♗g4 (6…♘f6 7.c3 also transposes) 7.c3 ♘f6 8.h3 and again we have transposed to the main line below. 6.c3 a6 7.♗f1

A) 7…e5 The only serious alternative to 7…♗g4, staking a claim in the centre. 8.h3

White spends a tempo avoiding the pin after d2-d4. You may play 8.d4 immediately, but after 8…cxd4 9.cxd4 207

♗g4 you need to deal with the c6knight coming to d4, in many cases. It is something I like to avoid, but it is not objectively bad for White. 8…g5? This has quite some shock value, but I am still amazed that GM Simon Williams has played it four times(!). If White reacts well it is almost unplayable. Black’s better option is to play 8…♗e7, for example: 9.d4 0-0 10.d5 ♘a7 11.a4 b5 12.axb5 axb5 13.b4 and White is slightly better here at least. It looks a lot like an old-fashioned line of the Chigorin Ruy that I used to play as Black. Note that White has an idea of ♖e2-a2 there that is not available in the Ruy. 9.d4 I think this is easier to handle than taking on g5. 9…g4 10.hxg4 ♗xg4 11.d5 ♘e7 12.♗g5 ♘g6 13.♘bd2

The general problem with Black’s position is that he has a bunch of longterm weaknesses and there is no adequate counterplay. 208

Chapter 12: Rossolimo with 3…d6

13…h6 14.♗xf6 ♕xf6 15.♕a4+ ♗d7 16.♕b3 b5 17.a4 c4 18.♕b4 a5 19.axb5 And White was winning in the game Burg-Williams, Amsterdam 2014. B) 7…♗g4

By far the most common move. In many cases Black will take on f3, giving us the bishop pair. In return, Black gains some time to develop, as we will see in the main line. If Black tries to keep the bishop by retreating to h5, White has excellent chances to get an early attack. 8.h3 B1) 8…♗h5?! Retreating is very risky, but White needs to be very energetic now.

209

9.g4! ♗g6 10.d4! The important point! Taking on e4 is bad now: 1) 10…♘xe4? 11.d5 ♘b8 12.♖xe4 ♗xe4 13.♕a4+; 2) 10…♗xe4? 11.♖xe4 ♘xe4 12.d5 ♘b8 13.♕a4+ ♘d7 14.♕xe4; 3) 10…e5 (not ideal, but relatively best compared to the alternatives) 11.d5 ♘b8 12.♘h4. White is better here due to his space advantage and excellent knight post on f5 to come. 10…cxd4 11.cxd4 d5 Taking the pawn fails, just like it does with c-pawns on the board. 12.exd5 Instead, 12.e5 and 13.e6! is strong as well, shutting in the f8-bishop. I think 12.exd5 is just as strong, and more straightforward. Black will have a very hard time surviving the next moves. 12…♘xd5 13.♘c3 e6 14.♘xd5 ♕xd5 15.♗g2

210

Chapter 12: Rossolimo with 3…d6

And White has a great attack, based on the ♘e5 idea and a potential d4-d5 after the queen leaves its post on d5. B2) 8…♗xf3 This is best. Black keeps control over d4 for a while and White will lose some time by moving the queen. 9.♕xf3

And we have reached the mentioned French type position. Black is OK here, but we still have the bishop pair for a long-term advantage. Note also that … f7-f6 is tougher to play without the light-squared bishop – there are downsides to trading this often problematic piece for Black. 10.d3 ♗g7 11.♗e3 Trying to play d3-d4 to build the ideal centre. 11…♘d7 Black has good control over d4, but we need to be persistent. 12.♘d2 0-0 13.♕d1 b5 14.♘f3 b4 15.♕a4

9…g6 Fundamentally di erent is 9…e6. Here Black prepares …d6-d5, which often leads to French-type positions. White needs to regroup now, and develop the queenside: 10.d3 ♗e7 11.♘d2 0-0 12.♕d1 (the plan is to go ♘f3 and d3-d4) 12…d5 13.e5 ♘d7 14.♘f3

And a later d3-d4 gives White the better game. The position is opening up

211

212

Chapter 13: Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7

and we already have the bishop pair. Please note that the immediate 15.d4 is also perfectly ne and covered in sample game 9, Bok-Tiviakov, Groningen 2014. Conclusion 3…d6 is fairly popular, but it does not equalize. In the main line, Black gives up the bishop pair and his compensation is not clear to me. Black has no pawn weaknesses and is solid that way, but it seems rather unattractive to me from the black point of view. For White I recommend to study the position type at the end of the main line, starting with the sample game provided. This will occur frequently, but there are many possible earlier deviations that are quite bad for Black – make sure to study these, quick wins will be the reward!

213

CHAPTER 13

Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7 1.e4 c5 2.♘f3 d6 3.♗b5+ ♘d7

4.c3 This is my suggestion at this point, but White has some interesting alternatives. The most principled move is 4.d4, but this was too close to an Open Sicilian for this repertoire. There is absolutely nothing wrong with it, but it just does not t well with our purposes. Also, there is 4.0-0, going for 4…a6 5.♗d3!?. This is quite interesting as well but a tad too arti cial for a ‘Keep It Simple’ repertoire. Again, nothing wrong with it, but it didn’t work out for me in this context. My chosen move opens up the c2-square for the bishop and leads to position types that are more typical of the Ruy Lopez than anything else. One important point to remember is that Black intended to play a Dragon or Najdorf, most likely, and might not be 214

Chapter 13: Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7

well versed in defending a Ruy Lopez structure. 4…♘gf6 There also is 4…a6. This has little independent value compared to 4… ♘f6 unless White tries to omit ♕e2, which makes little sense: 5.♗a4 b5 6.♗c2 ♗b7 7.0-0 ♘gf6 8.♕e2 and we will transpose to the main line. Instead, 8.♖e1 is possible, but not particularly better than the queen move. 5.♕e2

Now, and on the next couple of moves, Black has many options. He may play … b7-b5/…e7-e5 for the typical Ruy Lopez set-up (most popular and our main line), but there is also …e7-e6 or some combination of …♕c7 and …g7g6. Fortunately, White’s responses are all very natural – d2-d4, 0-0 etc. It is easy to play. It is impossible to cover all the move orders, so I will just give some sample lines. 5…a6 1) 5…e5 is badly timed. Black should get in …a7-a6/…b7-b5 rst and only later play …e7-e5: 6.d4 ♕c7 7.0-0 and here White has the option to retreat to 215

d3 and play a2-a4. Black does not manage to grab any queenside space, compared to the early …a7-a6/b7-b5 lines; 2) 5…e6 6.d4 a6 7.♗d3

White has an easy game with good central control. One key point in the later phase of the game is to answer the question if the push e4-e5 is helping White or not. Black will get the d5square often and might use the long diagonal with the bishop coming to b7. On the other hand, White gets space, the post e4 for a piece, and possible kingside play. 6.♗a4 b5 Natural, but clearly there are some alternatives, given the exible position type: 1) 6…e5 7.d4 b5 8.♗c2 leads to positions typical of the main line starting with 6…b5; 2) 6…e6 7.0-0 b5 8.♗c2 ♗e7 9.d4.

216

Chapter 13: Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7

as both 10.b3 and 10.d3 give White a comfortable game.

This is very similar to 5…e6 earlier. If/when to play e4-e5 is a key point. White however has lots of useful moves to play (♖e1/d1, ♘bd2-f1-g3 etc) and may wait for a good moment. He has more space after all. This type of position is analysed in sample game 10, Maze-Plenca, Germany Bundesliga B 2016/17; 3) 6…♕c7 is quite exible. Black may still play …e7-e5/e6 or …g7-g6 and transpose to one of those structures. There is an interesting and independent idea, though: 7.0-0 c4. This is the independent idea, but I think it may be a bit too ambitious. Instead 7…e5 was played by Topalov against Carlsen in 2012, leading to more standard positions examined in the main line. 8.♗c2 (this move is best. Now White is ready to attack the pawn on c4 with b2-b3 and a later a2-a4 against …b7-b5 pushes by Black; a2-a4 rst and then ♘a3 is also a good idea, as we will see) 8…e5 9.a4 ♕c6. This has been played most often here, but Black is in trouble now. Stock sh suggests 9…♘b8 with just half a pawn plus for White – nothing to be afraid of 217

10.♘a3 b5 11.axb5 axb5 12.d3 ♗b7 13.♗g5 and White will win a pawn on the queenside for no compensation; 4) 6…g6 is also quite respectable for Black. We have some easy moves, however, to complete our set-up: 7.0-0 ♗g7 8.d4 b5 9.♗c2 cxd4 10.cxd4 ♘b6 11.♗g5 with ♘bd2 and ♖f1-c1 to follow. This is not the only set-up, of course, but a logical one. 7.♗c2 ♗b7 8.d4

8…e5 This stops any White ambitions to push e4-e5 and leads to a very Ruy Lopez like position. 218

Chapter 13: Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7

1) 8…e6 9.0-0 we have already examined via 6…e6. Quite clearly there are many transpositions possible; 2) 8…cxd4 9.cxd4 ♖c8 is a very direct approach. Black immediately threatens e4 as the c1-bishop is unprotected: 10.0-0.

White should avoid 10.♘c3 b4 11.♘d1 ♖xc2, which is a worse version of the sacri ce. 10…♖xc2 (a radical solution. If Black does not play this move he will come under some pressure on the queenside with a later ♗d3 and a2-a4) 11.♕xc2 ♗xe4 12.♕e2 ♗b7.

13.♗d2!. Important, in order to get a2a4 in without allowing …b5-b4. Black has some compensation for the 219

exchange, but White can’t complain either. 9.0-0 ♗e7 10.♖d1 ♕c7 11.d5 Not strictly necessary yet, but grabbing space makes sense. 11…c4 We should also check 11…0-0 12.a4. White plays against the queenside now. Instead, 12.c4 is also quite good, trying to close this side of the board and then switch to kingside play: 12…c4 13.♘a3 ♕a5 14.♗e3.

Black is under heavy pressure on the queenside. A well-prepared b2-b3 lever will give White access to the b5-pawn, which is tough to defend long-term. 12.b3 12.a4 is less promising here due to 12…♘c5 and a later …♕d7. 12…♘b6 After 12…cxb3 13.axb3 0-0 14.♗d2 ♘b6 15.♘a3 White will open up the queenside with a later c3-c4, giving him a very nice initiative. 13.♘fd2 Not the most natural move ever, but it is strong. White increases the pressure

220

Chapter 14: Moscow Variation with 3…♗d7

on c4 and, compared to the more natural ♘bd2, keeps c3 covered. 13…cxb3 13…0-0 14.bxc4 bxc4 15.♘a3 is very unpleasant for Black. c4 is weak for no compensation. 14.axb3 0-0 15.♘f1

White is better here as well. Playing c3c4 will be the long-term plan. Conclusion Our choice against the combative 3… ♘d7 leads to positions that are rather ‘non-Sicilian’ in nature. The play is rather slow, with careful development by both sides. One point of playing this line with white is that this slow play is not what the black player originally intended when going for the Sicilian. He had a Najdorf or Dragon prepared, now he needs to deal with something that rather looks like a Ruy Lopez. If you study the plans and middlegame ideas, starting with the sample game, you will be the favourite against the typical opponent who faces 3.♗b5+ rather infrequently.

221

CHAPTER 14

Moscow Variation with 3…♗d7 1.e4 c5 2.♘f3 d6 3.♗b5+ ♗d7

4.♗xd7+ This is the normal move, and the main line by a mile. Only after I had completed work on this repertoire did I learn that 4.c4 is currently in fashion. It shares some similarities with our main line and is a good choice for a later addition to your repertoire. A) 4…♘xd7 5.0-0 ♘gf6 6.♕e2

222

Chapter 14: Moscow Variation with 3…♗d7

6…e6 1) 6…g6 7.c3 ♗g7 8.d4. White is quite comfortable here. Note that e5-e6 is an immediate idea as Black is weak on the light squares. 8…e5 is quite accommodating, but it is useful to know how to react to it as similar situations might occur from other lines. 9.dxe5! makes a lot of sense here, as with this trade White quickly exploits the weak d5-square, which is not defended by a light-squared bishop anymore. This trade would be far less promising with, for example, White having a bishop on g2 and Black on c8. 9…dxe5 10.♖d1 ♕c7 11.♗g5 (trading o a defender of the weak square d5) 11…0-0 12.♗xf6 ♗xf6 13.a4 and White has two knights to exploit the weak light squares, in particular d5. The black bishop on f6 is pretty useless; 2) A recent trend is the exible move 6…♖c8.

leading to a French-like structure without the c8-bishop. We rather want the Maroczy Bind instead. This can be started with c2-c4 and d2-d4 later, or d2-d4 immediately. Against 7.c4 there is the irritating move 7…♘e5, so we start with the d-pawn. 7…cxd4 8.♘xd4 Black will play …♗e7/…0-0/…♖c8/… a7-a6 on the next moves, with the move order not being particularly important. I will go with just one and not point out the maze of di erent move sequences. 8…♗e7 9.c4 0-0 10.♘c3 a6 11.b3 Without the light-squared bishop, the queen is well placed on the e2-square, and the anchetto of the c1-bishop becomes more attractive. Without the bishop trade White’s bishop will be on e2 and his queen on d2, where it covers c3 as well. In this situation, e3 is the most natural post for the bishop. Here b2 is ne, as it covers c3, and there are also some attacking chances against the king using the long diagonal. 11…♖c8 12.♗b2 ♖e8 13.♖ad1

This line is examined in the analysis of sample game 11, Xu-Goldbeck, Tegernsee 2017. 7.d4 Here a c2-c3 based approach makes less sense as Black will manage …d6-d5,

White may go for f2-f4, with a possible rook lift to g3. I prefer White’s chances

223

224

Chapter 14: Moscow Variation with 3…♗d7

here, but of course Black is quite solid. B) 4…♕xd7 This move signals that Black will play …♘c6 most of the time to emphasize his control over d4. This is most often combined with a kingside anchetto. 5.c4

A strategically very sound choice. White places his pawns on light squares after this bishop has been traded o . We are aiming for a Maroczy Bind structure, giving White comfortable control. 5…♘f6 This is the most popular move on the top level nowadays. There are some concrete lines where delaying …♘c6 is relevant, as an early …♖f8-c8 will attack the c4-pawn. More about that later. 1) With 5…e5 Black just stops d2-d4 and sets up a symmetrical structure in the centre. At rst it looks like Black should be worse here, because of the potentially bad bishop on f8, but in fact Black’s position is rather solid. White only has a small edge: 6.♘c3 ♘c6 7.d3 g6. 225

Black aims for a later …f7-f5, so this ts well. 8.a3 (we gain some space on the queenside. With the centre completely closed we need to play on the anks) 8…♗g7 9.♖b1 ♘ge7 10.b4 0-0 11.0-0 f5 (11…b6 12.♘d5 ♘xd5 13.cxd5 ♘d4 14.♘xd4 cxd4 15.♗d2 is very slightly better for White here, due to the better bishop. For an illustration of White’s prospects please look up the classic game Spassky-Byrne, Candidates Match 1974) 12.♗g5!? (intending simply to capture on e7, thus improving his light-square control. Black should prevent that scenario) 12…♗f6 13.♗xf6 (this starts concrete play, aimed at the somewhat loose pawn structure) 13…♖xf6 14.bxc5 dxc5 15.♘d5 ♖f7 16.a4 b6 17.a5!.

226

Chapter 14: Moscow Variation with 3…♗d7

And White is better as Black cannot take with the knight; 2) 5…♕g4 ‘Greed is good!’ Gordon Gekko says. Well, just based on the evidence provided by the analysis of this move we have to conclude he was wrong: 6.0-0 ♕xe4 7.d4. This opens up the position entirely. Black will face a tough defensive task, so we don’t have to check every single imaginable line from here. Just enjoy playing a true gambit-style position over the board! Let’s have a look at a nice line to illustrate the play: 7…cxd4 8.♘xd4 ♘f6 9.♖e1 ♕g4 10.♕a4+ ♕d7

11.♘b5! (the point of 10.♕a4+; 12.♘c7+ is the threat now) 11…♘c6 227

12.♘1c3 e6 13.♘d5 with a great attack for White; 3) 5…♘c6 is a very popular move, but I choose 5…♘f6 as the main move order to look at. Black will always play …♘f6 shortly after 5…♘c6, with a transposition. Where else than f6 could the knight possibly go? 6.♘c3 g6 (6… ♘f6 is examined under 5…♘f6; 6…e5 we have checked via 5…e5) 7.d4 cxd4 8.♘xd4 ♗g7 9.♗e3 ♘f6 10.f3 and we have reached the main-line position examined via 5…♘f6 rst. 6.♘c3

Black now has three basic approaches. He needs to develop the f8-bishop soon, so the rst two approaches are to play 6…e6 or 6…g6 now or after inserting 6…♘c6 7.d4 cxd4.8 ♘xd4. The third approach is to try some concrete play after 6…♘c6, based on the queen sortie …♕d7-g4!? to attack the still unprotected g-pawn. We’ll start with this approach. B1) 6…♘c6 In most cases Black will play …e7-e6 or …g7-g6 right after 7.d4 cxd4 8.♘xd4. 228

Chapter 14: Moscow Variation with 3…♗d7

These positions are examined via the 6….g6 and 6….e6 move orders, so the only unique lines starting with 6….♘c6 feature early concrete play by Black. 7.d4 White may also play 7.0-0 rst, but this is, in fact, not just a later transposition to a line examined via 6…g6. After 7.00 g6 8.d4 cxd4 9.♘xd4 ♗g7 Black is threatening ♘xe4!, and 10.♗e3 ♘g4 is awkward. White is stuck with 10.♘de2, which is not terrible for him, but there is little incentive to allow this if the immediate 7.d4 is a simple transposition. The lines allowed by 7.d4 involving … ♕g4 are not great for Black, so we don’t mind these additional options for our opponents. B11) 7…♕g4!?

against it is a bit more complicated than I would have liked, for such a rarity, but I didn’t nd a simpler solution. 8.d5 The only challenging approach, forcing the action. 8…♘d4 9.♘xd4 ♕xd1+ 10.♔xd1 cxd4 11.♘b5 ♖c8 11…♘xe4? 12.♖e1 wins material for White. 12.b3 This position White can force after the 7…♕g4 sortie. We are slightly better as, after the likely trade of the e-pawn for the d-pawn, we will have far more space, and Black has some issues with the f8-bishop. 12…♘xe4 There is no problem after 12…a6 13.♘xd4 ♘xe4 14.♔e2. We will go ♗e3, ♖ac1, ♖fd1 and play on the queenside, with a nice edge.

The computer era has brought many original ideas to light – this is one of them. White is slightly better though – let’s have a look. Please note that this is a rather obscure line that is played very rarely, but the comp sees it as an option to consider. My recommendation

13.♖e1! ♘xf2+ Again, 13…a6 14.♖xe4 axb5 15.cxb5 does not worry us. White has the better ending here. We will get d4 back and keep the more active pieces/king position.

229

230

Chapter 14: Moscow Variation with 3…♗d7

14.♔d2 Yes, a somewhat weird position with a two-pawn de cit. But note that d6 is hanging with check.a7 is threatened as well, and the f2-knight only has g4 as a possible escape route. 14…♖d8 15.h3 Cutting o the knight. 15…g6 16.♗b2 ♗g7 17.♗xd4 ♗xd4 18.♘xd4 f5 Securing escape for the adventurer on f2! 19.♔e3 ♘e4 20.g4 White keeps the initiative here. A really long line to study, but I didn’t nd anything simpler that was convincing. B12) 7…cxd4 8.♘xd4 ♕g4 Far more popular and better are 8….e6 or 8…g6. These positions we examine via the 6…e6 and 6…g6 move orders. The queen sortie is the only interesting move that does not transpose. We need to be precise here to keep an edge. 9.♕xg4 ♘xg4 10.♘xc6 bxc6

11.♗f4! This sequence is the key to remember after 8…♕g4. White intends either to play e4-e5, to trade the e-pawn for the 231

d-pawn, worsening Black’s structure, or to take the knight coming to e5, again playing for the better pawn formation. 1) 11…♖b8 is a more conservative approach compared to 11….g5: 12.h3 ♘f6 (12…♘e5 13.♗xe5 dxe5 14.0-0-0 and Black will su er forever with his bad structure) 13.0-0-0 ♘d7 14.♖he1 g6 15.e5 dxe5

16.♖xe5! A cute idea to remember! White is better with the more active pieces and the better pawn formation; 2) 11…g5 is actually not that good, but it is helpful to have seen it before the game: 12.♗xg5 ♘e5 13.b3 ♖g8 (this was Black’s idea) 14.♗h4 ♗g7 (14…♖xg2? 15.♗g3 and the rook is trapped) 15.♔d2 and Black does not have enough for the pawn. B2) 6…e6 Black chooses to develop the bishop to e7, setting up a pawn formation with both …d7-d6 and …e7-e6 being played, very typical of many Open Sicilian lines, like the Scheveningen Variation. 7.d4 cxd4 8.♘xd4

232

Chapter 14: Moscow Variation with 3…♗d7

Very natural play. Black will lose some time with the queen, leaving the d- le. 12…♕c7 13.f4 ♖ac8 14.b3

Note that …e7-e6 and ♗e7 look more passive than …g7-g6 and ♗g7, but there is a very important upside: Black controls d5, so the white knight is not coming to d5. In …g7-g6 formations this is always an idea, as Black cannot easily kick it away with a later …e7-e6 as this weakens d6 considerably. 8…♘c6 With 8…♗e7 Black may try to omit … ♘c6 altogether and later play …♕c7, to develop the knight to d7, starting with this move. However one wonders why 4…♕xd7 was played in the rst place if this was the desired set-up. White keeps a nice edge with standard moves: 9.0-0 0-0 10.♕e2 (this initiates our standard set-up against the Scheveningen centre) 10…a6 11.b3 ♕c7 12.♗b2 ♘bd7 13.♖ad1 ♖ac8 14.f4 and White has an excellent centre and good attacking chances on the kingside. 9.0-0 ♗e7 10.♗e3 10.b3 is also quite good, but, with the knight on c6, it is useful to cover d4 with ♗e3, so that ♕e2 and ♖ad1 can be played next. 10…0-0 11.♕e2 a6 12.♖ad1 233

I like White here, with a good centre and chances on the kingside. B3) 6…g6 7.d4 cxd4 8.♘xd4

B31) 8…♗g7 Black may choose this move order to launch quick play against the c4-pawn, delaying the knight development to c6 for a while. 9.f3! After 9.0-0?! ♘c6 Black is OK, as already examined via 8…♘c6 9.0-0?! ♗g7, which leads to the same position. 9…0-0 234

Chapter 14: Moscow Variation with 3…♗d7

It is advisable to check 9…♕c7.

more detail via the 8…♘c6 move order below. B32) 8…♘c6

This is rarely played, but it was used in a well-known game by Vishy Anand against Magnus Carlsen in 2012. This can be found further in this book as sample game 12. 10.♗e3 Again, it is important to get the move order right. 10.0-0?! is completely defanged by the genius 10…♖c8 11.b3 d5!!, a concept that was introduced by Vasily Ivanchuk. I don’t want to get into the details here, so please check for yourself if you are interested. 10…♖c8 Or 10…♘c6 11.0-0, a position we check via the 8…♘c6 move order (B32 below). 11.b3 a6 12.♕d3 Black was threatening to go …b7-b5, so we protect the c3-knight to stop this idea. 12…♘c6 Black does not have any better option now. 13.0-0 And we have reached the standard Maroczy centre that we will examine in

9.f3! This is an important move order detail! We want to play ♗e3 and f2-f3 to support the centre anyway, but it is important to start with this immediately and not waste a tempo on castling. After 9.0-0?! ♗g7 Black threatens 10… ♘xe4! and it is too late to play 10.f3. Instead, we need to go 10.♘de2, which is less testing for Black than the f2-f3 set-ups. 9…♗g7 9…♘xd4 10.♕xd4 ♗g7 11.♗e3 0-0 12.♕d3 leads to a standard Maroczy position that is slightly better for White. We will try to prevent black pawn breaks like …b7-b5 and maintain a space advantage. ♘d5 will be a key idea for changing the pawn structure, as after …♘xd5 White retakes with a pawn: cxd5 will often lead to a better endgame, exd5 might be an option to play against the e7-pawn. 10.♗e3

235

236

Chapter 14: Moscow Variation with 3…♗d7

We need to lend some protection to the knight and develop the bishop to the most natural square. 10…0-0 11.0-0

It is quite di cult to analyse these positions on a move-by-move basis. Black has di erent rook set-ups (… ♖fc8, …♖ac8 + …♖ad8) and moves like …a7-a6, …♕d8-a5 etc., that may be played in various move orders. White is quite happy to consolidate his space advantage and prevent pawn breaks like …b7-b5 or be ready for … e7-e6 and …d6-d5 by Black. We will centralize and often base our play on the knight coming to d5. The Maroczy centre is generally regarded as slightly favourable for White, but it requires patience, building up slowly. It is very useful to study some complete games in order to get a better feel for the position. It is important to note that the play is quite strategical and slow in nature, a type of position that, in many cases, the black player did not intend to reach when he learned his main-line Open Sicilian theory. Defending the black side of the 237

Maroczy is quite di erent to a Dragon or Najdorf main line. We will check some possible set-ups now. B321) 11…♖ac8 12.b3 The pawn on c4 needs some cover. 12…♖fd8 This signals that Black is trying to play in the centre with …e7-e6 and …d6-d5 to come. 13.♕d2 e6 14.♖ad1 After Black has revealed his plan, we know how to place this rook. White needs more control of the d- le as we will see in the coming sequence of moves. 14…d5 15.cxd5 exd5 16.♘xc6 bxc6 17.e5 ♘e8 18.f4

Here it is clear why the rook is needed on d1 and why ♖ad1 was more appropriate than ♖fd1, in this speci c situation. White is better here, with the superior pawn structure and more active pieces. B322) 11…e6

238

Chapter 14: Moscow Variation with 3…♗d7

Going for the …d6-d5 break immediately. 12.♕d2 ♖fd8 13.♖ad1 Against …e7-e6 this set-up with ♖ad1 is best. We will see why, shortly. 13…d5 14.cxd5 exd5 15.♘xc6 bxc6 16.e5 ♘e8 17.f4 And it’s very similar to 11…♖ac8 above. B323) 11…♖fc8 This move is usually a sign that Black aims for queenside play. Often, his next moves will be …♕d8-a5 and …a7-a6, supporting …b7-b5. 12.b3 ♕d8 13.♕d2 ♕a5 14.♖fd1 This is the most popular, but it is equally good to play 14.♖ac1. Again, it is not so move-order-critical as long as you address the breaks that Black may have – here we need to keep an eye on …b7-b5, of course. 14…a6 15.a4 Stopping the b-pawn by radical means. Weakening b4 is not ideal, but keeping a grip on the position is more important. 15…♘d7

239

16.♘de2!? This is an interesting idea. White avoids further trades and tries to play ♖ab1 and ♘d5 soon. If Black covers d5 with …e7-e6, the d6-pawn is a potential target. Black is quite solid here, but I still think White has the easier game. B324) 11…a6

Against this move we need to be precise. Black has the simple idea to take on d4 and then go …b7-b5 right away. 12.♕d3! This is not a very common square for the queen, but here it is most important to keep an eye on b5, thus 240

Chapter 14: Moscow Variation with 3…♗d7

preventing the b-pawn push. Note that 12.♕e2? blunders to 12….♘xe4!. 12…♘e5 12…♖ac8 is threatening 13…♘e5, picking up the c-pawn. Now 13.b3 ♘e5 14.♕d2 looks like a loss of time, but in fact Black has not gained that much. … b7-b5 is impossible and the knight is actually quite unstable on e5. White is slightly better with normal moves to come, like ♖ac1 and a later ♘d5 in most cases. 13.♕e2 ♖ac8 14.b3 b5 Instead, 14…e6 sees Black taking control of d5 and preventing ♘d5. 15.♖ac1 (it is important to cover c3, so …b7-b5 is again impossible) 15…♕c7 16.♖fd1 ♕a5 17.♕d2. We have prevented …b7-b5 and are better now, due to the weak pawn on d6 and the space advantage typical of this line.

15…♘xd5 16.cxd5 ♕b7 16…e6 is probably best, but not the kind of move a human would consider immediately over the board: 17.dxe6! (in fact, 17.f4?, the move that probably stops you from calculating 16…e6, is not that great due to 17…exd5 18 fxe5 dxe5 and Black has good play and some pawns for the knight) 17…fxe6 18.♖ad1 and White enjoys a small but stable edge with the better pawn structure in the centre. 17.♖ac1

And White is better here, f3-f4 and ♘c6 being the main idea. Black is under pressure.

It looks like Black has reached his goal, but White still retains a slight pull. 15.♘d5 This is available, and it saves the day. The knight now threatens the fork on b6, and b5 is hanging, so Black has to take it. 241

Conclusion The Maroczy Bind is the most frequently seen pawn structure in this chapter. The Maroczy centre always had the reputation to be at least slightly favourable for White. His space and excellent control accounts for a lot. The main challenge is to not lose track of the move orders at the beginning. The correct timing of f2-f3/0-0/♗e3 is important to put the most pressure on Black. In the main-line middlegames in 242

Chapter 15: 2.♘f3 e6 3.c3

the Maroczy, it is crucial to keep an eye on the pawn breaks …b7-b5 and …d6d5 and look for a favourable moment to go ♘d5.

CHAPTER 15

2.♘f3 e6 3.c3 1.e4 c5 2.♘f3 e6

3.c3 The Delayed Alapin is our line against 3…e6. Other interesting choices here (besides the Open Sicilian, of course) are 3.b3 or 3.g3. These are worth considering if you want to expand the repertoire. I didn’t choose them as my main lines because I felt they did not t into the general concept of this repertoire as well as 3.c3 does. With Black committed to …e7-e6 he will usually play …d7-d5 on move 3 or 4. The only other very serious alternative is 3…♘f6, attacking the pawn and unbalancing the pawn structure, as White needs to advance the e-pawn. A) 3…b6 This is an independent option that is comfortable for White. Black does not ght for the centre and White develops easily. 243

244

Chapter 15: 2.♘f3 e6 3.c3

4.d4 ♗b7 5.♗d3 ♘f6 6.e5 And this position we examine via a completely di erent move order, starting with 1…b6, the Owen Defence. I was not sure where to place it (Owen or c3 Sicilian), but at the end decided it probably ts more to the 1…b6 move order – so please see this chapter for more information. B) 3…a6 4.d4 d5 5.exd5 And it transposes to 3…d5. C) 3…♘c6

A natural developing move that will very often transpose to lines examined via 3…d5. 4.d4 cxd4 This is the only independent option, and regarded as imprecise, as White should bene t from the trade. The c3square is now available to White’s knight. It is best for Black to play 4…d5 5.exd5 and we will transpose to 3…d5 after either recapture. 5.cxd4

245

C1) 5…♘f6 6.♘c3 Instead, 6.e5 transposes to 3…♘f6 4.e5 ♘d5 5.d4, which is not part of our repertoire. Our move avoids this transposition and is objectively stronger as well. 6…d5 7.e5 ♘e4 8.♗d3 ♘xc3 8…♗b4 9.♕c2 is also ne for White. 9.bxc3 ♗e7 10.h4

And White has a promising position on the kingside. Black must avoid castling now; after 10…0-0 the classic sacri ce 11.♗xh7+ wins for White. C2) 5…d5

246

Chapter 15: 2.♘f3 e6 3.c3

c3: 7…♗d7 8.a3 ♘ge7 9.♘a4 ♕c7 10.♗d3 and White enjoys a nice space advantage. 7.♘c3 ♘f5 8.a3 This is useful in order to have ♘a4 available after a possible …♕b6. Without a2-a3 Black may respond with …♕a5+/…♗b4+ ideas. 8…♗d7 9.h4

6.e5 This is the theoretical argument against Black’s move order. We get to an Advance French with an early trade on d4 that bene ts White, as he is able to play ♘c3 early on. However, it needs to be said that it is not a situation where White is clearly better and just needs to learn some moves to prove it. In fact, I am unsure if White has anything better than in a regular French if Black is precise. If you feel uncomfortable in this structure, after 6.e5, you also have the option to simply take on d5, which is playing for a tempo edge in a symmetrical position, or go 6.♘c3, which very likely gets to an IQP after Black takes on e4. I don’t want to examine all this in detail as this move order, for Black, is quite rarely seen. If you want to keep it very simple, 6.exd5 is the way to play. 6…♘ge7 This seems to be best choice for Black, immediately aiming for f5 with the knight. After 6…♕b6, 7.♘c3 is possible now compared to the French Advance where there are still pawns on c5 and 247

The intended follow up is g2-g4, as now the annoying response ♘h4 is blocked. This position is slightly better for White, but you see that it is not at all that obvious. You have various options at move 6, so there are other position types available if you like to avoid this French Advance structure. D) 3…d5 This is the ‘French’ way of playing. I suggest to take on d5, avoiding a possible transposition into a French Advance Variation after the alternative 4.e5. 4.exd5 Now we will quite often reach an IQP position. After 4…exd5 Black often goes for the IQP, while after 4…♕xd5

248

Chapter 15: 2.♘f3 e6 3.c3

we will play with the isolated pawn in many cases. D1) 4…exd5 5.d4

White is very comfortable here, being better developed and coordinated. 12…0-0 13.dxc5 ♗xc5 14.♖ad1 ♖ad8 15.♗g5 f6 16.♗e3

Now Black has a wide range of moves to check.

And White managed to get a good antiIQP game.

D11) 5…♘f6 6.♗b5+ White does not mind trading this bishop against its c8 counterpart, or pinning the knight on c6 if Black avoids the trade. Playing against the IQP by trading minor pieces is a ne approach. 6…♗d7 6…♘c6 we check via the 5…♘c6 6.♗b5 move order (line D14 below). 7.♗xd7+

D112) 7…♘bxd7 8.0-0 ♗e7 Black quickly closes the e- le before it is too late. 9.dxc5 ♘xc5 10.♗e3 0-0 11.♘bd2

D111) 7…♕xd7 This is somewhat questionable as now there is no knight on d7 that is able to recapture on c5 comfortably. 8.0-0 ♘c6 9.♖e1+ ♘e4 9…♗e7?! 10.dxc5 0-0 11.♗g5 is very bad for Black already, being a pawn down or allowing ♗xf6, destroying the pawn structure. 10.♗f4 ♗e7 11.♘bd2 ♘xd2 12.♕xd2

249

This position is roughly equal, but easy to play for White. You go ♕c2, centralize rooks and maybe go for ♗d4 and ♘d2-f1-e3, to put pressure on the IQP. Black has got active pieces, of

250

Chapter 15: 2.♘f3 e6 3.c3

course, so we cannot count on any substantial edge. D12) 5…c4 This is not well-timed at this particular moment. 6.b3 cxb3 This is forced. After 6…b5? 7.a4 Black will lose the queenside. 7.axb3 ♗d6 8.♗d3 ♘e7 9.0-0 0-0 10.♘a3

White is also better after other moves, but this is a nice move, intending ♘b5 or just ♘a3-c2-e3 to move the pieces to the kingside. White is better in the centre and has nice chances after the natural… 10…♘bc6 11.♘h4!? This controls f5 and prepares f2-f4, to attack. D13) 5…a6 This move avoids an early ♗b5+ and is often quite useful later on. This position also arises frequently after 2… a6, the O’Kelly Variation. 6.♗d3 A natural move, but other normal developing moves would be ne as 251

well, for instance 6.♗e2 or 6.♗e3. D131) 6…♘c6 7.0-0 cxd4 8.♖e1+ ♗e7 9.♘xd4 ♘xd4 9…♘f6 10.♗e3 0-0 11.h3 leads to a quite typical position. White is very slightly better. 10.cxd4 White has an annoying pull here. 10…♘f6 11.♘c3 0-0 12.♗g5 ♗e6 13.♗c2

Intending ♕d3 and ♗b3 ideas. Black is doomed to passive defence. D132) 6…c4 The most ambitious move, which is quite in line with the early …a7-a6. 7.♗c2 ♗d6 8.b3 The normal reaction to …c5-c4. 8…cxb3 8…b5 9.a4 is ne for White. He opens up the queenside while being better developed. 9.axb3 ♘e7 10.0-0 0-0 11.♗a3

252

Chapter 15: 2.♘f3 e6 3.c3

Black is close to being equal now, but we still have the slightly better structure and easy moves like ♖e1 will follow. D14) 5…♘c6

This is the most popular move, placing the knight on the most logical square, looking at the crucial d4 spot in front of the future isolated queen’s pawn (IQP). 6.♗b5 ♗d6 Black needs to get his kingside developed in order to get castled quickly. With …♗d6 he usually intends quickly to play …♘ge7 to avoid the issues that we discuss following 6…♘f6. 253

Let’s check 6…♘f6. This is the move that Black ideally wants to play. The knight has its most active post on f6, but it allows a pin by ♗g5 and does not block the e- le, like …♘ge7 in the main line does: 7.0-0 ♗e7 8.dxc5 ♗xc5 9.♗g5 ♗e6 10.♘bd2 0-0 11.♘b3 ♗b6 12.♘bd4 and White retains a comfortable edge here as the IQP is safely blockaded and the pin on the f6knight is quite unpleasant for Black to handle. 7.dxc5 ♗xc5 8.0-0 ♘ge7 8…♘f6 9.♗g5 we examine via the 6… ♘f6 move order in the paragraph above. 9.♘bd2 0-0 10.♘b3 Now we are at the nal important point where Black needs to choose where to place the bishop.

D141) 10…♗d6 11.♗d3 This retreat is in fact the most active move. It prevents 11…♗g4 due to 12.♗xh7+, and prepares the battery with ♗c2/♕d3 to provoke weaknesses. 11…h6 In order to play …♗g4 Black avoids ♗xh7+. 254

Chapter 15: 2.♘f3 e6 3.c3

12.h3 No pin please – this is an important move. 12…♘f5 A good move that prevents 13.♗e3. Instead with 12…♗c7 Black can aim to set up a battery himself: 13.♖e1 ♕d6 14.♗e3 b6 15.♘bd4 a6 16.♗c2 and White is in excellent shape. He is very well coordinated and has no weaknesses, while Black has multiple problems with the pawn structure. 13.♗c2 ♖e8

14.♕d3 This is quite inconvenient for Black to meet. White has also tried the more technical move 14.♖e1. This was features in sample game 13 between c3 specialist Sergey Tiviakov and GM Ferenc Berkes, played in Plovdiv 2003. 14…g6 It may be better for Black to play 14… ♕f6. The best reply is 15.♗d2. Grabbing the pawn on d5 does not lose immediately but gives Black strong compensation. We should rather focus on development. Note that 15.g4 ♘h4 leads to absolutely nothing as we only 255

have a harmless check on h7. The battery, here, is a constant worry for Black, but we need to be cautious how we actually execute the ‘threat’, especially if we have to weaken our position like 15.g4 does here. After 15…♗e6 16.♖fe1 g6 17.♖e2 I prefer White but Black is not in terrible shape either. His pieces are active and the weaknesses are not an immediate problem. 15.♗d2 ♗e6 15…♕f6 is better here again, transposing to 14…♕f6. 16.♘bd4 This is strong, inducing multiple trades on d4. 16…♘cxd4 17.♘xd4 ♘xd4 18.♕xd4

An excellent scenario for White. Without knights the position becomes very static and the IQP weakness becomes more pronounced. D142) 10…♗b6

256

Chapter 15: 2.♘f3 e6 3.c3

This move and 10…♗d6 are of roughly equal value. 10…♗b6 looks more principled, though, as it still connects with the d4-square. In any case, White would like to trade this bishop – it follows the general rule to trade minor pieces vs the IQP and gets rid of Black’s better bishop. 11.♖e1 To play ♗e3 is a key idea to trade the bishop, which makes 11.♖e1 a very natural choice. 11…♘f5 Directed against 12.♗e3. I think this is better than 11…♗g4 12.h3 ♗h5 13.♗e3.

Here, White is comfortable again, being able to trade the bishops on e3 or push Black’s bishop back to c7. 12.♗d3 Against …♘f5 set-ups the battery with ♗c2/♕d3 is usually a good idea. 12…h6 13.♗c2 ♗e6 14.♕d3 ♖e8 Note that after 14…♕f6, 15.♗e3 is an important point. The ♘f5 is basically pinned, so ♗e3 is possible now. (Note that 15.g4 is answered by 15…♕g6! and White gets nothing out of this weakening pawn push.) 15…♗xe3 16.♖xe3 g6 17.♖e2 is no fun for Black. He is quite passive and has no dynamic chances to compensate for the IQP weakness. 15.♗d2 ♖c8 16.♘fd4 This is an idea if Black avoids …g7-g6. 16…♘cxd4 17.♘xd4 ♗xd4 18.cxd4 ♕f6 19.♗c3 This is slightly better for White due to the bishops, but Black is quite solid, obviously. D2) 4…♕xd5

Black avoids the IQP with this move and in fact White could be the one with a (favourable) IQP soon. Most of the 257

258

Chapter 15: 2.♘f3 e6 3.c3

time however we play with an early ♘a3, intending to harrass the queen with ♘b5, leading to an open centre without an isolated pawn. 5.d4 D21) 5…cxd4 This is a quite clear concession, giving White the good c3-square for the knight. White will easily get a comfortable IQP position. 6.cxd4 The IQP structure has been discussed a lot in chess literature or media, so I only want to give a quick summary of what is most important. The side with the IQP enjoys uent piece development and does not have any problem pieces. With many pieces on the board the weakness of the isolated pawn is not that pronounced, as it is di cult to attack it multiple times. This is not the case after minor pieces are traded, when the rooks and queen may focus on attacking the pawn. Following this logic, the IQP side should keep minor pieces on, and, if there are no active opportunities for the rooks, allow them to be traded. The square in front of the pawn is very important as it may serve as an outpost for the opponent. If the IQP side manages to push the pawn forward and break this potential blockade it is often favourable as piece activity persists, while the weakness gets removed. This is just a brief intro to the IQP, please see some classical examples or books on strategy for more information. 6…♘f6 7.♘c3 ♕d8 259

Probably the wisest decision. Other squares o er additional tempo-gaining possibilities for White. 8.♗d3 ♗e7 9.0-0 0-0 10.♖e1 ♘c6 11.a3

This is a rather favourable IQP position, but it is useful for your general chess knowledge to study these structures to understand them better. I will give a typical move sequence to illustrate the dangers for Black, but further study of classical games is very bene cial here. 11…b6 12.♗c2 ♗b7 13.♕d3 ♖c8?

This normal-looking move already puts Black in grave danger. 14.d5! exd5 Instead, 14…♘a5 is the most resilient, but still very bad for Black. 260

Chapter 15: 2.♘f3 e6 3.c3

Now it’s just over immediately. 15.♗g5 g6 16.♖xe7 ♕xe7 17.♘xd5 And Black loses decisive material. D22) 5…a6 This is quite rare in this particular position, but the position sometimes arises from 2…a6 3.c3, which is also part of our repertoire. 6.♗d3 Other developing moves are also ne, of course. 6…♘f6 7.0-0 ♗e7 8.dxc5 This capture is a bit more promising for White here compared to other, similar positions due to the weakness on b6. White should be slightly better with other sensible moves as well. 8…♕xc5 9.♗f4 0-0 10.♘bd2

White is slightly better here. He is well developed and Black still has no great option for his bishop on c8. White’s next moves will likely be ♕e2, a2-a4, ♘c4, to increase the pressure on the queenside. D23) 5…♘c6 This has little independent value as … ♘f6 will be the next move most of the 261

time. 6.♘a3 Now 6…♕d8 7.♘c2 ♘f6 or 6…♘f6 7.♗e3 will transpose to the main line D24 below. The two rare alternatives are bad for Black: 1) 6…a6 addresses ♘b5, but the cure is worse than the disease: 7.♘c4 (b6 is a glaring weakness) 7…♕d8 8.♗e3 (White’s developmental advantage is reaching alarming proportions) 8…b5 9.♘ce5 ♘xe5 10.♘xe5 cxd4 11.♗xd4 ♘f6 12.a4 b4 13.a5.

And Black will probably only last for a few more moves, facing ♕a4+ and ♗e2-f3; 2) 6…cxd4 is ill-timed: 7.♘b5 ♗d6 (the only move already; 7…♕d8 8.♗f4 and it’s ‘game over’) 8.♗c4 ♕e4+ 9.♗e3 ♗b8 10.♘bxd4 and White is a lot better with excellent mobilization. D24) 5…♘f6 This is the main line by a mile. This follows the logic that this knight will be best on f6 with almost 100% certainty, while the optimal position of the other knight is not as obvious and 262

Chapter 15: 2.♘f3 e6 3.c3

bishop moves make little sense at the moment. 6.♘a3

D241) 6…♘c6 With this move Black keeps the queen on d5 as long as possible. 7.♗e3 Very natural play, but it is important to note that instead 7.♘b5 ♕d8 8.dxc5 is less promising than it looks due to 8… ♗xc5 9.♕xd8 ♔xd8, and this endgame does not lead to much. It is, of course, an option if you like a quiet game, but I think 7.♗e3 is objectively more promising. 7…cxd4 8.♘b5 ♕d8 9.♘bxd4 ♘d5 9…♘xd4 is too accommodating. White retains a good game with simple moves: 10.♗xd4 ♗e7 11.♗d3 0-0 12.♕c2.

White is better both with 0-0 or 0-0-0 next. Choose according to your taste – for a nice example of 0-0-0 see the game So-Akobian, St Louis 2016. 10.♘xc6 bxc6 11.♗d2 11.♗d4 is also interesting, but plans involving …f7-f6 are a bit murky. Our move is simple, but also good for a slight pull. 11…♕b6 Black may also develop the f8-bishop, of course, but this move is the only one that requires a concrete reaction. After a move like 11…♗c5 12.♗d3 0-0 13.0-0 White enjoys a nice structural edge. 12.♗d3! We happily o er the b2-pawn as we are much better developed. 12…♕xb2 13.0-0 With excellent attacking chances for White. D242) 6…♕d8 7.♘c2

263

264

Chapter 15: 2.♘f3 e6 3.c3

pieces and a small advantage. Simple moves, like ♕c2/♗e5, and centralization of the rooks will follow. 9.0-0

The knight on c2 will recapture on d4 if Black takes this pawn. Here both 7.♘c4 and 7.♗f4 are very decent alternatives if you like to vary your play. I went for 7.♘c2 as the next moves are simple to play and yet still good for a comfortable game. We have the general development plan to go for: ♗d3, 0-0, ♕e2, ♗f4/g5, and then centralize the rooks. This will work pretty much regardless what Black does. 7…♘c6 With 7…♘bd7 Black opts for a set-up with the knight on d7. He could have started this set-up with 7…♗e7 and 8…0-0 as well, of course. Our set-up is the standard one mentioned earlier. After 8.♗d3 ♗e7 9.0-0 0-0 10.♕e2 b6 11.♗f4 ♗b7 12.♖ad1 White is slightly better. We will play ♖fe1 and ♘e5 next, and see how Black reacts. Our good centralized pieces give us some options, from piece-based kingside play to playing on the queenside with ideas like ♘f3-e5-c4-d6. 8.♗d3 ♗e7 Starting to trade with 8…cxd4 9.♘cxd4 ♘xd4 10.♘xd4 ♗e7 11.0-0 00 12.♗f4 gives White the more active 265

Very natural, but White does not always need to commit his king early on. I’d also like to refer you to sample game 14, Godena-Ter Sahakyan, Plovdiv 2013. In this game White delayed castling in a very similar position and managed to reach a promising position. 9…0-0 10.♕e2 b6 This is seen most often, but Black has tried some other moves. I don’t think we need to examine them all, as White will go ♗g5 and ♖ad1 anyway and be a bit better. 11.♗g5 As mentioned before, this standard setup is always possible. Note that 11.dxc5 is also ne, and should be slightly better for White as well. 11…♗b7 12.♖ad1 cxd4 13.♘cxd4 ♘xd4 14.♘xd4 ♕d5 This is Black’s most active set-up. 15.f4 With an interesting game. White has some attacking chances on the 266

Chapter 15: 2.♘f3 e6 3.c3

kingside, but Black is quite solid. I have selected three sample games for the Delayed Alapin and only after annotating them all I recognized that all three have been played in Plovdiv, Bulgaria – a weird coincidence. Maybe we should rename this line to ‘Plovdiv Variation’? E) 3…♘f6 This is best in my opinion. Of course, 3…d5 is very respectable as well but I feel that White has more chances for an edge against this move. After 3… ♘f6 there is no path to a white advantage, according to modern theory. 4.e5 ♘d5

5.g3 White’s big main line is 5.d4, of course, but it is rather complicated and Black gets to choose the course of play at important junctions. My choice, 5.g3, is a rather modest move but it is easy to play and learn. If you want to sharpen up your repertoire, 5.d4 is the way to go.

Black quickly tries to put pressure on the e5-pawn. 6.♗g2 ♕c7 7.0-0!

O ering the e5-pawn for excellent compensation. Black should probably decline, but we need to check the capture, of course. E11) 7…♘xe5?! This is extremely risky, giving White a great initiative. 8.♘xe5 ♕xe5 9.♖e1 ♕d6 10.♘a3 Instead, 10.d4 is ne as well. Just open it up as White is so much ahead in development. 10…♗e7 11.♕b3 ♘f6 Essentially forced as ♗xd5/♘b5 was in the air. 12.d4

E1) 5…♘c6 267

268

Chapter 15: 2.♘f3 e6 3.c3

And Black still has some questions to answer. E2) 5…d6 6.exd6 ♗xd6 7.♗g2 ♘c6 8.0-0 0-0 9.d4

With excellent compensation for the pawn. E12) 7…d6 Not very principled, but probably better than taking. We now get a somewhat di erent version of …d7-d6 lines as the black queen is not necessarily well placed on c7. 8.exd6 ♗xd6 9.d4 0-0 10.♘a3 Highlighting the issue with the early … ♕c7. 10…a6 11.♘c4 ♗e7 12.dxc5 ♗xc5 13.♕e2

White’s idea in this line is to use the better bishop position on g2 – better, that is, than its counterpart on c8. We will also try to make use of the queenside pawn majority, which is often nicely supported by the g2bishop. 9…cxd4 Black may keep the tension with 9… ♗d7 or something like 9…h6. Against this, capturing on c5 is a simple and good choice: 10.dxc5 ♗xc5 11.♕e2, intending to play ♖d1 and expand on the queenside. 10.♘xd4 ♘xd4 11.♕xd4 White recaptures with pieces on d4 in order to avoid the IQP. After 11.cxd4?! Black easily develops with …♗c8-d7c6 and blockades the pawn. 11…♕c7

Black is OK here, but White still has some ideas to try, for example: 13…b5 14.♘e3 ♘ce7 15.♘xd5 ♘xd5 16.a4 269

270

Chapter 15: 2.♘f3 e6 3.c3

An interesting try. Of course, you may also just retreat with 15.♕e3, leading to equal play. 15…♗xf4 No better is 15…♗xd4 16.♗xc7 ♗xc3 17.bxc3 ♖ac8 18.♗d6 ♖fe8 19.♗xb7 ♖xc4 20.♗b4 when White has an edge with the bishop pair. 16.♕xf4 ♕xf4 17.gxf4

12.c4 This is a rare move, but I like it better than the more popular 12.♘d2, which is, of course, not bad either. I simply feel that I would rather have the knight on c3 in this position. The sample game 15, Spasov-Szostko, Plovdiv 2015, provides a good illustration of 12.♘d2 and this position type in general. 12…♘f6 13.♘c3 ♗d7 13…e5?! is questionable as it weakens d5 permanently: 14.♕d3 a6 15.♗g5 and Black is already in serious trouble. 14.♖d1

This is equal, but there is still some play left. The doubled pawns are not exploitable while White has the more active pieces and some pressure. E22) 14…♗c5 15.♕d3 ♗c6 16.♕e2 ♖fd8 17.♗f4

Black is at a crossroads: E21) 14…♗e5 15.♗f4!?

271

272

Chapter 16: 2.♘f3: other lines

This is an equal position. White will try to exploit his majority on the queenside in the endgame. Conclusion The Delayed Alapin is a sound choice versus 2…e6. Black is limited to a smaller range of defences compared to the Alapin on move 2. Black should be able to equalize, or very close to that, after 3…d5 4.exd5 ♕xd5 and 3…♘f6. These positions however are not an immediate draw or anything very forced. They just lead to middlegames with equal chances where you still have ideas to outplay opponents. White is extremely solid and can play a riskfree game with only a low theoretical workload needed.

CHAPTER 16

2.♘f3: other lines 1.e4 c5 2.♘f3

This chapter deals with the rare and o beat lines that Black may try. A) 2…a6

This is the O’Kelly Variation. The main reason for Black to choose this line is that it contains a little trap: 3.d4?! is met by 3…cxd4 4.♘xd4 ♘f6 5.♘c3 e5! and after 6.♘b3 ♗b4 Black is in very

273

274

Chapter 16: 2.♘f3: other lines

good shape. Not a big deal for us, as we don’t go for Open Sicilians anyway. Even though 2…a6 is not very popular, it is quite playable, so we cannot expect to refute it, but we can get a comfortable game for sure. 3.c3 This ts well with our repertoire. 3.c4 leads to Maroczy Bind/Hedgehog positions, while 3.♘c3 will most likely later transpose to an Open Sicilian. Our Alapin-style approach is also a very respectable choice. Compared to 2.c3 we now have ♘f3 and …a7-a6 inserted, which means it must be at least slightly more favourable for White. ♘f3 is useful all the time, while …a7-a6 delays Black’s development and weakens the queenside, in particular the square b6. We will have a closer look at 3…d5 now. Black may try moves like 3…d6 or 3…♘f6, but you will be ne by just playing normal moves that develop and control the centre – there is no need to learn moves there by heart for such a rare line. 3…d5 This is the most popular move, even though it opens the position while remaining rather less well developed, compared to 2.c3 d5. Probably, 3…e6 is the most solid here. It transposes to positions that we will get via 2.♘f3 e6 3.c3 most of the time: 4.d4 d5 5.exd5. Instead, 5.e5 is an Advance French with Black having played …a7-a6 early on. This is actually not a bad line, so there is no huge incentive to go there. After the capture

we will transpose to 2.♘f3 e6 3.c3 d5 territory. 4.exd5 ♕xd5 5.d4 The only independent lines now are the ones where Black develops with …♗g4. If he goes …e7-e6 instead, we will transpose to the lines we check via 2.♘f3 e6 3.c3 d5 4.exd5 ♕xd5. 5…♗g4 6.♗e2 e6 7.h3 It is useful to insert this as h2-h3 is a little improvement for White in all cases. It also has a concrete plan attached to it, as we will see later. 7…♗h5 8.0-0 ♘f6 9.♗e3 cxd4 If Black tries to delay this we might take on c5 and keep the pawn. c3-c4 and ♘c3 is also an idea. 10.cxd4 ♘c6 11.♘c3 ♕d6

275

276

12.g4 White is also OK after a more standard approach, but this is strong. 12…♗g6 13.♘e5 The point. The pawn on e5 cannot be taken and White prepares moves like ♗f3 and/or ♗f4 to increase the pressure. 13…♘xe5? This fails completely, but we need to check why. If Black avoids this he is just

Chapter 16: 2.♘f3: other lines

slightly worse. 14.dxe5

A2) 14…♕xe5 15.♕a4+ ♘d7

A1) 14…♕xd1 15.♖fxd1 ♘d7 16.♖ac1 This is almost winning for White. The threat is 17.♘b5!. 16…♘xe5?

16.♗f4! This is very important. Against 16.♖ad1 or 16.♖fd1 the reply 16… ♗d6 still continues the game. 16…♕c5 17.♖fd1 ♖d8 18.♖ac1 White’s attack is overwhelming. B) 2…♘f6

This fails in spectacular fashion. 17.♘b5 The coming sequence is forced. 17…axb5 18.♗xb5+ ♔e7 18…♘c6 19.♖xc6 bxc6 20.♗xc6+ and mate to come! 20…♔e7 21.♖d7+ ♔f6 22.g5+ ♔f5 23.♖d4 ♖xa2 24.♗e4+ ♔e5 25.f4#. 19.♗g5+ f6 20.♖c7+ ♘d7 21.♖dxd7+ ♔e8 22.♖d1#

This rare move introduces the Nimzowitsch Variation. It is occasionally played by stronger players to surprise the opponent. Black does not equalize if White plays the most principled line. 3.e5 ♘d5 4.♘c3 e6

277

278

Chapter 16: 2.♘f3: other lines

This is the traditional main line of the Nimzowitsch. The bad news is that now we have to memorize some moves. The good news is that it leads to a very good position for White – so there is little choice. Instead of 4…e6 the modern way to play it is 4…♘xc3 5.dxc3 (of course, we take this way to speed up development) 5…♘c6 6.♗f4 ♕b6 7.b3. I think that is all one needs to know. Next moves are ♕d2/h2-h4, and White has an edge in the centre. It is not terrible for Black but White is certainly to be preferred.

5.♘xd5 exd5 6.d4 ♘c6 Black o ers a pawn with this move. It is not correct, ultimately, but the alternatives are not great either. After 6…d6 7.♗b5+ ♘c6 8.0-0 Black will be in trouble on the e- le. We are just much better developed. 7.dxc5 ♗xc5 8.♕xd5 ♕b6

279

That is the point of Black’s play. 9.♗c4! ♗xf2+ 10.♔e2 It looks a bit wild now, but move-bymove analysis shows that White is better. The king is quite safe on e2, and Black has problems on f7. 10…0-0 11.♖f1 ♗c5 12.♘g5 ♘xe5 A bit of a shocker, but it does not fully convince. Black has also tried 12… ♘d4+ 13.♔d1 ♘e6 14.♘e4 d6 15.exd6 ♖d8 16.♗d3 ♗xd6 17.♕h5, but White is much better here, with threats on the kingside. 13.♘xf7! It is important to avoid 13.♕xe5? d5!. 13…♘xf7 14.♖xf7 ♕e6+ 15.♕xe6 dxe6 16.♖xf8+ ♔xf8 17.♗g5

280

Chapter 16: 2.♘f3: other lines

This long line is the reason why Black’s pawn sac is unpopular. Black is eternally worse here – not by a mile, but it’s no fun at all, obviously. I would have liked to present something shorter than a 17-move sequence against the Nimzowitsch, but I don’t see anything else convincing. C) 2…b6

This is a rather uncommon variation, but it is slightly annoying for our repertoire. The reason is that the way to punish Black is to play 3.d4, heading for an open Sicilian where …b7-b6 is a slow set-up. Still, I am not going to recommend that way to play as you will probably never face 2…b6 anyway. If you know beforehand that your opponent will play this line, you might want to prepare 3.d4 in detail for such a game. The critical line goes 3.d4 cxd4 4.♘xd4 ♗b7 5.♘c3 a6 6.♗g5 ♕c7 7.♘d5, which is very tough for Black, but something you will want to study in more detail – it is quite tricky. If your opponent surprises you with 2…b6 I recommend a simple approach: 3.♗c4 281

The general plan is the very natural 00, ♖e1, c2-c3, d2-d4. Note that 3.c3 does not work all that well, here, as after 3…♗b7 there is no great way to react to the threat against e4. 3…♗b7 4.0-0 e6 4…♗xe4? 5.♗xf7+ ♔xf7 6.♘g5+ ♔e8 7.♘xe4 is excellent for White. 5.♖e1

And we’ll just play normal moves from here. A sample line, which is not forced, could run like this: 5…♘f6 6.d3 ♗e7 7.♗b3 0-0 8.c3 d5 9.e5 ♘fd7 10.d4 with a ne French-type position that looks quite passive for Black. D) 2…g6

282

Chapter 16: 2.♘f3: other lines

The Hyper-Accelerated Dragon with 2….g6 is mostly played in order to reach a Dragon set-up while avoiding the Rossolimo after 2….♘c6 3.♗b5 or the Moscow after 2…d6 3.♗b5+. It is the most common among the secondtier options for Black. 3.c3 White now has two main tries for an advantage: the rst is going for a Maroczy Bind with 3.c4, followed by 4.d4. This is probably the most critical for Black, but White needs to study quite a body of theory, as Black chooses the set-up. This means that you need to be ready for a range of lines. I decided against covering this complex option here as 2…g6 is a line you won’t face very frequently. Fortunately, there is a second option in 3.c3, which ts nicely with our repertoire choices against 2… e6 and 2…a6. After 3.c3 we again get a c3-Sicilian while avoiding the most respectable set-ups available to Black. In this version of the Alapin, Black is committed to …g7-g6 and has trouble equalizing completely. Black now needs to choose when to play …d7-d5: right away, in order to recapture with the queen, or wait until White has played d2-d4.

White’s centre at all and will su er in a cramped position. I like the prophylactic 6.h3 best. White prevents …♗g4, and Black struggles to nd a good square for this piece now: 6…♘f6 7.♘c3 0-0 8.♗e3.

D1) 3…♗g7 This natural move is by far the most popular, but White has good chances for an edge against it. 4.d4 cxd4 5.cxd4 d5 A frequently seen set-up below master level is starting with 5…d6. This set-up is too passive. Black does not attack

We rarely go for these closed centre positions that are similar to a French or Caro-Kann Advance. But here it is the strongest option as the g7-bishop is not well placed. If you prefer a more open centre solution, 6.exd5 ♘f6 7.♗c4!? ♘xd5 8.♕b3 is a low-maintenance alternative that still puts some

283

284

White is better with natural moves, supporting the centre. 6.e5

Chapter 16: 2.♘f3: other lines

pressure on Black. However I need to emphasize that advancing is stronger, and Black struggles to equalize here. D11) 6…♘c6 7.♗b5!

This is a very important point in this line. White gets the option of taking on c6 now, creating a potentially weak pawn on the c- le, and obtaining control of the c5-square as well. D111) 7…♗g4 By far the most popular move here, but is does not address White’s intentions. 8.♗xc6+! bxc6 9.♘bd2 Black will not manage …c6-c5 under satisfactory conditions. Let’s look at a sample line. 9…♘h6 10.0-0 0-0 11.h3 ♗xf3 12.♘xf3 ♕b6 13.b3

285

This prepares both ♗a3 and ♗f4. If Black is stuck with the pawn on c6 he will be worse forever. 13…c5 14.♗f4! White threatens 15.♕c1/♕d2 and is close to winning already. D112) 7…♘h6 8.0-0 0-0 8…♗g4 9.♗xc6+ bxc6 10.♘bd2 transposes to 7…♗g4, with an excellent game for White. 9.h3 This is not strictly necessary, but with …♗g4 prevented it is not clear how Black will develop harmoniously. 9…f6 10.♘c3 We go for a more sophisticated solution than 10.exf6, which is also a bit better for White. 10…fxe5 Black goes for a forced line. 11.♗xc6 bxc6 12.♗xh6 ♗xh6 13.♘xe5 ♕d6 14.♘a4

286

Chapter 16: 2.♘f3: other lines

♘c3-e2-f4 is a key idea to follow up with. Black is in pretty bad shape as he has no safe place for his king. D12) 6…♗g4

With a very interesting position. Usually, two bishops are a lot stronger than two knights, but here we have fantastic outposts on c5 and e5. I prefer White here (I am rather known for my bishop pair addiction, so that tells you something). D113) 7…♗d7 8.♘c3 e6

A very logical move. Black pins the knight that supports d4. 7.♗b5+! This a very important move to remember. Black will be OK if White does not opt for this disruptive check. 7…♘d7 This is better than …♘c6, but the knight does not put any pressure on d4 now, and is rather misplaced. 7…♘c6 8.♗xc6+ bxc6 9.♘bd2 again transposes to 6…♘c6 7.♗b5. 8.0-0 a6 9.♗e2

9.h4! Why not? Black has weakened the kingside, so let’s go! 9…h6 To have …g6-g5 after h4-h5, but Black’s kingside is even weaker now. 10.♗e3 ♘ge7 11.♗d3! ♘f5 12.♗xf5 gxf5 13.0-0

287

288

Chapter 16: 2.♘f3: other lines

White is better here due to his space advantage and Black’s slightly awkward pieces. Note that it is harder for Black to play …f7-f6 after a trade on f3 as the light squares are weakened a lot. D2) 3…♘c6 This is actually an interesting move order twist, but most of the time Black players just go for it without any speci c idea in mind, and then get into trouble. 4.d4

D21) 4…cxd4 This leads to an excellent game for White. 5.cxd4 ♗g7 The issue is that 5…d5? 6.exd5 ♕xd5 7.♘c3 leads to a disaster for Black. White will push the d-pawn next. 6.♘c3 d6 7.h3 And White is comfortably better, as is usual after a passive set-up with …d7d6.

Instead 5.e5 is less convincing. It is not bad at all, but Black has some pluses compared to other similar lines. He has not traded on d4 yet and the bishop is often better on f8 than on g7, at least for the moment. We instead capture on d5, getting a good IQP position that we examine via 3.c3 d5 4.exd5 ♕xd5 5.d4 ♘c6. 5…♕xd5 And we have transposed to line D31 below. D3) 3…d5

This is Black’s best option, I think. 4.exd5 Again, 4.e5 is possible, but the same logic as after 3…♘c6 4.d4 d5 5.e5 applies – I rather want to have Black’s bishop on g7 before I push the e-pawn. 4…♕xd5 5.d4

D22) 4…d5 Very rare, but not bad for Black. 5.exd5 289

290

Chapter 16: 2.♘f3: other lines

10.0-0 e5

D31) 5…♘c6 This is very rare, but Raja Panjwani makes a good case for this move in his book The Hyper-Accelerated Dragon. Black develops the queenside rst, which is signi cant in lines like 6.♘a3 ♗g4 7.♘b5 0-0-0! and Black is ne. 6.♗e2 This looks very modest, but it makes some sense here. White’s idea is to play c3-c4 and d4-d5 to kick the knight, but we need to cover 6…♕e4+ rst. Also, it is important to note that exchanging on d4 is unattractive for Black, as this would allow a quick ♘c3 and d4-d5 push. 6…♗g7 As mentioned, 6…cxd4?! is not advisable due to 7.cxd4 ♗g7 8.♘c3 ♕d8 9.d5 and White is much better, with chances for a quick knockout. 7.c4 ♕d8 This is probably best, as the queen would be exposed on d6. ♘c3-b5, for instance, might be a nuisance. 8.d5 ♘d4 9.♘xd4 cxd4 9…♗xd4? loses too much time: 10.♘c3 ♘f6 11.♗h6 and Black is in terrible shape. 291

A very unbalanced position! White can try various moves, but I think a simple solution is possible that still poses problems for Black. 11.dxe6 ♗xe6 12.♕a4+ ♕d7 13.♕xd7+ ♗xd7 14.♘a3 We will go ♖d1 and ♘b5; ♗f3 also ts into this position. I think we still have a slight pull. Panjwani assesses 11.dxe6 as equal – it is certainly somewhere between these two evaluations! D32) 5…♗g7

6.♘bd2 This is a rare move at this point, but it is not worse than the more popular 292

Chapter 16: 2.♘f3: other lines

moves, and probably less likely to be expected by the black player. It is mentioned in Raja Panjwani’s book, as one reason why he prefers the 5…♘c6 line over 5…♗g7. It is also what Stock sh recommends after a while, so why not go for this fresh move? 6…cxd4 This looks like the critical choice. Taking on d4 after ♘c3 is out of the picture makes a lot of sense. If Black plays di erently, ♗c4 and 0-0 are easy moves to make. White should be ne there with excellent development. 7.♗c4 ♕h5 Or 7…♕d8 8.♘xd4 ♘f6 9.0-0 0-0 10.♖e1 and White is slightly better, enjoying the more active pieces. 8.♕b3 A good zwischenzug, forcing 8…♘h6. Panjwani likes 8.0-0 even more, but I think White is somewhat better in both cases. 8…♘h6 9.♘xd4

Conclusion Amongst the rare lines, 2…g6 is the most challenging. Against the other moves we will get a better or comfortable position without huge e ort. The delayed Alapin serves us well again against 2…g6, leading to interesting positions. In fact many opponents won’t know this line very well and will opt for set-ups that are questionable. Even against higherrated opponents you will get lines like 3.c3 ♗g7 4.d4 cxd4 5.cxd4 d5 frequently – and this is excellent for White. Probably the IQP positions are the most critical, but they are ne for us and still not that well-known.

White is slightly better due to the more active pieces and quicker development. The moves 0-0 and ♘2f3 are coming next. 293

294

Sample games

 

Sample games Game 7 SI 31.7 (B31) Eduardas Rozentalis 2619 Magnus Carlsen 2552 Malmö 2004 (7)

Instead, 12…♘d4 13.♘xd4 cxd4 14.♗h6 is the main line of the theoretical section. White’s position is easier to play. Sometimes Black goes for 12…♖e8 to preserve the bishop after 13.♗h6 ♗h8. Now 14.♘eg5 is important. White needs to remove the e6-knight in order to play ♕f4 next. White enjoys a zerorisk small pull in this position type. 13.exf6 exf6 14.♗h6

1.e4 c5 2.♘f3 ♘c6 3.♗b5 The 16th World Champion Magnus Carlsen has been very successful with Sicilian sidelines on the White side. In this game he defends Black’s cause against the Rossolimo Variation. 3…g6 4.0-0 ♗g7 5.♖e1 ♘f6 6.e5 ♘d5 7.♘c3 ♘c7 8.♗xc6 dxc6 9.♘e4 ♘e6 10.d3 0-0 11.♗e3 b6 12.♕d2

12…f5 The upside of this move is that it absolutely forces White to take and thus Black gets rid of the cramping e5pawn. It has drawbacks, as Rozentalis will demonstrate in the game.

295

Very natural play. Black’s bishop is an important defender of the kingside, so we trade it o . Note that Black has played many pawn moves that created small weaknesses like on d6, e6, c6 and some more. White’s pawn structure, on the other hand, is completely sound and impossible to attack. One important piece is the knight on e4, of course. Black has no clear way to get rid of it, because …f6-f5 would weaken his position too much. 14…♕d7 A creative set-up by the 13-year-old later World Champion. It seems better to play more conventional though.

296

Sample games

Here 14…a5 is Black’s main move. One idea is to play …♖a8-a7 to enter this rook into the play. 15.♗xg7 ♔xg7 16.♘g3 ♖a7 17.♖e2 ♖e8 18.♖ae1 ♖ae7 19.♕e3 is just a slight pull for White. 15.♗xg7 ♔xg7 Black’s intention is to place the bishop on b7 after …♘d4 has undoubled the cpawns. 16.a3 19…♖ad8 Stopping 20.♘d6. Allowing it with 19…♖ae8 20.♘d6 ♖xe1+ 21.♖xe1 c5 22.h4 is also quite uncomfortable to play for Black. One issue for Black in this line is to determine which mostly passive defence is the most precise while you never have any active play. 20.h4 h5

A exible move by Rozentalis, introducing the idea b2-b4 into the position. The more direct 16.♖e3 ♗b7 17.♖ae1 ♖ae8 18.h4 was a good alternative. 16…♘d4 Clarifying the central pawn structure. It is always tough to keep the tension. After 16…♗b7, 17.h4 or 17.♖e3 are easy moves to play. One point of ♖e3 is that after …♘d4 you don’t need to trade on d4 immediately but can rather go ♖ae1 instead. 17.♘xd4 cxd4 18.b4 ♗b7 19.♕f4

297

Almost an automatic response. Allowing h4-h5 can’t be good. 21.f3 This is inaccurate, but it is extremely tough to spot why. It was better to play 21.a4. Now if we get the sequence played in the game, we spot the di erence: 21…c5 22.bxc5 298

Sample games

♗xe4 23.♖xe4 bxc5 24.♖ae1 and now the game move 24…♖de8 fails due to 25.♖xe8 ♖xe8 26.♖xe8 ♕xe8 27.♕c7+ ♔h6 28.g3 (the white king is safe now compared with f2-f3 being played) 28…♕xa4 29.♕d8 ♔h7 30.♕e7+ ♔g8 31.♕xf6 with an excellent ending for White. 21…c5 22.bxc5 ♗xe4! 23.♖xe4 bxc5 24.♖ae1 ♖de8 25.g4 Now 25.♖xe8 ♖xe8 26.♖xe8 ♕xe8 leads nowhere as Black has …♕e1+ available. 25…♖xe4 25…f5 was possible as well. After 26.♖xe8 ♖xe8 27.♖xe8 ♕xe8, again Black has enough play with the white king being exposed. 26.♕xe4

the correct reply and got a very bad endgame. The alternative was to play 28.fxg4, keeping the tension and an annoying edge. White has the only open le and h4-h5 is an additional option. One possible line is 28…♕c7 29.♕e7+ ♖f7 30.♕xc7 ♖xc7 31.♖e6

analysis diagram

and Black needs to defend this worse rook endgame. 28…♕xe7 29.♖xe7+

In this endgame Black still has some questions to answer. He should be OK, but it can become tricky – especially with the clock ticking down. 26…♕d6 27.♔g2 hxg4 28.♕e7+ Objectively speaking, this leads to a clean draw. On another level, it actually WON the game as Carlsen failed to nd

299

29…♔h6? Carlsen rejected 29…♖f7, probably seeing a ghost in the king and pawn

300

Sample games

ending. 30.♖xf7+ ♔xf7 31.fxg4 ♔e6 32.♔f3 f5 is a dead draw. 30.fxg4 Now Black is in deep trouble. He is passive and his pawns are much easier to attack than White’s. 30…c4 This is the best try, trading pawns. 31.♔f3 31.♖xa7 cxd3 32.cxd3 ♖e8 gives Black more chances to hold compared to the game. 31…cxd3 31…a5 was the toughest defence, but most likely Carlsen was under time pressure. There is no way he would have misjudged 29…♖f7 with time on the clock. 32.cxd3 ♖c8 33.♔f4 33.a4 was even better, but Rozentalis keeps a large advantage with his move, too. 33…g5+ 34.hxg5+ fxg5+ 35.♔e4 ♖c1

36.♔xd4 This seems to spoil it to a draw, which is easy to gure out with computer help – but not at all in an actual game with low time. 301

36.♖e6+ ♔g7 37.a4 was still giving excellent winning chances. 36…♖g1 37.♖e4 ♖a1 38.♔c5 38.♔e5 ♖xa3 39.d4 ♔g6 also holds for Black. 38…♖xa3 39.d4

39…♔g6 This loses, but the computer lines after the drawing move are too hard to nd with low time. 39…♖c3+ was the only way to draw: 40.♔d6 (40.♔b4 ♖d3 41.♔c4 ♖a3 and White can’t push the pawn now) 40… ♖c4 41.♔e5 ♔g6 and again the dpawn is stopped. 40.d5 Now the d-pawn is too fast. 40…♖c3+ 41.♔b5 ♔f6 42.d6 ♖d3 43.♔c6 a5 44.d7 ♖c3+ 45.♔b7 ♖b3+ 46.♔c7 ♖c3+ 47.♔d8 ♔f7 48.♖e7+ ♔f8 49.♖e5 1-0 A ne game by Rozentalis, who kept annoying pressure throughout. Of course there was an up and down in the technical phase, but it is no coincidence that Black made a mistake – he had to defend all of the time, which very likely took a lot of time. 302

Sample games

The practical problems in seemingly simple endgames should not be underestimated. Getting into a slightly worse endgame with a huge time disadvantage is awful for the opponent to handle and often simply disastrous. Game 8 SI 30.5 (B30) Magnus Carlsen 2872 Boris Gelfand 2740 London ct 2013 (10)

1.e4 c5 2.♘f3 ♘c6 3.♗b5 Boris Gelfand played the Sveshnikov in his 2012 World Championship Match against Vishy Anand very successfully. In fact, Anand quickly switched to the Rossolimo and managed to pose more problems there. 3…e6 4.0-0 4.♗xc6 bxc6 5.b3 was Anand’s choice in the match and a good alternative if you like to add something new once you’ve mastered the lines presented here. 4…♘ge7 5.♖e1 a6 6.♗f1 d5 7.exd5 ♘xd5 8.d4 ♘f6 9.♗e3

303

9…cxd4 Some months after this game, Gelfand tried to improve with 9…♗e7 in Anand-Gelfand, Tal Memorial 2013. Anand played 10.c4, while 10.dxc5 is my recommendation in the theoretical section. Both moves are ne and give White chances for a small pull. 10.♘xd4 ♗d7 11.c4 ♘xd4 12.♗xd4 ♗c6 13.♘c3 This set-up is very common in the 6.♗f1 line. White has some space advantage and his queenside majority is a bit easier to utilize than Black’s extra pawn on the kingside/in the centre. Objectively speaking, Black is about equal, but I think White’s ideas are clearer and thus he has the easier task ahead. Carlsen beautifully uses White’s assets in this game. 13…♗e7 14.a3 a5 The type of decision that you can burn serious time on. Pawns don’t go backwards, so Black had to decide if it was worth it to weaken the queenside in order to stop White’s b2-b4 push. I would have been reluctant to play … a6-a5, but the alternative 14…0-0 15.b4 is also more comfortable for White. The following idea is worth remembering: after, for example, 15… ♕d7

304

Sample games

analysis diagram

16.♖a2!? ♖ad8 17.♖d2 is an elegant way to use the rook in the centre. 15.♕d3 Connecting rooks and preparing to go to g3. 15…0-0 Here 15…a4 looks interesting, trying to freeze the white queenside and maybe use the a5-square for the queen. There are some issues though: 16.♖ad1 0-0 17.♕g3 ♕a5? (this fails. Black should have played 17…♕c8 and limited the damage to a slight disadvantage) 18.♘d5! and Black is close to lost here. 16.♖ad1 ♕c7 17.♗e5 ♕b6 18.♕g3 ♖fd8 19.♖xd8+ ♕xd8 20.♖d1

305

The play has developed quite logically so far. White has the more active pieces and Black tries to trade down to reduce the pressure. 20…♕b6? This turns out to be a serious mistake. Black should have played 20…♕f8 according to the engines. However the coming complications are very di cult for a human player to assess: 21.b4 (this is critical, of course) 21… axb4 22.axb4 ♘h5 (otherwise b4-b5 will be very annoying; Black needs to go for the coming sequence) 23.♕h3 ♗xb4 24.♕xh5 f6 (Black will get the piece back, but White has more than one critical test now) 25.♗d3 (25.♗d4 e5 26.♗xe5 fxe5 27.♕xe5 is a di erent way to be a pawn up against the bishop pair; Black has enough compensation here) 25…g6 26.♕h3 fxe5 27.♕xe6+ ♔h8 28.♕xe5+ ♕g7 29.♕xg7+ ♔xg7 30.♘e4 ♖a3 31.f3 and again, Black has enough for the pawn to hold. 21.♗d4 ♕b3 This is forced as 21…♕a6 22.b4 is awful for Black. 22.♖d3! ♘d5 is the threat now – Black has no choice. 22…♕c2

306

Sample games

move available. 29.♕b6 e5 30.♖d1 Here, 30.♖d8 g6 31.b5 ♗e4 32.♖d7 was more precise.

23.b4 axb4 24.axb4 ♘h5 Again the only move. After something slow like 24…h6 White wins: 25.b5 ♗e8 26.♘e4 ♔h7 27.♖f3 with a deadly attack. 25.♕e5 ♗f6 26.♕xh5 ♗xd4 27.♖xd4 ♕xc3

There was no real alternative along the way after Black had played 20…♕b6. Is Black’s position all that bad? 28.♕a5!! It is bad! What a shot – Black’s weak back rank is exploited nicely with this queen sacri ce. 28…♖f8 The rook is forced back into this passive position; there was no other 307

analysis diagram

Black is in dire straits and will very likely lose a pawn soon. 30…g6 31.b5 ♗e4 32.♕f6 32.♖d7 would have been stronger, transposing to the note to move 30. 32…h5 Gelfand should have played 32…♗f5 here. White has excellent winning chances anyway, but this is not so easy to convert: 33.h3 (the obvious 33.♖d5 might fail to do the job in view of 33… ♗e6! 34.♕xe5 ♕xe5 35.♖xe5 ♖c8 36.c5 ♔f8 and Black will hold) 33… ♗e6 34.♕e7 ♕c2 35.♖d8 ♖xd8 36.♕xd8+ ♔g7 37.♕d6 ♕e4 38.c5 ♗d5 (this is the important defensive resource) 39.c6 bxc6 40.b6 c5 41.♕xc5 ♗b7 and I am not sure if White will be able to win this. b7 is very safely blockaded. 33.h4 ♗f5 34.♖d5

308

Sample games

34…♕c1 Black has to give up the e5-pawn. Just covering it won’t help as White just pushes the c-pawn: 34…♖e8 35.c5 ♕b2 36.c6 and so on. 35.♕xe5 ♗e6 36.♖d4 ♖a8 37.♕e2 ♔h7 38.♖d1 ♕c3 39.♕e4 ♖a1 39…♖a7 puts up more resistance, but after 40.♕f4, with ♖c1 and c4-c5 next in mind, White should win anyway. 40.♖xa1 ♕xa1 41.c5 ♕c3 42.♕xb7 ♕e1 Gelfand tries for some play on the back rank. After 42…♕xc5

♗d5 45.♕e5+ f6 46.♕c7+ ♕xc7 47.bxc7 ♗b7 48.♗b5 ♗c8 49.♗e8 and the passed pawn decides. 43.b6 43.♕e7 was quicker, but the result is not in doubt. 43…♗c4 44.♕f3 ♕xf1+ 44…♗xf1 45.♕xf7+ ♔h8 46.♕f6+ ♔h7 47.♔h2 also wins convincingly for White. 45.♔h2 ♕b1 46.b7 ♕b5 47.c6 ♗d5 48.♕g3 1-0 This was Carlsen’s best game of the Candidates Tournament, London 2013, one of the best and most exciting tournaments ever. Carlsen used White’s asset in exemplary fashion, posing small problems that were not all that easy to solve. Gelfand needed to defend very imaginatively and ultimately failed in nding the best way to do it. White has the slightly more active pieces and may use the queenside pawn majority, gaining more space on this part of the board. I think White can be quite satis ed in general, playing a position without any weaknesses or problems and being able to ask some questions. Game 9 SI 1.6.8 (B30) Benjamin Bok 2572 Sergey Tiviakov 2677 Groningen 2014 (9)

analysis diagram

43.b6 White wins in a purely technical way, for instance: 43…♔g7 44.♕b8

1.e4 c5 2.♘f3 ♘c6 3.♗b5 d6 4.0-0 ♗d7 The main line to break the pin. 4…♗g4 has been played by Tiviakov as well. As

309

310

Sample games

shown in the theoretical overview, it is very risky for Black if you react forcefully. 5.c3 ♕b6 6.♗a4 ♘f6 7.d4 is the sequence to remember. Black’s position is shaky. 5.♖e1 ♘f6 6.c3 a6 7.♗f1 ♗g4 8.h3 ♗xf3 8…♗h5 9.g4 ♗g6 10.d4! is a key line to remember! Some black players who don’t know the precise theory will stumble into this position and su er greatly against this powerful central play. See the theoretical section for details. 9.♕xf3 And we have gained the bishop pair, as it happens quite frequently in our repertoire. It’s good to have some longterm assets to play for. 9…g6 10.d3 ♗g7 Black’s compensation for giving up the bishop are his uent and quick development and sound structure. Still, White’s chances are slightly preferable, as is nicely demonstrated in this game. 11.♗e3 0-0 12.♘d2 ♘d7 13.♕d1 b5 14.♘f3

White’s regrouping over the last moves is important to remember. We need to get d3-d4 in, and this set-up gives maximum control over this key square. It is a slow set-up, but Black has no way to stop it. 14.a3 is also possible. I prefer to play d3-d4 immediately, as 14.a3 makes it more likely that more pawns are traded on the queenside. 14…b4 15.d4 15.♕a4 is the move I recommend in the theoretical section. It’s a close call between it and the game move, which is of course also perfectly logical. Let’s check a sample game for 15.♕a4. After 15…♕c7 16.d4 ♖fb8 17.♖ac1 bxc3 18.bxc3 ♘a7 19.♕c2 ♕a5 20.♘d2 White was in excellent shape in Vocaturo-Colon Garcia, Sitges 2015. The position opens up and the c4square is a great place for White’s knight. 15…bxc3 16.bxc3 ♕a5 This is now possible, but it does not help all that much. The centre and bishop pair still remain the most important features of the position. 17.♕d3 ♖fc8 Now 17…cxd4?! only makes sense if it is continued with 18.cxd4 ♘b4?.

311

312

Sample games

control and again Black’s direct counterplay ideas don’t work: 19… cxd4 20.cxd4 ♘b4 21.♕d2 ♘xa2 22.♖xc8+ ♖xc8 23.♕b2!.

analysis diagram

However this worsens Black’s cause considerably. After 19.♕b3, intending 20.♗d2, the pin is already deadly. 18.♖eb1 Black was now threatening the …cxd4/ …♘b4 sequence for real, so putting a rook on b1 makes a lot of sense. It seems however that using the other rook was even better: 18.♖ab1 and now: A) The immediate 18…cxd4 19.cxd4 ♘b4 again does not net much due to 20.♕d2 ♖ab8 21.♖b2 and White has ♖eb1 to follow. Also note that ♗g5 may be a quite annoying move as well. Black has no comfortable way to deal with the threat to the e7-pawn; B) The greedy 18…♕xa2 is not stumbling into a queen trap due to 19.♖a1 c4, but White is still clearly better after either 20.♕d1 or 20.♖xa2 cxd3 21.♗xd3. The point of 20.♕d1 is that 20…♕b3 21.♕d2 ♖cb8 22.♖eb1 ♕xb1 23.♖xb1 ♖xb1 24.♕a2 will win the c4- and a6-pawns at the end; C) 18…♕a4 19.♖ec1. Compared to the game White’s rooks are more actively placed. White keeps the central 313

analysis diagram

This is extremely strong. White not only has ♖a1, but also ♕b7 on the cards. After this move Black is actually already lost, for example: 23…♕c2 24.♗xa6 ♖c7 25.♕xc2 ♖xc2 26.♖b7 ♘f8 27.♖xe7 and White is winning. 18…♕a3 Black provokes the following doubling on the b- le. 18…♕a4 was a good alternative. White now does not have the easy doubling available and needs to nd a way to use the a1-rook. 19.♖b2 is my suggestion. I’d still prefer White here, but Black is not badly placed. 19.♖b3 ♕a4 20.♖ab1 ♕xa2 This was tempting and of course very principled. Black also had the attractive looking move 20…♘ce5, but it does not solve the problems: 21.♕d1 ♘xf3+ 22.gxf3! (this keeps the a-pawn protected due to concrete lines) 22…cxd4 23.cxd4 ♕xa2 314

Sample games

24.♖b7 ♖cb8 25.♗d3 and White is in good shape, retaining pressure while winning the e7-pawn back in most lines. 21.♖b7 White also had the less obvious move 21.♕d1, threatening to trap the queen with 22.♖a1. The somewhat computerish continuation 21…♖ab8 22.♗c4 ♕a4 23.♕f1 ♘a5 24.♖xb8 ♖xb8 25.♗xa6 ♖xb1 26.♕xb1 ♘b3 27.♕f1 leads to an equal position. White still has the pair of bishops, but it is di cult to make them count.

21…cxd4?! This is the rst serious inaccuracy of the game. 21…♘f6 was better, keeping equal chances. After 22.♘d2 White has enough for the pawn, but he is not better. (It is important to check 22.d5; now after 22…♘a5, 23.♖xe7? is bad due to 23…♗f8, trapping the rook.) 22…♘a5 23.♖7b2 ♕a4 is a possible continuation. Stock sh now suggests 24.♕e2.

315

analysis diagram

Planning to go ♕e1+♖a1. This is a tough position to analyse, with many possibilities for both sides. It is however clear that Black is better o here than in the game. 22.cxd4 ♘cb8? After this move Black is lost. A) The best defence was 22…♖ab8 with the likely continuation 23.♘d2 ♖xb7 24.♖xb7 ♘f8 25.d5 ♘e5 26.♕xa6 ♕xa6 27.♗xa6.

analysis diagram

White is much better, but it is of course not lost yet for Black; B) 22…♘f6 23.d5 (compared to the position with the c-pawns still on the 316

Sample games

board there is an important di erence) 23…♘a5 24.♖xe7 ♗f8 25.♖a7 and White saves the rook, with an overwhelming position. 23.♖c1 Now Black is falling apart. His pieces have no coordination. 23…♖d8 Or 23…♖xc1 24.♗xc1. White is now threatening to play 25.♕c3 to invade on the c- le. 24…♕a4 25.♕c3 ♘f8 26.♕c8 ♕c6 27.♖c7 ♕b6 28.♖xe7 is a possible line, winning for White. 24.♕c3

Too many threats. ♗c4 is on the cards as well. 24…♕e6 25.♗c4 ♕xe4 26.♗xf7+ ♔h8 27.♕b3 ♕f5 28.♗e6 ♕f6 29.♘g5 1-0 The defensive task for Black in this line is not easy. White enjoys some longterm pluses and Black’s position is solid, but somewhat passive. White’s game is easy to play, hoping to slowly open up the position for the bishop pair. If you play precisely you should be able to get a small, but stable advantage.

Sebastian Maze 2613 Julian Plenca 2411 Germany Bundesliga B 2016/17 (3)

1.e4 c5 2.♘f3 d6 3.♗b5+ ♘d7 4.c3 I recommend to start with this move to create a safe space for the bishop on c2. There are many alternatives of similar value, with quite some scope for creativity. Continuations like 4.♗a4!? or the fashionable 4.0-0 a6 5.♗d3 are well worth exploring if you look for new ideas at a later stage. 4…♘gf6 5.♕e2 a6 6.♗a4 b5 7.♗c2 You need to start with 4.c3 to be able to play this set-up. 7…♗b7 8.d4 I like the classical, harmonious nature of White’s position. You get a nicely supported centre and a good amount of space. Black’s set-up is quite exible though. He’s gained some space on the queenside, which might be the start of some counterplay, but on the other hand also has weakened this ank to some extent. I believe the chances are roughly equal, but given the choice I’d still rather be White – pawns on d4 and e4 are nice to have. 8…e6 9.0-0 ♗e7 10.♖e1 ♕c7 One of my games continued 10…0-0 11.♘bd2 ♕c7 12.♘f1 ♖ac8 13.♗d3 ♖fe8.

Game 10 SI 1.3.3 (B51) 317

318

Sample games

analysis diagram

Now I played 14.a4, which I don’t like with hindsight. Today I prefer to just go 14.♘g3 and leave the a-pawn at home. There followed 14…♘b6! 15.axb5 c4 16.♗c2 axb5 17.♘g3. Now the a- le is open, with trades to follow. This is helpful for Black, and the drawback of my 14.a4, allowing equality. After some later weirdness this ended in a draw in Sielecki-Holtel, Germany Bundesliga B 2017. 11.♗d3 e5

Black is worried about e4-e5 and decides to stop the advance by changing the central structure. This was not neccessary at all. He could 319

have played 11…0-0, like Holtel did against me. Now 12.e5? is ill-timed and gives Black the upper hand: 12…♗xf3 13.gxf3 dxe5 14.dxe5 c4 15.♗c2 ♘d5, and White still has most of his pieces at home on the queenside and his pawn structure is damaged. Black is clearly better. 12.dxe5 As, on a very general level, d4 is the better pawn compared to d6, a trade on e5 should be clearly motivated by a concrete idea, like taking over control over d5 or a favourable piece trade. Here White’s idea is revealed on the next move. Instead, White also had options to keep the tension with moves like 12.a4 or 12.b3, in both cases with a very complex position and about equal chances. I believe those moves were to be preferred compared to the game continuation. 12…dxe5 13.c4 It is important to note that something slow like 13.♘bd2?

analysis diagram

320

Sample games

13…c4! 14.♗c2 0-0 15.♘f1 ♘c5 16.♘g3 ♖fd8 is just awful for White. The text move is the point of White’s play. 13…bxc4?! It was much better to play 13…b4. It’s hard to say what Black disliked about this advance. It keeps the pawn structure sound and should be absolutely ne for Black, for example: 14.♘bd2 ♘f8 (Black may also castle rst and only play this after …♖fd8. In any case the knight will be ideal on e6) 15.♘f1 ♘e6. Here I’d rather have Black, but of course it is quite hard to get something tangible in this symmetrical centre situation. As mentioned above, I recommend to not trade on e5 in the rst place. 14.♗c2 14.♗xc4 is a nice alternative that can get easily overlooked. Doesn’t it just blunder e4? 14…♘xe4 15.♗xf7+ ♔xf7 16.♕c4+ ♔e8 17.♖xe4 ♗xe4 18.♕xe4.

analysis diagram

This is a forced sequence after 14.♗xc4. White is an exchange down, but Black has lots of weaknesses and 321

cannot castle anymore. If White had better development he would certainly be for choice, but his queenside is not yet developed. Stock sh gives the expected 0.00. It is not terribly relevant theoretically, but a fascinating longterm sacri ce. On a more sober note: the actual game move 14.♗c2 is just a bit better for White. 14…0-0 15.♘bd2

White just intends to regain the pawn and keep the better pawn formation. 15…♘b8 Heading towards d4. The knight certainly had little scope on d7. Instead, 15…c3 looks like a good idea, but White does not have to take it. If 16.♘c4! cxb2? is very compliant, but it needs to be checked: 17.♗xb2 ♖fe8 18.♘fxe5 and White will follow up with a later e4-e5 after a knight move/capture from e5. Black faces a terribly dangerous attack. 16.♘xc4 ♘c6 17.♗d2 ♖ac8 Or 17…♘d4 18.♘xd4 (Stock sh also likes 18.♕d3, but this feels very computerish. Just taking on d4 is better 322

Sample games

for White as well) 18…cxd4 19.♗d3 a5 20.♖ac1 and White has a comfortable game. Note that f2-f4 is a powerful resource for White. 18.♖ac1 ♘d4 Black does not want to just sit with the worse pawn formation. 19.♕d1 19.♘xd4 cxd4 20.♗d3 instead was simple and strong. ♘a5 will be very irritating for Black. 19…♘d7 20.♗d3 ♗c6 21.♘e3

This was probably the regrouping that White had in mind when he opted for 19.♕d1. The knight is aimed at f5, switching the play to the kingside. A surprising turn of events, given that Black’s weaknesses were on the queenside. 21…♕b7? It looks natural to get the queen o the c- le, but it fails due to concrete tactics. It was better to play 21…♖fe8, after which White is still for choice, but at least there are no immediate problems. 22.♘xd4 exd4

323

22…cxd4 23.♘f5 ♖fe8 24.♗h6 is no fun either. 23.♘f5

‘Knife’ to f5 was the term coined by Greek GM Vasilios Kotronias. This is certainly a piece that might hurt Black soon. 23…♖fe8 23…♘e5? 24.♗xa6 is one of the tactics allowed by the knight. 24.e5 24.♕g4 g6 25.♘h6+ ♔g7 26.e5 leads to the game position by transposition. 24…g6 Tough to defend this. The alternatives don’t convince either: A) 24…♘f8 fails due to 25.♕g4 ♘g6 26.e6 (the decisive break) 26…♗f8 27.♘xd4! and White wins as 27… cxd4? 28.exf7+ ♕xf7 29.♗c4 ♗d5 30.♗xd5 ♕xd5 31.♖xc8 loses the house; B) 24…♔h8.

324

Sample games

course can not take the piece. 29… gxf5?? 30.♗f6+ ♔xh6 31.♕g5# 27…♗f8! An excellent defence; the only one to put up resistance. 28.♕g3 ♔h8

25.♗h6! (one of those shots that look simple once the computer points you to it. Still not at all easy to gure out over the board) 25…gxh6 26.♘xh6 (Black is defenceless here against the white forces pouring in) 26…♘f8 27.♘xf7+ ♔g7 28.e6 is one line to check. White wins after 28…♗xg2 29.♕g4+ ♘g6 30.♘e5. 25.♘h6+ ♔g7 26.♕g4 ♗d5

27.b3 Here 27.♗f5! ♖cd8 28.b3 was a lot stronger, which is not obvious at all. White has a strong attack, avoiding the ♗f8 resource of the game. Now 28… ♗f8 is answered by 29.♗g5 with a great position for White. Black of 325

29.♘g4 Both 29.♗g5 and 29.h4 were promising as well. It is often tough to choose from di erent attacking possibilities. If you are unsure, maybe because of time issues, my advice is to play with a piece or pawn that has not yet taken part in the attack – let as many as possible join the party. In this case 29.h4 would t the bill. 29…♗g7 30.♗h6 ♗xh6! 31.♘xh6 a5 32.h4 You can safely assume that time trouble played a part here, given the complicated nature of the game. The next moves contain multiple mistakes by both sides. 32…♕b4

326

Sample games

33.♗c4 The shot 33.♗f5! would have ended the game right here: 33…gxf5 34.e6 ♗xe6 35.♖xe6!. Very nice! White wins in style. 33…♕d2? 33…♗xc4 34.bxc4 ♖e7 was neccessary to put up a ght. 34.f4? 34.♗b5 was one of the easier wins. 34…♗xc4 35.bxc4 ♖f8 36.h5 ♖c6 Black correctly avoided 36…♕xa2? 37.hxg6 fxg6 38.e6 and White wins. Best was 36…♖ce8, holding for the moment. 37.♖cd1 ♕c2 38.♕h3

327

White misses a hard to spot resource. 38.♕f3 was the right move, keeping a huge advantage. 38…♘b6? The surprising 38…gxh5! seems to save Black! The dramatic sequence 39.♘xf7+ ♖xf7 40.e6 ♖xf4 41.exd7 ♕f2+ 42.♔h2 ♖d6 43.♖e8+ ♔g7 44.♖e7+ ♔f8 45.♖e8+ ends in a draw. 39.hxg6? 39.♖c1 ♕d2 40.e6 was decisive. 39…♕xg6 40.♘f5 ♕e6 Black has survived the time control and is ne now. 41.♖d3 ♘xc4 42.♖g3

42…♘e3! A necessary move, forcing an equal endgame. After other moves the attack is too much to handle, for example: 42…♖g8 43.♖xg8+ ♔xg8 44.♖b1 ♘b6 45.♖b3. 43.♘xe3 ♕xh3 44.♖xh3 dxe3 45.♖hxe3 ♖d8 46.♖1e2 ♔g7 47.♔f2 ♖d4 48.♔f3 ♔g6 49.g4 h5 Black is on track to a draw, now getting rid of a pawn weakness. 50.♖g2 hxg4+ 51.♖xg4+ ♔f5 52.♖g5+ ♔e6 53.f5+ ♔e7 54.♖b3 c4 55.♖b2 328

Sample games

♖d7 56.♖b1 ♖d3+? 56…c3 secured the draw: 57.♖b8 ♖d3+ 58.♔e4 c2 59.♖g1 ♖d1 60.♖b7+ ♔f8 61.♖b8+ ♔e7 with a repetition. 57.♔e4 Now Black is lost, at least against perfect play. 57…♖h6

58.f6+? The nal moves of this game are not very precise, but this is very understandable as these types of endgames are extremely tough to play while being short of time. I will still brie y mention the improvements, as I need to do as an annotator. Here 58.♖b7+ was winning. 58…♔d7? 58…♔d8 draws. 59.♖b7+ ♔c6 60.♖xf7 ♖h4+ 61.♔f5 c3 62.♖f8 ♖dh3 63.♔g6? Here 63.e6 was the win. 63…♔b7? 63…♔d7 draws. 64.♖g2? 64.f7 wins. 64…♖c4? 329

64…♖h2 was the last chance to draw. 65.♖c2 ♖g3+ 66.♔f7 ♖d3 67.e6 1-0 This line starting with 3…♘d7 4.c3 leads to very complicated play with chances for both sides. With very few early piece trades, you will certainly get a complex middlegame that allows you to play for a win while still being solid. The central pawn structures are tough to handle for both sides and often feature ideas that are more common in other openings, like the Ruy Lopez. This game shows that the middlegames tend to be very complicated, often getting to a crisis later in the game. Make sure to study some more games to get a better feel for the manoeuvres. This will help you to play quicker and have more time available for tough decisions that may come up later. Game 11 SI 1.8.7 (B52) Xu Yi 2439 Lars Goldbeck 2191 Tegernsee 2017 (9)

1.e4 c5 2.♘f3 d6 3.♗b5+ ♗d7 4.♗xd7+ ♘xd7 5.0-0 ♘gf6 6.♕e2 ♖c8

330

Sample games

This is a exible move that does not commit to a set-up yet. Black still may play with …e7-e6 or …g7-g6, depending on White’s choices. 7.c4 The Maroczy set-up serves us well in other lines, so there is little reason to go for a di erent approach here. 7…g6 Black may try for a completely original approach with 7…♘e5, which prevents the Maroczy Bind. However after 8.♘xe5 dxe5 9.♘c3 e6 10.d3 ♗e7 11.f4 exf4 12.♗xf4 0-0 13.e5 ♘d7 14.♘e4

on the kingside, giving White attacking chances against the black king. Black may also play 7…e6, but after 8.d4 cxd4 9.♘xd4 we will transpose to normal positions that we have examined via the 6…e6 move order in the theoretical section. 8.♘c3 ♗g7 9.d4 cxd4 10.♘xd4 0-0 11.b3 Going for b2-b3 and ♗b2 feels very natural here, but there is nothing wrong with 11.♗e3 either. 11…♖e8 A universally useful move in these structures. It covers e7 in case of a later ♘d5 and in some cases …e7-e6/…d6d5 plans are supported. 12.♗b2

White’s game was easier to play in Ni Hua-Gao Rui, China tt 2016. The pawn on e5 establishes the space advantage

12…e6? This is badly timed. The pawn on d6 is very weak and that causes immediate concrete problems. Black’s best was probably 12…♕a5, putting some pressure on c3 and connecting the rooks. I still prefer White after it, but Black is very solid, of course. Let’s have a look at some lines: 13.♖ad1 a6 14.♖fe1. We have played

331

332

Sample games

useful moves and centralized the rooks. What now for Black?

The direct moves, aiming for breaks, don’t work that well: 14…e6 (14…b5? fails due to 15.cxb5 ♖xc3 16.♕d2!) 15.♘f3, with some useful pressure on d6. After 15…♘e5 16.♘xe5 dxe5 17.♘a4 White is slightly better due to his better structure. Of course Black does not need to play 14…e6, but could opt for a waiting strategy. In this case we are going to slowly aim to prepare ♘d5 for later. Watch out to prevent the …b7-b5 and …d6-d5 breaks – if you succeed in doing that you will often get help by Black lashing out incorrectly – just sitting forever is not what Sicilian players are out for. 13.♘db5! Not a di cult move by any means, but it needs to be played now to exploit Black’s misstep. 13…♕b6 14.♖ad1 ♗f8

Forced moves up to this point after 12…e6. It needs to be emphasized that on general grounds, going …e7-e6 is not terrible for Black. Yes, it leaves d6 weak, but in Hedgehog structures this is not that relevant. The main issue here is the timing. 15.♔h1?! The engine’s suggestion of 15.♕f3! is very strong, with the brutal threat of 16.♘a4, attacking the knight on f6. After, for example, 15…♗e7 16.♘a4 ♕a6 17.♕e3, Black is busted. There are too many weaknesses to cover. 15…♖c6?! With 15…♖cd8 Black had better ghting chances, probably. After 16.♘a4 ♕c6 17.♕f3 ♗e7 18.♘xa7 ♕xe4 19.♘b5 White is much better though. 16.f4! The key idea is to push e4-e5, freeing up e4 for the knight and going for an attack on the black king. 16…a6 17.♘d4 ♖cc8 18.e5

333

334

analysis diagram

Sample games

21…♕a5 22.♕f3 ♗a3 23.♗a1 ♖cd8 24.♘xd5 ♘xe5

18…♘h5 Sadly, the only square for the knight. 19.♘e4 d5?! Black had to go for 19…dxe5 20.fxe5 ♘xe5. It is a winning position for White anyway, but this is murky at least, and requires precision: 21.g4 f5. White now wins with 22.gxh5 fxe4 23.♘b5 axb5 24.♗xe5

25.♘f6+! It is almost always possible to screw up – here 25.fxe5? ♕xd5 is not all that clear. The correct move ends the game quickly. 25…♘xf6 26.fxe5 ♘d5 We should check 26…♕xe5 brie y.

analysis diagram analysis diagram

and White crashes through on the kingside. All this is messy, while in the game it is a white win without any e ort needed at all. 20.cxd5 exd5 21.♘c3 Material is dropping now, as the d5pawn is too weak to defend. 335

This is a case where you should not just blitz out the obvious move, but rather look for other candidates. White wins immediately with 27.♘c6!, while the ‘obvious’ 27.♕xf6? ♕xf6 28.♖xf6 ♖e2! still prolongs the game. Sure, it is winning in the long run, but Black 336

Sample games

won’t resign anytime soon – which will probably happen after the best move. 27.♕xf7+ ♔h8 28.e6 ♘c3 29.♘e2 ♗b4 30.♖d7 1-0 Black went wrong with the badly timed 12…e6 move and was duly punished. Most of the thirty selected games feature GMs on both sides, but it is good to also check how the suggested repertoire fares on levels below that. The player on the black side, in most cases at least, is still higher rated than your typical opponent – and he lost quite drastically to a swift attack.

assessment much. This line is between slightly better for White and equal. I’d always prefer White because of the space, but Black is very solid, of course. 10.b3 ♕a5 11.♗b2 ♘c6 12.0-0 0-0

Game 12 SI 1.9.4 (B52) Magnus Carlsen 2843 Viswanathan Anand 2780 Bilbao 2012 (9)

1.e4 c5 2.♘f3 d6 3.♗b5+ ♗d7 Some months before, Anand had tried 3…♘d7 against Sergey Tiviakov and lost a true nightmare of a game. I rejected 4.d4 for the repertoire because it is too close to an Open Sicilian for my taste. There are no objective drawbacks to Tiviakov’s choice: 4…cxd4 5.♕xd4 a6 6.♗xd7+ ♗xd7 7.c4 e5 8.♕d3 h6 9.♘c3 ♘f6 10.0-0 ♗e7 11.a4 b6 12.b3 and White had excellent control in Tiviakov-Anand, Germany Bundesliga 2011/12. 4.♗xd7+ ♕xd7 5.c4 ♘f6 6.♘c3 g6 7.d4 cxd4 8.♘xd4 ♗g7 9.f3 ♕c7 At the time when this game was played, this move was still quite rare. In recent years there have been some new games that didn’t change the 337

13.♘ce2 13.♔h1 is a good alternative that has been tried in some games. With the bishop being on b2 the a7-g1 diagonal is more vulnerable than usual. The more common position of the bishop is on e3, where this is less of a problem. Carlsen plays ♔h1 just two moves later, it just is part of White’s general set-up at some point. 13…♖fd8 Anand aims for a …d6-d5 break. 14.♗c3 This is actually a crucial decision as the following sequence is quite forced. Again, going for 14.♔h1 was a way to prevent the …d6-d5 break that Black manages in the game. 14…♕b6 15.♔h1

338

Sample games

and threatens 18.♗a5 at the same time. 17…♖dc8?! It is tough to say why Anand rejected the more normal-looking 17…a5. If White plays in a similar fashion to the game it will be much easier to defend: 18.e5 ♘e8 (instead 18…♘d7 is probably less precise: 19.e6 d4 20.♗d2 ♘c5 21.exf7+ ♔xf7 No choice, as …e7-e5 was threatened. 15…d5! Black has managed his key break with some nice tactics to support it, as we will see. 16.♘xc6 16.exd5 ♘xd5 17.cxd5 ♖xd5 was the justi cation of the …d6-d5 break. Black will win the piece back with a good game due to the rook vs queen situation on the d- le. Carlsen’s move is the only way to ask some questions. 16…bxc6

analysis diagram

22.f4!. An instructive move, combining f4-f5 ideas with improving the knight via g1 to f3 in some cases. I’d rather be White here, but things are not clear as Black has some assets as well). Now 19.e6?! is actually dubious because of 19…f5!. The rook is just so much better on d8 now instead of c8. It supports the d-pawn and might even help to win e6 soon. White should avoid the e5-e6 push here and rather play a slow move like 19.f4, with about equal chances. 18.e5 ♘e8

17.♕e1! The crucial move to gain a tempo. White removes the queen o the d- le

339

340

Sample games

19.e6 Going for it! Carlsen could also have played the calm 19.f4 with a comfortable, slightly better game. 19…fxe6?! After this Black’s defensive task becomes very di cult in a practical game, as we will witness. The engine still evaluates 19…f5 as roughly equal. It is a tough call to allow the pawn to stay on the board, but White has no clear idea how to drum up play connected to it: 20.cxd5 cxd5 21.♗xg7 ♘xg7 22.♘f4 ♕d6 and it’s about equal. 20.♘f4 ♗xc3 21.♕xc3

weak themselves, they leave weak squares (like e5) open for White and they disturb the communication between both sides of the board. The black king is on the kingside and it is not so easy to get defenders closer to it as the pawns are in the way. 21…d4 22.♕d2 It is quite likely that Carlsen at this point already had seen the key ideas about to occur in the next moves. The queen aims for h6. 22…c5 23.♖ae1 ♘g7 24.g4

Not a di cult move, but a very aesthetically pleasing one. Black’s knight is on a dead end street. 24…♖c6?

The doubled pawns are problematic for Black in more than one way. They are 341

342

Sample games

The most resilient defence was probably 24…♖f8. Now the attacking plan seen in the game is no longer winning: 25.♘h3 ♕c6 26.♕h6 ♖xf3 27.♔g1 ♖xf1+ 28.♖xf1 ♕e4 and Black is holding. White is still somewhat better after 29.♕xh7+ ♔xh7 30.♘g5+ ♔g8 31.♘xe4 but it’s not over yet. It needs to be mentioned that 25.♘h3 is not mandatory, of course. Moves like 25.♘d3 or 25.♖e5 are promising as well, keeping the pressure up. 25.♘h3! Black’s king is in deep trouble now. It’s too late to save the game. 25…♘e8 26.♕h6 ♘f6 27.♘g5 White is going to pile up on the e6pawn next with ♖e5. At the same time, ♘xh7 becomes a serious idea as well. 27…d3 28.♖e5

The direct 30.♖xd3 was winning as well, but Black is entirely defenceless after the game move. White will win d3, next e6, and has an attack on the king on top. Anand resigned at this point, somewhat early for most people’s standards. He is, however, without any active ideas and Carlsen has easy winning moves, so it is quite understandable. In general Black is believed to be slightly worse against the Maroczy Bind. In this game Anand managed to play …d6-d5 due to tactics, but still it was not easy to handle. Two bad decisions were enough to allow Carlsen an attractive win. Game 13 FR 23.9 (2) Sergey Tiviakov 2601 Ferenc Berkes 2616 Plovdiv Ech-tt 2003 (2)

28…♔h8 After a non-move like 28…a5 White wins with 29.♘xh7 ♘xh7 30.♕xg6+ ♔h8 31.♖h5. 29.♖d1 Just picking up the lone d-pawn. 29…♕a6 30.a4 1-0

343

1.e4 c5 2.c3 Sergey Tiviakov is one of the leading experts on the c3-Sicilian and the French Tarrasch. His games are excellent study material for the IQP positions reached via our move order 2.♘f3 e6 3.c3 d5 4.exd5 exd5 5.d4. 2…e6 3.d4 d5 4.exd5 exd5 5.♘f3 ♘c6 6.♗b5 ♗d6 7.dxc5 ♗xc5 8.0-0 ♘ge7 9.♘bd2 This position is frequently reached via a completely di erent move order, starting with a French Tarrasch Variation. Have a look: 1.e4 e6 2.d4 d5 3.♘d2 c5 4.exd5 exd5 5.♘gf3 ♘c6

344

Sample games

6.♗b5 ♗d6 7.dxc5 ♗xc5 8.0-0 ♘ge7 9.c3 and we reach the current position. 9…0-0 10.♘b3 ♗d6 11.♗d3 h6 12.h3 ♘f5 13.♗c2 ♖e8

analysis diagram

14.♖e1 The theoretical section features 14.♕d3. I believe this poses more problems than Tiviakov’s move, but his choice is of course perfectly playable. 14…♖xe1+ This is the most obvious reply because the white queen looks misplaced on e1 at rst sight. We’ll see that this is not the case though. 14…♗e6 was the main alternative, but 15.♕d3 is already awkward to face. 15.♕xe1 ♗e6 16.♗e3 This is the positive point of the queen being on e1. It allows this development of the bishop to its best square, controlling d4 and c5. 16…♕f6 It was very principled to take the knight: 16…♘xe3 17.♕xe3.

345

In such a position, Black is very close to equality. There is just one weakness on d5 and Black is reasonably active. However there is also a practical side: this position is much easier to play for White, who has no weaknesses at all and will often get slightly better endgames, quite like in the actual game. Please note that Black’s bishop on e6 is a rather passive piece that does not have much of a perspective to improve. Let’s continue this line a bit: 17…♕f6 18.♖e1 ♖e8 19.♕d3 g6 20.♘bd4 ♘xd4 21.♕xd4 ♕xd4 22.♘xd4. Black should successfully defend this position, but White can try forever to play for little things. One idea is to regroup against the d5-pawn, for example with ♗b3, ♖d1 and routing the knight to e3 if possible. Black’s bishop pair is not important here as the pawn structure is quite static and the bishops don’t have any targets to attack. Now back to the game:

346

Sample games

17.♗c5 Trading o the defenders of the d4square is a classic anti-IQP strategy. One very common idea is to occupy the square in front of the IQP with a piece, most frequently a knight. 17…♘e5 As trades are not desirable with the isolated pawn, Black could have considered to preserve the bishop with 17…♗c7. White is ne after 18.♕e2 ♖e8 19.♖e1, keeping a nagging edge. 18.♘xe5 ♗xe5

19.♗xf5 ‘Trade the minor pieces against the IQP’ is a very well known guideline. Tiviakov is happy to employ this strategy. 19…♗xf5? This normal-looking move is a mistake, at least it turns the game into a strictly ‘two-results’ a air. White will press forever with zero risk. 19…♕xf5! was better, keeping the dark-squared bishop on the board. Now there is no straightforward way for White to improve: 20.♘d4 (this is not enough for anything tangible, however I’d like to show an idea worth knowing. More ambitious than 20.♘d4 is 20.♗d4, but after 20…♗c7! 21.♘c5 ♖e8 22.♘xe6 ♖xe6 23.♕d2 it is just equal as well). Now after 20…♗xd4 21.♗xd4 we get this position.

analysis diagram

It was worse to play 18…♕xe5 19.♗xd6 ♕xd6 20.♖d1 and White’s play against d5 is uncomfortable to face. 347

This kind of opposite-coloured bishops position can be quite favourable for White if you are able to quickly start a rook lift to g3, but here White’s queen is sitting on the key square: 21…♕g6 22.♔h2 a6 23.♕d2 ♖e8 24.♖e1 ♗d7 348

Sample games

and Black is quick enough to neutralize any ideas left. 20.♗d4!

The exclam is not so much for the strength of the move, but rather for its instructional value. Black now has to accept the eternal knight on d4 against the toothless light-squared bishop. White’s play from here is extremely comfortable. There is constant nagging pressure and zero danger of losing. 20…♖e8 21.♕e3

The simpler 21.♗xe5 ♖xe5 22.♕d2 was more precise, making sure you actually get the desired knight vs bishop scenario. 21…♔h7 349

This is too accommodating. 21…♗c2! was a lot tougher. Now 22.♗xe5 ♖xe5 23.♕xa7 ♗xb3 24.axb3 ♖e2 leads nowhere for White. He should play 22.♘d2, but this is not ideal obviously. 22.♗xe5 ♖xe5 23.♕f4 One of the most frequent reasons why attacks are successful are badly positioned queens that are too far o to help in the defence. Here 23.♕xa7?? gets heavily punished by 23…♗xh3! 24.gxh3 ♕g5+ 25.♔h1 ♕h4 and Black wins. 23…♖e4 24.♕f3 ♕e5 25.♘d4 ♗g6?! Black’s best way to defend here was to trade the major pieces and then try to put as many pawns as possible on dark squares (b6/a5/f6 etc.). White may still press, but it should not be convertible into a win. 25…♖e1+ 26.♖xe1 ♕xe1+ 27.♔h2 ♗g6 28.h4 is the line to check. Here White should try to keep queens on, still being an annoyance for Black.

26.♕d1! This is a high-class move! Tiviakov prevents further trades, thus being able to exert more pressure on the d5-pawn with the major pieces. Don’t get me 350

Sample games

wrong: White is not clearly better after this, but he retains more opportunities to manoeuvre and keep posing problems. 26…♕e7 27.♕d2 f6 28.♖d1 a6 29.a3 Black’s bishop has no targets at all to attack. Not right now, and most likely not in the future. 29…♕e8 30.f3 ♖e5 31.♔f2

A useful centralization of the king. In case of further trades it is closer to the centre and the queenside. 31…♕e7 32.♘e2 ♕e6 33.♘d4 33.♘f4 ♕b6+ 34.♔f1 ♕b5+ 35.♔g1 ♗f7 was possible as well, keeping the slight but permanent edge. 33…♕b6 Black is not playing for a win here, of course. He could have just returned to e7 to repeat the position, but White for sure wouldn’t have continued to go along. Berkes probably just felt the queen was better on b6 than on e7. 34.♖e1 Not strictly neccessary at this moment, but the rook trade is just ne for White. 34…a5 After 34…♖xe1 35.♔xe1, doing a move-by-move analysis is not very 351

practical. White will combine play against d5 with expanding on the kingside, keeping the pressure up.

35.a4 I was surprised by this move. You would expect that White avoids pawns on light squares, being a possible target for Black’s bishop. A closer inspection however shows that Black has no way to attack the pawns for real, and in fact White has now prepared a later b2-b4, opening the position. 35…♔h8 36.b3 ♔h7 37.♖e2 ♕c5 38.♖e3 ♕a3 39.♕e2 ♕d6 40.♕e1 ♖g5 41.♔g1 ♖e5 42.♕d2 ♕e7 43.♔f2 ♕c5 44.♘b5 b6 45.♘d4 ♕d6 46.♘e2 ♕c5 Black has defended excellently so far. White now tries for a di erent approach. 47.♘f4

352

Sample games

47…♖xe3 Let’s check the alternatives: A) 47…♗f7 was di cult to assess during the game. Black allows the trades that actually help White to improve his king. It seems that after 48.♘d3 ♕xe3+ 49.♕xe3 ♖xe3 50.♔xe3 ♔g6 (with the idea to go to d6, defending the pawn and covering potential entry squares on c5 and e5) 51.♔d4 ♔f5 52.b4 ♔e6 Black should hold; B) There also is 47…h5, an engine suggestion that probably no human would play. The following line is the justi cation of the move: 48.b4 ♕c6! 49.♖xe5 fxe5 50.♘xd5 ♗f7 51.♕c2+ ♗g6 52.♘e7 ♗xc2 53.♘xc6 ♗xa4 54.♘xe5 ♗b3 55.♔e3 a4 56.♔d2 a3 57.♔c1 ♔h6 58.f4 g5 and Black indeed holds – giving just an impression by quoting one line of many possible ones. 48.♕xe3 ♗c2 49.♕xc5 bxc5 50.♘e6

Probably the best try. Instead, 50.♘xd5 ♗xb3 51.♘b6 was no improvement. The pawn on a4 is too weak. 50…c4? This is the only clear mistake in the whole game, and it is fatal, at least by human standards. 50…♗xb3 was the way to draw: 51.♘xc5 ♗c4 52.♘b7 ♗b3 53.♘xa5 ♗xa4 54.♔e3 ♔g6 55.♘b7 ♔f7 56.♔d4 ♗b3. There is no progress to be made from here. Black will defend his only weakness. 51.b4! axb4 52.cxb4 ♗xa4

53.♘d4! Very nicely played. The b-pawn is unstoppable. 353

354

Sample games

53…♗d7 54.b5 ♔g6 55.b6 ♗c8 56.♘b5

while avoiding clear draws. Defending this with black is no fun at all, so keep on searching for new problems you might pose. Game 14 SI 46.6 (B22) Michele Godena 2508 Samvel Ter Sahakyan 2578 Plovdiv ch-EUR 2012 (10)

56…♗a6? The question mark has to be there because this is objectively a mistake, but the save is absolutely amazing and inhuman to nd: 56…f5! 57.♘d6 c3 58.♘xc8 c2 59.b7 c1♕ 60.b8♕ ♔h7! is still putting up resistance and probably holds. The d-pawn is enormously strong! 61.♔g3 d4 62.♔h2 ♕c5 63.♘b6 d3 64.♘d7 ♕d5 65.♘f8+ is the engine’s main line and leads to a draw! 57.♔e3 ♔f5 58.f4! Now Black is lost against good technical play, which Tiviakov managed to demonstrate. 58…h5 59.♘d6+ ♔e6 60.b7 ♗xb7 61.♘xb7 h4 62.♘a5 ♔d6 63.♔d4 g6 64.f5 gxf5 65.♘b7+ ♔c6 66.♘c5 ♔b5 67.♘e6 ♔b4 68.♘f4 ♔b3 69.♘xd5 f4 70.♘xf4 c3 71.♘e2 c2 72.♔d3 ♔b2 73.♔e4 1-0 A real grind by Tiviakov. White had a very comfortable game, just playing for two results. At the end Black just made one serious mistake and it cost him the game. One key virtue in the anti-IQP positions is patience. You need to keep on pressing, playing for little things 355

1.e4 c5 2.c3 Italian GM Godena is an expert on the Alapin. He also plays some other interesting low-maintenance lines like 1.e4 c5 2.♘f3 d6 3.♗c4, an even simpler move than our choice 3.♗b5+. 2…d5 3.exd5 ♕xd5 4.d4 ♘f6 5.♘f3 e6 6.♘a3 ♕d8 7.♘c2 ♗e7 8.♗d3 0-0

Black has quite a wide choice of move orders here. One upside for White in this line is that you have a standard set-up, almost regardless of what Black is doing. You place the queen on e2, the bishop to f4 or g5, and usually castle kingside and centralize the rooks – the most natural placement you can imagine. 356

Sample games

9.♕e2 b6 10.♗f4 White is still not committing to castling to either side. 10…♗b7 11.dxc5 The white d-pawn will very likely be exchanged against Black’s c-pawn anyway, so White goes ahead and gives Black the choice how to recapture. 11…bxc5

A surprising decision, at least in my eyes. There are positions where the black formation with a-, c- and e-pawn are ne, because the open les provide counterplay. Here, however, White will have an excellent knight on c4 that covers b2 and Black’s inferior structure is not easily repaired. I would have expected 11…♗xc5 12.♖d1 ♘bd7 13.0-0 and I’d slightly prefer White, but Black is solid. 12.♖d1 ♕a5 The black queen will be o side in some moves, but this still looks like the most natural move. Instead, 12…♘bd7 13.0-0 ♕c8 looks quite passive, and White has the simple plan to shift pieces to the kingside, for example starting with 14.♘e3. 13.♘a3 ♘bd7 14.0-0 357

Black now has to worry about ♘c4, but at the same time he does not want to move the d7-knight as it guards the e5square as well. 14…♘b6 Black decides to concede the e5-square. Against 14…♗d5 the engine points out the excellent resource 15.♘d2!. Now, grabbing on a2 leads to serious problems: 15…♗xa2 16.♘ac4 ♗xc4 17.♘xc4 ♕d8 18.♘e5 and White is much better. The threat is to win material on the d- le. In fact Black only has one way to avoid an immediate loss: 18…♘d5 19.♘xd7 ♕xd7 20.♗g3 ♕a4 21.♗e4 and Black su ers. 15.♘e5 ♘bd5 16.♗g3 ♖ac8 17.♘ac4 White’s opening concept was successful. He’s got very active pieces and excellent posts on c4 and e5 for his knights. 17…♕a4?! Black should have played 17…♕d8. It is not a happy move to play, but Black needs defenders near his king.

18.♖fe1 Getting the rook centralized is very natural, but there was an even better 358

Sample games

alternative in 18.f4!. White’s play in this opening is mostly based on piece play, so pawn pushes like that are not very common. Here, f2-f4 is very strong, as f4-f5 would open up the kingside for the attack while the black queen is not well placed at all. 18…g6 stops f4-f5, but 19.♗h4 allows g2-g4 next for a strong attack. 18…♗a8? This has a tactical problem. After a move like 18…♖cd8 White is better, but it is nothing decisive. One risk-free option (probably not the best one though) is to play 19.♕c2, when Black can’t avoid the trade: 19…♕xc2 20.♗xc2. Now plans with ♗a4 and ♘a5 are very annoying for Black to meet.

19.♗c2? White certainly calculated 19.♘xf7!, but must have missed something. In fact this knight sacri ce is close to winning for White: 19…♖xf7 20.♕xe6 (one key point is the unprotected rook on c8. Black’s last move was 18…♗a8, undefending this piece) 20…♖cf8 (20…♕d7 21.♗f5! is a complete 359

disaster) 21.♘e5 and White will get more than enough material just by taking on f7, but the main problem is Black’s uncoordinated piece placement. 21…♘b6 22.b3 ♕e8 23.♘xf7 ♖xf7 24.♕h3.

analysis diagram

Black is helpless against ♗d6 and ♗f5 ideas. 19…♕a6 20.♗d3 ♕b7 It looks like Black has spotted the ♘xf7 idea. He prevents it and even threatens himself now. 21…♘xc3 is on. 21.♕f3 ♕a6 22.♕e2 ♕a4? Black allows the strike again. 23.♗c2 No, thank you. 23…♕e8 Maybe Black needed a win in this game? He avoids a repetition but misplaces the queen in the process. 24.♗d3 So White allows Black to repeat again. Maybe he was thinking that Black will avoid it anyway, given how he played the last moves? In any case, White could have avoided this possibility

360

Sample games

with many possible moves – he is still comfortably better. For example, 24.♗b3 was a way to continue: 24…♖d8 25.♘a5, intending 26.♕a6. This is tough to stop, as 25… ♖d6 26.♘xf7 ♕xf7 27.♗xd6 ♗xd6 28.♕xe6 is better for White as well. 24…g6? Black wants/needs to win, it seems. He could have claimed a draw with 24…♕a4, allowing 25.♘xf7 again – but White wouldn’t get the chance to play it after the claim. Objectively speaking he should be happy with the draw. 25.♘e3 ♘c7 Here 25…♘xe3 was neccessary, but no fun either: 26.♗b5 ♗c6 27.♗xc6 ♖xc6 28.fxe3 ♖c8 29.e4 and the pressure is mounting.

26.h3! This is very strong, thinking about prophylaxis in the middle of a brewing attack. 26…♘h5

361

27.♗h2 White now threatens 28.♘d7!. 27…f6 Going 27…♘f6 wasn’t any better. After 28.♗f4 White has 29.♗h6 as the next threat. Black is on the ropes. 28.♘5c4 28.♘d7 ♖f7 29.♗c4 was winning quickly, but Godena’s move is strong as well. The engine con rms that the slower attack is also decisive. 28…e5 29.♘g4 ♕c6 30.f3 ♖fd8

31.♘cxe5! And nally the oodgates are opened. 31…fxe5 32.♗c4+ ♘d5 33.♘xe5 ♕e6

362

Sample games

34.♘xg6! ♕xe2 If 34…♕xg6 35.♕xe7 ♘hf6 36.♕xa7 it also falls apart. 35.♘xe7+ ♕xe7 36.♖xe7 ♘hf6 37.♖xa7 ♔f8 38.♗e5 The bishops are completely dominating. 38…♖d7 39.♖xd7 ♘xd7 40.♗d6+ ♘e7 41.♖e1 ♖e8 42.♗b5 1-0 In the …e7-e6 Alapin White has a simple, yet e ective set-up, as demonstrated in this game. All his pieces are well placed, and his superior activity might become the starting point of an attack on the black king. Game 15 SI 47.3 (B40) Vasil Spasov 2559 Jakub Szostko 2160 Plovdiv 2015 (2)

12.♘d2 12.c4 is my recommendation in the theoretical section. Spasov’s move is more popular, but I liked the idea to place the knight on c3. Objectively speaking these lines are of about equal value. 12…♗d7 13.♕c4 13.♗xd5? exd5 14.♕xd5 ♗c6 gives Black great play for the pawn. 13…♗c6 13…♕xc4?! 14.♘xc4 ♗c7 15.♘e3 is good for White. His bishops are putting too much pressure on Black’s queenside. 14.♘e4 White had a good alternative with 14.♕e2 ♘f6 15.♗xc6 ♕xc6 16.♘f3 ♖fe8 17.♘d4 ♕d7 18.♖d1 a6 19.♗g5 and a small plus for White in PröhlOsmanodja, Germany 2014. 14…♗e7 15.♖d1 ♖fd8 16.♕e2

1.e4 c5 2.♘f3 e6 3.c3 ♘f6 4.e5 ♘d5 5.g3 ♘c6 6.♗g2 d6 7.exd6 ♗xd6 8.0-0 0-0 9.d4 cxd4 10.♘xd4 ♘xd4 11.♕xd4 ♕c7

363

364

Sample games

White has regrouped to allow c3-c4, starting to push the queenside majority. 16…h6 I think Black should prevent White’s idea and play 16…b5. It’s not clear what he feared now. The game move suggests that maybe 17.♘g5 was seen as an issue, but 17…♕b7 prevents sacri ces and is ne for Black. Instead of 17.♘g5 I would suggest 17.♗g5 for White, with about equal chances. 17.c4 ♘f6 18.♗f4

It puts a pawn on a dark square and weakens the light squares, as we will see later in the ending. The current position still looks completely harmless, but small things may easily pile up for Black. This game is a good example for drifting into trouble slowly. The alternative was 18…♕a5. I suggest 19.a3 for White now. Instead 19.♗d6 ♗xd6 20.♘xd6 ♗xg2 21.♔xg2 ♕b6 22.♘b5 ♕c6+ is just too equal. In such a non-forced position it is tough to analyse move-by-move. Let’s have a look at 19…♕a6, allowing to take on c6 with the queen if needed: 20.b4 ♕a4! (20…♘xe4 21.♗xe4 ♗xe4 22.♕xe4 b6 23.♗c7 is still annoying for Black) 21.♘c3 ♖xd1+ 22.♕xd1 ♕xd1+ 23.♖xd1 ♗xg2 24.♔xg2 a5 and it’s equal. 19.♘xf6+ ♗xf6 20.♗e3 ♗xg2 21.♔xg2 ♕c6+ 22.♔g1 This was a very natural sequence of moves after 18…e5.

18…e5 This is not bad objectively, but it’s still a move that I would have liked to avoid.

22…a5 Again giving White a little something by weakening the light squares, in

365

366

Sample games

particular b5. The obvious alternative was to play 22…a6. I’d still take White because the plan of advancing the queenside is easy to play while Black has no clear active idea. Of course, it is nothing serious yet, but if anybody, Black needs to be alert. 23.b3

23…e4? The question mark is maybe a bit harsh, but after this move it gets uncomfortable for Black. The e-pawn is just weak on e4 while the long diagonal for the f6-bishop is rather worthless (the staring into thin air phenomenon). 24.♖xd8+ ♖xd8 25.♖d1 The vulnerability of the three splintered pawns starts to be felt. The queen cannot cover all of them. 25…♖d6 Instead 25…♖xd1+ 26.♕xd1 leads to a position very similar to 26…♖d7 below. 26.h4

367

26…♕d7? This leads to a lost position. Black had to wait with a move like 26… ♖d7. The coming sequence is quite logical, I think: 27.♕c2 ♖xd1+ 28.♕xd1 a4 (at least getting rid of one weakness) 29.♔g2 axb3 30.axb3. White has nagging pressure in this endgame. b7 and e4 are weak and need to be covered. At the same time White has the easy plan to advance the queenside. Note that after a queen trade the white king is closer to the centre/queenside as well. It is an interesting exercise to play out such a position against a computer with both sides. You will see that with white you can press with a draw in hand, while with black you will lose every single time. 27.♖xd6 ♕xd6 28.♕g4 Too many loose pawns! There is e4, b7 and even h6 hanging. 28…♕c6

368

Sample games

Instead, 28…♕e6 saves the pawn for the moment, but not for a long time. 29.♕xe6 fxe6 30.♔f1 ♔f7 31.♔e2 h5 32.♗b6 ♗c3 33.♔e3 wins easily. 29.♗xh6 ♗d4 30.♗f4 ♕c5 31.♕e2 f5 32.♗e3 ♗xe3 33.♕xe3

That’s the point of taking on a4. There is just no good reply available for Black now. 38…♔h6 38…f4 was a better defence and a good try to play for a trap: 39.♕f5+ (the pitfall to avoid is 39.♕xb7?? ♕c2! and Black draws all of a sudden, based on the threat …e4-e3) 39…♔g8 40.♕xf4 ♕xc4 41.♕b8+ ♔h7 42.♕xb7 e3 43.♕b1+ ♔h8 44.fxe3 ♕e2+ 45.♔h3 ♕xe3 46.♕f5 and White is winning, but there is still some work to do. 39.♕xf5 39.♕e6+ was easier, but Spasov’s move wins as well. 39…♕xc4 40.g4

White is completely winning here. In addition to the extra pawn there are multiple weaknesses to exploit, in particular around Black’s king. 33…♕d6 34.♕f4 ♕d7 35.♔g2 a4 36.bxa4 This looks a bit odd, but it works based on concrete lines. 36.♕e5 was the lazy man’s move, keeping everything under control. 36…♕xa4 37.♕b8+ ♔h7 38.♕c8

40…g6 41.♕f8+ ♔h7 42.♕e7+ ♔g8 43.♕xb7 e3 44.♕b8+ ♔h7 45.♕a7+ ♔g8 46.♕xe3 From this point there are some moves that are tough to explain, even if we assume that there was heavy time pressure involved. 46…♕xa2? Black should have gone 46…♕xg4+ 47.♕g3 ♕e4+ 48.♔h2 ♔h7. Here at least White has to worry about a

369

370

Introduction

perpetual, while the white king is perfectly safe in the game. White should win anyway, but still it is a strange choice. 47.♕e8+ ♔g7 48.♕e5+ ♔h7 49.h5 gxh5 50.♕xh5+ ♔g7 51.♕e5+ ♔h7 52.♔g3 ♕b3+ 53.♔h4 ♕d1 54.♔h5 I wonder if the coming moves were correctly entered in the database, or if there is some mistake. This already looks a bit odd, as 54.♔g5 is a lot more natural to just avoid a check on h1. 54…♕d2 55.f4??

55…♕d1?? 55…♕h2+ 56.♔g5 ♕h6+ 57.♔f5 ♕g6 and mate! It is certainly possible that both players missed this if there was heavy time pressure involved, but my bet is there is something wrong with the entry into the database. 56.♕e7+ ♔g8 57.f5 ♕d4 58.♔g5 1-0 This game is a good example how easy it is to drift into a worse position if you don’t have a very clear plan available to do something active. Black was certainly ne after the opening, but some imprecise pawn moves got him into grave trouble. 371

PART III

Caro-Kann Defence: 1.e4 c6 Introduction

The Caro-Kann Defence is a very well respected choice for Black. For many years it had a very solid, but also somewhat dull reputation. This has changed a lot in recent years, as white players have recognized that the safe options don’t promise much – so they have sharpened up the game. The main weapon of the top GMs is the Advance Variation, but this is not an easy choice for us as it requires intensive study. I decided to go for the Two Knights Variation. This line sets some pitfalls for Black and the best lines are not all that well known. 2.♘c3 d5 3.♘f3 This is the starting position of the Two Knights Variation. It used to be a favourite of Bobby Fischer, but it fell out of fashion for a long time.

372

Introduction

The most natural move probably, reaching the typical Caro-Kann pawn structure right away. This option is quite popular on the club level and is discussed in Chapter 18. Chapter 19: 3…♗g4

In recent years there has been renewed interest in the line and many strong players use this, at least occasionally. Nowadays, top players play a lot more games than decades ago and everybody needs backup lines or systems that are not applied in every single game. The Two Knights is a good line because it is relatively simple to learn and still contains some poison. The Caro-Kann is examined in four chapters. Chapter 17 is the ‘odds and ends’ section that deals with move orders and rare black choices. However, Black has three main answers to the Two Knights that are of about equal value and popular on all levels. Chapter 18: 3…dxe4

373

This has been the main line for ages, and was played almost exclusively at the times of Fischer and the BotvinnikTal matches. It has retained its solid reputation, but has lost some ground recently to the alternatives, in particular 3…♘f6. The 3…♗g4 line is analysed in Chapter 19. Chapter 20: 3…♘f6

374

Chapter 17: 2.♘c3: sidelines

This is the move of recent years. It o ers Black a bit more excitement compared to the other options, but White has no reason to complain about the complex positions either. We will take a closer look in Chapter 20.

CHAPTER 17

2.♘c3: sidelines 1.e4 c6 2.♘c3

Instead, 2.♘f3 will lead to the same thing almost all of the time. However Black may omit …d7-d5 and go for a … g7-g6/…d7-d6 based set-up when we have forfeited the option of pushing f2f4. Please see the chapters on the Pirc and Modern Defence for details. 2…d5 1) 2…e5 is very rarely seen, for good reason: 3.d4 d6 4.f4 (instead, 4.♘f3 would transpose to a Philidor that is quite reasonable for Black) 4…♘d7 5.♘f3 and White is just better with his big centre; 2) 2…g6 3.d4 d5 4.♘f3 we examine via 2…d5 3.♘f3 g6 4.d4 below. 3.♘f3

375

376

Chapter 17: 2.♘c3: sidelines

A) 5…dxe4 The conservative choice, steering the structure towards typical Caro-Kann play. 6.♘xe4

3…g6 This move is not very frequently seen here, but this set-up, involving …c7c6/…d7-d5 and the anchetto, may arise from 1…g6 as well – this is indeed the more common move order. 4.d4 ♗g7

5.h3 This is an important move as otherwise Black might simply play 5… ♗g4 to increase pressure on the d4pawn. Note that, via the 1…g6 move order, we also play h2-h3 early on, so this ts well. Now Black has no natural move for the c8-bishop, so he either takes on e4 or develops the king’s knight. Those knight moves lead to quite sharp play. 377

A1) 6…♘f6 7.♘xf6+ exf6 7…♗xf6 8.c3 0-0 9.♗h6 ♖e8 10.♗c4 is a more attractive version of the 6… ♘d7 line, where Black prepares to recapture with the knight from d7. White is slightly better with extra space. 8.♗d3 This is slightly better for White and very comparable to 3…dxe4 4 ♘xe4 ♘f6 5 ♘xf6 exf6. It is a slightly better version as the bishop is less attractive on g7 than on d6. A2) 6…♘d7 7.♗c4 ♘gf6 8.♘xf6+ ♘xf6 9.0-0 0-0 10.♖e1

378

Chapter 17: 2.♘c3: sidelines

6…f6 The principled move, but, objectively, taking on e4 is probably better. This leads to a position similar to 5…dxe4, but with the knight already on h6. 7.♕d2 ♘f7 8.0-0-0 0-0 9.♔b1

A very natural sequence of moves up to this point. White is slightly better as he controls more space. The outpost e5 is an important factor as well. We will continue with centralization like c2-c3, ♗f4, ♕e2 and ♖ad1. This is always useful to play. I think White is better here. He’s got the simple h4-h5 available but also the subtle ♗g3, which reliably prevents … e6-e5.

B) 5…♘h6

C) 5…♘f6

This looks a bit strange, but it involves an ambitious plan. Black intends to play …f7-f6, …♘h6-f7, and ultimately …e7-e5 to strike in the centre. White needs to be precise here. 6.♗f4 I like this very straightforward approach. The ♗f4/♕d2 set-up forces the knight back and prepares queenside castling very quickly. 379

6.e5 This leads to favourable complications, but you need to check the coming moves carefully. I would have liked a 380

Chapter 17: 2.♘c3: sidelines

simpler solution, given that this is a fairly rare line from Black’s perspective. Unfortunately, there is none, as the alternative 6.♗d3 allows easy equality after 6 …dxe4 7.♘xe4 ♘xe4 8.♗xe4 00 9.0-0 c5! – so we have to learn these forcing moves. 6…♘e4 7.♘xe4 This is the strongest. Instead, 7.♗d3 leads to complicated play as well, and might be a worthy alternative if you forget your preparation. 7…dxe4 8.♘g5 c5 Forced, otherwise White will grab e4 for nothing. 9.dxc5 The big alternative is 9.♗c4, of course, but it did not convince me. After capturing on c5 White might just keep the pawn, if Black is not careful. 9…♕c7 The most testing move, and the only one where further knowledge is useful. 10.♗f4

10…0-0 10…♗xe5? is a tactical mistake that might easily happen, as the coming motif is not that common: 11.♗xe5 381

♕xe5 12.♕d8+! ♔xd8 13.♘xf7+ ♔c7 14.♘xe5 with a huge advantage for White. 11.♕d4 ♘d7 After 11…♘c6 12.♕e3 ♗xe5 13.♗xe5 ♕xe5 14.0-0-0 White is better as e4 will fall, or Black needs to make serious concessions, like playing …f7-f5. 12.♘xe4 ♘xe5 13.♕e3 We have kept the extra pawn and I don’t see enough compensation for Black. Moves like ♗e2, 0-0 and maybe c2-c3 will follow.

Conclusion 3…g6 is the only move that requires some study in this chapter. In fact I was surprised that the road to an advantage is rather narrow. I think the lines provided here give White a good game, but the play is still quite tricky. The main line is rather long and contains some crucial decisions that are not 100% intuitive. So make sure you memorize them – it will pay o if you get it on the board!

382

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

CHAPTER 18

Two Knights: 2.♘c3 d5 3.♘f3 dxe4 1.e4 c6 2.♘c3 d5 3.♘f3 dxe4

Threatening 8.♘xf7. Black needs to react to this very precisely. 7…♘d5 Amazingly, the terrible 7…♘b6? has been played dozens of times. After 8.♘e5 the pawn on f7 falls! 8.d4 h6 9.♘e4 ♗e7 10.0-0 Like in the Classical main line with 2.d4 d5 3.♘c3/d2 dxe4 Black captures on e4. This move is ne for Black, if he is aware that there are some crucial di erences between our position at hand and the main-line Classical. This mostly concerns 4…♗f5, as we will examine now. 4.♘xe4 A) 4…♘d7 Very solid. Black prepares 5…♘gf6, avoiding the doubling of the f-pawn as seen after 4…♘f6. 5.♗c4 This is the most challenging move. 5…♘gf6 6.♘eg5 e6 7.♕e2

383

White enjoys a small advantage here. We have easier development and more space. Also note that h6 might be a weakness on the black kingside, and White has ideas like ♘e4-g3-h5. B) 4…♘f6

384

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

This line has enjoyed some recent popularity, mostly due to the revaluation of the traditional main line, 3…♗g4, and the resultant search for alternatives. 5.♕e2!? This is the move currently preferred by the top players. It looks a bit strange, but it has some interesting points, as we will see. That being said, just taking on f6 with the knight is also ne, of course. B1) 5…♗f5 This is the ghting move, encouraging White to double the f-pawn. 6.♘xf6+ A natural move, but 6.d3, continuing with g2-g3 and maybe even b2-b3 later, is a good alternative, aiming to get a quiet game without pawn structure imbalances. 6…gxf6 7.d3 One point of White’s set-up. Going d2d3 instead of d2-d4 blunts the bishop on f5. 7…♘d7 8.g3 This not only prepares to develop, but also supports a possible ♘h4 followed by f2-f4 to harass the bishop. 385

8…♘e5 A very concrete solution that was played by Carlsen and Anand – and they know what they are doing ☺. Instead, 8…e6 is met by 9.♘h4 ♗g6 10.♗g2. This line is as yet untested in practice. I think White’s position is slightly easier to play here. Let’s look at a sample line: 10…♗e7 11.0-0 f5 12.♘xg6 (instead, 12.♘f3 is also playable, claiming that it was worth investing two tempi to provoke 11…f5) 12…hxg6 13.d4 with the idea to play c2-c4 and d4-d5, opening the position for the bishops. 9.♘xe5 ♕a5+ 10.♗d2 ♕xe5 11.♗c3 Instead, 11.♕xe5 has been played by Maxime Vachier-Lagrave, and this is also ne. Black still has the worse pawn structure. 11…♕xe2+ 12.♗xe2

In this endgame position White is slightly more comfortable. Note that … e7-e6 is tough to play and …♗g7 – and then …e7-e6 to protect f6 – looks passive. If Black goes for …e7-e5, a White f2-f4 will be annoying. B2) 5…♗g4 386

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

Similar to 5…♗f5, Black allows the pawn structure to be weakened. 6.h3 ♗h5 The simplistic 6…♗xf3 7.♕xf3 gives White the standard advantage that we often have in the Two Knights. You have the bishops and Black is solid, but without any clear and active idea. We play c2-c3/d2-d4/♗d3 as usual. 7.♘xf6+ gxf6 8.d3 Anticipating that the bishop will go to g6 later. On g6 it will now look at this solidly protected pawn. 8…e6 9.♗d2 ♘a6 10.g4 ♗g6 11.0-0-0

With an interesting position. White has ideas of ♘h4, maybe f4-f5 later to play against the bishop on g6. B3) 5…♘bd7?? 6.♘d6# I’ve had this multiple times in blitz, so it does actually appear on the board!

387

B4) 5…♘xe4 The most conservative move. Black keeps his pawn structure sound and simpli es a bit. 6.♕xe4

B41) 6…♕d5 Well, with clear intentions to trade the queen. We should not trade on d5 in such a situation as this would improve Black’s central control, with a pawn coming to d5. 7.♕f4 ♕f5 8.♕e3 ♕e6 Very consistent play by Black. Now we can no longer avoid the trade, but we are now able to allow it on our terms. I feel that 8…♕xc2?! is too risky: 9.♗d3 ♕a4 10.b3 ♕a5 11.♗b2 and 388

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

White has excellent development for the pawn. 9.b3 ♕xe3+ Black will have to trade anyway, soon. Once White has castled we threaten to move the queen away and make the queen on e6 look silly. 10.fxe3

8.♘e5! This zwischenzug forces a concession from Black. 8…e6 9.♕e2 White is slightly better here with more space and a nicely placed knight on e5. C) 4…♗f5?! This position looks harmless at rst, but a closer look reveals that Black is not well developed, and White has an extra centre pawn. It is not much, but White is a bit more comfortable to play. Sample game 16, Mamedov-Paichadze, Nakhchivan 2014, shows how the game develops from here. B42) 6…♘d7 7.♗c4 An important nuance in move order, as we will see. 7…♘f6

389

Yes, this natural move is somewhat dubious – still, it is played very frequently as most Caro-Kann players have 1.e4 c6 2.d4 d5 3.♘c3 dxe4 4.♘xe4 ♗f5 in their repertoires, so going …♗f5 here as well looks very useful. Let’s check why 4…♗f5 is questionable here. 5.♘g3 ♗g6?! 390

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

In fact, 5…♗g4 is the best move, limiting the damage considerably, for example: 6.h3 ♗xf3 7.♕xf3 ♘f6 8.c3 e6 9.d4 with a typical slight advantage for White due to his space surplus and bishop pair. This is very comparable to the 3…♗g4 line, but here Black has lost some time with …♗f5-g4. 6.h4! h6 7.♘e5 ♗h7? After this the game is basically over. 7…♕d6 is relatively best, but still very nice for White: 8.d4 ♘d7 9.♘xg6 ♕xg6 10.♗d3 ♕d6 11.c3 and White enjoys a typically comfortable edge due to his extra space and bishop pair.

8.♕h5 g6 9.♕f3 Instead, the attractive 9.♗c4 is also very favourable for White. We are spoilt for choice at this point! 9…♘f6 10.♕b3 This is White’s point: a deadly double attack. 10…♕d5 11.♕xb7 ♕xe5+ 12.♗e2 And Black has no way to save his rook on a8. However, there is one line where knowing an important point is necessary. 12…e6 391

13.a4! Not intuitive at all, but it is very strong. If White grabs the rook immediately, with 13.♕xa8, Black has 13…♕c7! and it will be very tricky to get the queen out. 13.a4! actually prepares for this scenario by intending to play the pawn up to a6 and then go ♕b7. In addition, now Black has problems nding a move without serious drawbacks. 1) For example, 13…♗e7 does not prepare …0-0, as e7 would hang; 2) 13…♗d6 simply allows 14.♕xa8 without any problems. 3) 13…♗c5 also leads to a lost position, but it is at least somewhat messy: 14.c3 0-0 15.d4 ♗xd4 16.cxd4 ♕a5+ 17.♗d2 ♕b6 18.♕xb6 axb6 19.♗xh6 and White is winning. Note how bad the h7-bishop is. Conclusion Taking on e4 is a ne line for Black, but he needs to understand that this is not the same as the main line with 1.e4 c6 2.d4 d5 3.♘c3. If he fails to do so, he will be in trouble. Black’s best line is probably 4…♘f6, but after the fresh

392

Chapter 19: Two Knights with 3…♗g4

5.♕e2 there are many interesting ideas to try.

CHAPTER 19

Two Knights with 3…♗g4 1.e4 c6 2.♘c3 d5 3.♘f3 ♗g4

This is the traditional main line. Nowadays 3…♘f6 is a more fashionable move as it o ers Black more chances for a ght, while 3…♗g4 is very solid, but not very active by comparison. With 3….♗g4 Black usually signals his intention to take on f3 and give up the pair of bishops. Retreating the bishop is possible, but has a dicey reputation, as we will see later. 4.h3 Recently, the alternative 4.d4 has been played as well. This is also a good option for varying your play. The main di erence between this, and our choice of 4.h3, is that we force a decision immediately about whether Black wants to keep his bishop (risky) or stay solid, taking on f3 but giving up the bishop pair. 393

394

Chapter 19: Two Knights with 3…♗g4

A) 4…♗h5 This is actually a somewhat dubious line, but it is still played quite frequently as it leads to di cult positions in which White may go wrong. If White knows what to do Black will be the one in trouble. 5.exd5 This initiates a forced sequence aimed at playing against the soon-to-be o side bishop on g6. 5…cxd5 6.♗b5+

6…♘c6 6…♘d7?? loses on the spot: 7.g4 ♗g6 8.♘e5 ♘gf6 9.♘xd5! ♗e4 10.♘xf6+ exf6 11.♘xd7 ♗xh1 12.♘e5+ ♔e7 13.♕e2 and White is completely winning, threatening multiple discoveries with the knight on e5. 7.g4 ♗g6 8.♘e5

395

8…♖c8 This is almost forced, as 8…♕d6?! already puts Black in danger of su ering a drastic loss. After 9.d4 we are threatening 10.♗f4, so a slow move won’t help: 1) 9…a6? 10.♗xc6+ bxc6 11.♗f4 and White is not threatening to move the knight immediately, but ♕d2 would be very awkward to meet next move. White also threatens h4-h5, winning the bishop. Black is in dire straits already: 11…♕b4 12.♕d2 ♕xb2 13.0-0 ♕xc2 14.♕e3 and White has a winning attack; 2) 9…f6 10.♘xg6 hxg6 11.♕d3 0-0-0 (11…♔f7

396

Chapter 19: Two Knights with 3…♗g4

12.♘xd5! – a cute point to remember) 12.♗xc6 ♕xc6 13.♕xg6 e5 14.♗d2. White has won the g6-pawn for little compensation and is better. 9.d4 e6 The most logical and best move. After 9…f6?! 10.♘xg6 hxg6 11.♕d3 ♔f7 12.♘xd5! we get the same tactic that we have seen in other lines. White nets a pawn for nothing.

10.♕e2! White has an important choice here. We may play 10.♕e2, as I recommend, or go for 10.h4. The move 10.h4 after 10…f6! 11.♘xg6 hxg6 12.♕d3 ♔f7 13.♗d2 leads to a complicated middlegame where Black’s chances are not worse. It is interesting, though, so if you dislike my long main line that leads to an edge for White, this is worth checking out. I recommend 10.♕e2!, because it is objectively better, but the arising positions are maybe not to everybody’s taste, so mentioning an alternative seems like a good idea. Let’s check out what I am talking about. 10…♗b4 397

Best; alternatives are unconvincing: 1) 10…♗d6 11.♘xg6 hxg6 12.♘xd5 (I don’t see any serious compensation for the pawn) 12…♗b8 13.♘b4 ♕xd4 14.c3 ♕f6 15.♗e3 and White enjoys a healthy advantage; 2) 10…♘f6? 11.h4 and White wins already as the bishop on g6 is trapped – this can only be stopped by …f7-f6, which highlights why the queen is on e2. 11…h6 12.♘xg6 fxg6 13.♕xe6+ – a bad day at the o ce for Black. 11.h4 ♘ge7 Again, this is forced. White is able to refute 11…♕a5 in more than one way, but there is one line that has a great nishing shot: 12.0-0! ♗xc3 13.♗xc6+ bxc6 14.bxc3 ♕xc3 15.♖b1 ♗xc2 16.♖b7 ♘e7 17.♗g5 f6 18.♘g6!!. What a shot! 12.h5 ♗e4 13.f3 So this bishop is ultimately lost. Things are still very tricky, though. 13…0-0 14.♘xc6 Please avoid 14.fxe4 ♘xd4!. 14…♘xc6 After 14…bxc6 15.♗a6 Black loses material for no compensation. 15.♗e3

398

Chapter 19: Two Knights with 3…♗g4

Again, necessary to cover d4. Now the ♗e4 is hanging ‘for real’. 15…♕f6 This move attacks d4 and f3, so the following line is completely forced. Black has alternatives to 15…♕f6 though: 1) 15…♕a5? 16.0-0 and Black does not get enough pawns for the bishop about to be lost; 2) 15…e5!? ultimately leads to a white advantage as well, but it is tricky: 16.♗xc6 ♖xc6 17.fxe4 ♖xc3 18.0-0-0! (the key idea that is tough to anticipate. White does not win any material, but is better in this oppositecastling scenario) 18…♖c4 19.exd5 ♕xd5 20.dxe5. It is a complicated position, but White’s attacking chances are better here. 16.fxe4 Black will now get some pawns for the piece. 16…♘xd4 17.♗xd4 ♕xd4 18.♖d1 All pretty much forced. 18…♗xc3+ 19.bxc3 ♕xc3+ 20.♔f1 dxe4 21.♕xe4 ♕xc2

About 10.♕e2 I wrote that the resulting positions might not appeal to everybody. This is what I was referring to. White has a bishop for 3 pawns, but no pawn cover for his king. 22.♕e2 This is actually a novelty. Up till now white players have traded queens on c2. This leads to an endgame where Black’s sole idea is to get rid of the white a-pawn and draw while being a bishop down. Note the wrong colour of the corner for the bishop, if it’s about the h-pawn at the end. This is a supersafe option to go for – you have winning chances and can’t lose, given you don’t blunder a rook or something like that! 22…♕xe2+ Of course, Black can keep the queens on, but it’s important to show the di erence to taking on c2. If Black keeps the queens on, White will post the bishop on d3 to cover c2. Black does not get anything against White’s king and the bishop is stronger than the pawns. 23.♔xe2 ♖c2+ 24.♖d2 This way it is a bit more di cult for Black to get the a-pawn. White has a freeroll to play for a win. B) 4…♗xf3 5.♕xf3

399

400

Chapter 19: Two Knights with 3…♗g4

So after just 5 moves we have a longterm asset: the bishop pair. This is nice, but the position is closed at the moment and Black’s set-up is very sound with …e7-e6 to come. It is very important to understand that pushing e4-e5, in these sets of lines, is rarely a good idea as White is not well equipped to hold that pawn centre. Black will push …c6-c5 and knights on d7 and c6 will put pressure on the central pawn. Our knight is misplaced on c3 for that scenario, and we don’t have the pawn move c2-c3 to support d4. So be very careful with this advance; it’s almost always bad. B1) 5…dxe4 A bit simplistic, but there is one little trap that White should avoid. 6.♘xe4 ♘d7 7.d4! With the usual slight edge that we aim for in the Two Knights. Instead 7.♘g5?! ♘gf6 8 ♕b3 e6 9 ♕xb7 ♘d5 leads to trouble for our queen – to be avoided! B2) 5…d4?!

401

This is just a mistake as it allows an attractive tempo-gaining move. 6.♗c4 e6 6…dxc3? 7.♕xf7+ ♔d7 8.dxc3 and Black gets wiped out with ♗f4/0-0-0, and so on. 7.♘e2 c5 8.d3 With a vastly improved version of the main lines, as the bishop is on c4 outside the pawn chain, and after ♕g3 White can aim for f2-f4. B3) 5…♘f6 This will very likely transpose to 5…e6, the most popular move. 6.♗e2 We play this move against 5…e6 as well. Instead, 6.d4 dxe4 is either a pawn sac after 7.♘xe4 ♕xd4, which is not entirely clear, or you can go for 7.♕e3 to take one move later. These are alternatives worth looking into, if you are not happy with my ♗e2 set-up. I went for 6.♗e2 because it is the easiest to play as it is possible against both 5… e6 and 5…♘f6 without any major di erence between the two. By the way, 6.♗e2 is not covered in Jovanka Houska’s book about the Caro-Kann. It 402

Chapter 19: Two Knights with 3…♗g4

is a thoroughly researched and wellwritten book, so this omission is surprising.

6…e6 Black may also play for trades with 6… dxe4 7.♘xe4 ♘xe4 8.♕xe4 ♕d5

9.♕g4!. It is important to remember that trading on d5 in such a situation does not help our cause. Black’s pawn structure would be improved and the knight gains a good square on c6. Here we continue with c2-c3/d2-d4 and try to open up the position slowly, to enhance our bishops. 7.0-0 With a transposition to 5…e6.

403

B4) 5…e6

Black’s main moves here are 5…♘f6 and 5…e6. These two moves have some subtle di erences depending on the set-up that White goes for. I have selected a line where it makes no di erence whether Black starts with the pawn or the knight. 6.♗e2 A good alternative against the 5…e6 move order is 6.d4. I am going for 6.♗e2 instead as this is playable against both 5…♘f6 and 5…e6, so we reduce the theory to learn. That being said, going for 6.d4 set-ups is a worthy alternative to explore if you like to expand your repertoire. B41) 6…d4 This leads to a complicated strategic battle. 7.♘b1 ♘f6 8.d3 White’s main idea is to play with f2-f4 later, in the style of the King’s Indian. Note that Black takes a long time to set up with …e6-e5, …c6-c5, compared to a King’s Indian proper. 8…c5

404

Chapter 19: Two Knights with 3…♗g4

There is something interesting to check after 8…♗d6 9.0-0 ♘bd7. White has the interesting 10.♗d1, combining ♕e2, f2-f4, c2-c3 and ♗b3. For example, 10…0-0 11.♕e2 e5 12.c3 I like for White. It is a complicated position though, with chances for both sides. 9.♕g3 ♘c6 10.0-0

Black just stops d2-d4, but the bishop is not on a very stable square, as we will see. 7.0-0 ♘e7 Or 7…♘f6 8.♖d1, which we examine via 6…♘f6 7.0-0 ♗c5 8.♖d1. 8.♘a4 ♗d6 9.d4 And White is comfortable again, having gotten in d2-d4. B44) 6…♘f6 7.0-0 ♗e7 The main move, modestly preparing to castle. Others will transpose most often, only 7…♗c5 has some independent value. 1) 7…♗c5 is interesting, but Black will not be able to prevent d2-d4 completely: 8.♖d1 ♗d4 (the only principled move, otherwise we just go d2-d4 and make ♗c5 look silly) 9.exd5 cxd5 10.♗b5+ ♘c6 11.♘e2 ♗b6 12.d4.

With good play for White, as f2-f4 is coming while g7 is still under pressure. B42) 6…♘d7 7.d4 This is strong, as now after 7…dxe4 we may comfortably recapture with the knight to enjoy a standard advantage. B43) 6…♗c5

White will play c2-c3 and later reroute the bishop to c2 or d3. With ♘e2-g3-h5 coming, there are chances on the kingside after Black has castled. Also note that ♗g5 is an annoying pin that Black may want to avoid. 2) 7…♘bd7 8.d4 is very similar to 6… ♘d7 7.d4 and leads to the usual 405

406

Chapter 19: Two Knights with 3…♗g4

comfortable edge; 3) 7…d4 8.♘b1 – this we have examined via 6…d4 already; 4) 7…dxe4 is Lars Schandor ’s recommendation in his Caro-Kann book, going for the standard position type of the Two Knights after 8.♘xe4 ♘xe4 9.♕xe4 ♘d7 10.d4. He says that by classical standards White should be slightly better (agreed!), but in a practical game chances are equal. I understand the reasoning to some extent, as Black’s position is solid and easy to play. But if you like to be mean you can also put it like this: he basically has to sit and wait anyway for White to do something. Probably this is also a matter of taste. I like space and bishops, you may also argue in favour of a compact structure.

after 8…dxe4 you need to play 9.♕e3 and recapture on e4 later. This basically loses a tempo, compared to 8.♖d1, if Black chooses not to push 8… d4. This tempo is not terribly important though… go 8.d4 if you don’t like allowing Black to push …d5d4. It is easy to play. 8…d4 1) 8…b5 was played by Topalov, but it is not a great choice for Black: 9.e5 (in this rare case, e4-e5 is quite favourable for White as Black cannot push …c6-c5 easily due to b5 hanging. Black needs to lose some time to cover b5, enough for us to support the centre and enjoy a nice edge) 9…♘fd7 10.d4 a6 11.♕g3 and White was already considerably better in Vachier-Lagrave-Topalov, Paris blitz 2017. 2) 8…0-0 is the routine move by Black, but it allows us to reach an optimal version of d2-d4: 9.d4 dxe4 10.♘xe4 ♘xe4 11.♕xe4.

8.♖d1 This is quite ambitious as White wants to play d2-d4 comfortably. It allows 8… d4, though, which might not be to everybody’s taste, as the rook is clearly not optimally placed in this pawn structure. It is still interesting to play, so I choose to give it as the main line. If you dislike it, you may also go 8.d4, but

We have reached the standard comfortable position from the Two Knights, with the bishop pair and some space advantage. Note that White often plays modestly with c2-c3, not

407

408

Chapter 19: Two Knights with 3…♗g4

weakening d4 early on. You can always play c3-c4 later, according to circumstances. There are many ways to reach this structure and piece distribution. I recommend to study sample game 17, Vachier-LagraveRiazantsev, Doha 2016. It is an excellent showcase on how to handle these situations. 9.♘b1 This leads to a complicated strategic battle. I also analysed the wild 9.e5 ♘fd7 10.♕g3, but it was a bit too complicated for the purposes of this repertoire. But it is very interesting to analyse, so go ahead, if you are after some unusual complications. 9…0-0 It is not so easy to analyse a position of this type, as the opponent has many possible moves, and it is not forcing at all. However, I think that Black playing …c6-c5 to go …♘c6 makes a lot of sense, so this is my main line. 10.a4 c5 11.♘a3 ♘c6 12.d3

Conclusion This chapter consists of two very di erent parts. If Black retreats to h5 the play becomes very sharp, but favourable for White. It is absolutely necessary to learn these lines by heart; they are not intuitive. My experience is that you get 4…♗h5 more frequently than you expect, so be ready to face this line with con dence. After the far more solid capture on f3 the play is slow and more strategical in nature. White has good chances of a slight advantage, I think. One key decision for White is whether to allow …d5-d4 or not. I suggest to do so, but you may also go the other way and be ne as well.

White’s idea is to play f2-f4 and expand on the kingside. It will be a complicated game with most pieces still left on the board. 409

410

Chapter 20: Two Knights with 3…♘f6

CHAPTER 20

Two Knights with 3…♘f6 1.e4 c6 2.♘c3 d5 3.♘f3 ♘f6

This is the current fashion, but it is probably too early to call it ‘the modern main line’. Amongst the choices on move 3, it is the move that leads to the greatest complications. We get mostly French-type positions that give White a space advantage, but Black has chances as well. Compared to some lines with 3…♗g4 Black has more counterplay. Currently, there is no way for White to get an advantage, as is the case with most main-line openings. I suggest that we head for a complicated middlegame that is di cult to handle for both sides and o ers some chances for Black to go wrong. 4.e5 ♘e4 4…♘fd7 does not t well with Black’s moves so far. We will get a French with Black being a tempo down – or even more: 5.d4 (simple and good, but 5.e6 411

isn’t bad either, and probably good fun in a blitz game) 5…e6 6.♘e2 (preparing c2-c3 and a transfer of the knight towards the kingside where we have the space advantage) 6…c5 7.c3 ♘c6 8.a3. The position after 7…♘c6 is usually reached with Black to move. This happens via 1.e4 e6 2.d4 d5 3.♘c3 ♘f6 4.e5 ♘fd7 5.♘ce2 c5 6.c3 ♘c6 7.♘f3. So we are getting an extra move now. I suggest playing 8.a3. This is useful in many ways, like preventing a check on b4, and preparing b2-b4 in some cases. We will play on the kingside, and ♘f4 is coming.

5.♘e2 Not the most obvious move, maybe, but it is the most testing choice. This prepares play against the knight on e4 by going d2-d3, without allowing the capture on c3. Also note that with a closed structure it is useful to have c2c3 later, to support the centre. The knight will also do a good job on the kingside, as we will see. 5…♕b6 Threatening mate in one! The main point is that White is forced to go 6.d4 412

Chapter 20: Two Knights with 3…♘f6

to cover that, and in turn loses the option of d2-d3 to kick the knight away. 1) 5…♗g4?!.

A very natural Caro-Kann like move, but it is ill-timed, at this point: 6.♘fg1! Backwards! This backward-looking move actually initiates a sequence of moves that make Black go back even more. First of all we threaten 7.f3, winning a piece. 6…♘c5 (6…♗xe2 7.♘xe2 e6 8.d4 and now with f2-f3 and h2-h4 the knight would be trapped, so Black has to play something that he really does not want) 8…f6 9.♘f4 and Black is in trouble) 7.d4 ♘e6 8.c3 c5 9.f3 ♗f5 10.♘g3 and Black is struggling again; 2) 5…e6, again, does not t with Black’s earlier play: 6.d3 ♘c5 7.c3 and Black’s knight is misplaced: 7…♘cd7 8.d4 c5 9.a3. This would be a French with two extra tempi – compare with 4…♘fd7, please! 6.d4 e6 Instead, 6…c5 is a bit premature, as White is better developed and the exposed queen will be a target, gaining 413

even more tempi: 7.dxc5 ♕xc5 8.♘ed4. We will develop with moves like ♗e3, c2-c3, ♗d3 etc., and be better with even more time gained on the black queen. Note that 6…♗g4? is again answered by 7.♘fg1! and the e4-knight is in grave danger!

7.♘g3 This looks very normal, but usually 7 ♘fg1!? is viewed as the critical line. I do agree, this is a test for Black, but I didn’t nd all that much for White in these lines. Considering that it is a huge e ort to learn and understand everything there, I checked the more natural move 7.♘g3 and decided to go with that. It does not lead to an objective advantage, but it is a good practical decision. 7…c5

414

Chapter 20: Two Knights with 3…♘f6

Black may go for 9…cxd4 10.♘xe4 (instead, 10.cxd4 ♗b4+ 11 ♔f1 allows the additional option 11…f5 for Black. Taking on e4 radically changes the pawn formation, but I think it is a favourable transformation) 10…dxe4 11.♗xe4 dxc3 12.bxc3.

8.c3! Well, this does not really deserve an exclamation mark, as it looks perfectly normal. But there is a reason: I tested this line in a league game and without thinking blitzed out 8.♗d3 here, simply confusing the move order. My opponent just took on d4 and was ne – so this exclamation mark is for me to remember it next time! 8…♘c6 Or 8…cxd4 9.♘xd4. This is forced, as 9.cxd4?? loses immediately to 9… ♗b4+. This is already a very rare position, so we’ll just look at a logical sequence to show the ideas: 9…♗d7 10.♗d3 ♘xg3 11.hxg3. Now Black will not castle queenside anymore. This promises an interesting game with a very likely opposite-side castling scenario: 11…♘c6 12.♘f3 d4 13.♕e2 with a double-edged position. I prefer White, but it is tough to play for both sides. 9.♗d3 Forcing some action by the e4-knight. Black will have to trade on g3 or allow a transformation in the centre. 9…♘xg3

White’s pawn structure is severely weakened, but we have the more active pieces (♗e4!) and it is not easy for Black to develop quickly. 1) 12…♕a5 is too greedy. We happily sacri ce the e5-pawn: 13.0-0! ♘xe5 14.♖b1 and White has a very dangerous attack for the pawn; 2) 12…♕a6 was Grischuk’s choice in a blitz game. At least it poses some problems as we cannot castle immediately: 13.♗f4 (simply covering e5. White’s plan is to play ♕c2, ♗d3 and then castle) 13…♗e7 14.♖b1 h6 15.♕c2. With ♗d3 and 0-0 to come, White is better. Black’s pieces are passive and the e5-pawn is very annoying to play against. 10.hxg3 cxd4 11.cxd4 ♗d7 11…♗b4+?! is quite questionable as it provokes a move that White likes to

415

416

Chapter 20: Two Knights with 3…♘f6

play anyway: 12.♔f1. As we will see in the main line, this is quite a normal set-up for White. However, the bishop is misplaced on b4, as White may, at least, get a useful tempo with a2-a3.

12.♔f1! This is important to remember. This sidestep actually protects the d4-pawn, as we will see. With the h- le open we don’t need to castle anyway. Now there are some options for Black:

16.♖h4! would also win, but it is good to remember this ♖c1 idea, as it might happen in similar cases where b2-b4 is not available, like after 12…h6 13.g4, where the 4th rank is blocked and ♖h4 at the end is no longer e ective. B) 12…h6 It is actually not that clear if White threatens to take on h7, but many Black players may go for this cautious approach. 13.g4 ♘b4 This makes the most sense to me. Black intends to activate or trade the d7bishop. 13…♘xd4?? still doesn’t work: 14.♗e3 ♗c5 15.♖c1. The same idea as seen before. ♖xc5 is a killer that Black can’t avoid anymore. 14.♗e2 ♗b5 15.a3 ♗xe2+ 16.♕xe2 ♘c6

A) 12…♘xd4?? 13.♗e3 ♗c5 14.♖c1

And White threatens 15.♖xc5 with a winning position – Black is helpless against this idea. In this particular situation, 14.b4 ♕xb4 15.♗xd4 ♗xd4 417

Black has managed to trade his potential problem piece on d7. It has taken a lot of time though, so I think we still have a slight pull. 17.g5 This poses the most problems. 17…h5

418

Chapter 20: Two Knights with 3…♘f6

There also is the fearless 17…♘xd4. Black is objectively OK after it, but I think White’s position is much easier to play. Also, take into account that this is a line up to move 22. If you actually remember this in a game, and play all moves without spending much time, you will be ahead on the clock almost for sure: 18.♘xd4 ♕xd4 19.♕b5+ ♔d8 20.♗e3 ♕c4+ (Black cannot allow 21.♕xb7, so a trade of queens is mandatory) 21.♕xc4 dxc4 22.♖h4 and White will get the pawn back with an initiative. The engine just calls it +0.2, but Black has no fun here. 18.♗e3 The immediate 18.g6 is also interesting, but I did not want to carry this line to move 25. Feel free to explore it yourself. I think it is equal at the end, but Black needs to solve problems. The approach I suggest is interesting as well, but needs less memorization. 18…g6 19.g4

With an interesting struggle. One idea for White is to take on h5 and then regroup the knight to f4, trying to win h5. Black has pressure on d4, though, 419

so this is most likely an idea for the endgame. This is an interesting position to analyse further, but you need to make a cut somewhere and say: ‘OK, let’s go for it!’. Still, it is always a good idea for a training session to continue to analyse from the nal position of your prep to get a better feel for the arising positions. C) 12…♗e7

13.a3 Computers suggest taking on h7, but imprisoning your own bishop on h7 looks dubious to me, to say the least. 13.a3 threatens to expand with b2-b4, which would box in Black too much, so he needs to react. Note that Black can’t castle queenside, so any pawn play on the queenside is very tricky for Black to go for if …0-0-0 is on the agenda. I’d like to mention that 13.g4 was played in the high-level game AdamsMotylev, Germany Bundesliga 2014/15. This is a very decent alternative, so I annotated this game and included it in the book. Please see

420

Chapter 20: Two Knights with 3…♘f6

sample game 18 for further information. 13…♘a5 14.♖b1 White does not need to play 14.b4 just yet. ♖b1 is useful anyway, and we rather want Black to commit to something rst. 14…♗b5 Not the only move, but the most direct one. 15.b4 ♕a6! 15…♘c4?! 16.a4 shows a key point of playing 14.♖b1 rst instead of the immediate 14.b4. 16.♗xb5+ ♕xb5+ 17.♔g1 ♘c4 18.a4 ♕d7 19.♕c2

White because I value space a lot, but objectively this line seems dynamically equal. It is crucial to look at the position after 11…♗d7 and the following moves. It seems deep already (move 12), but you will get this position a lot, because 3… ♘f6 is usually played by people who study current theory, and they will know up to this point and further. 3… ♘f6 is usually a signal that Black is booked up, so we need to go deep as well.

And we have reached a typically complicated scenario for this line. I always like to have space, so I prefer White. But the engines evaluate it as equal. This is fair enough, Black has his ideas as well – it will be a tough battle to look forward to. Conclusion 3…♘f6 is Black’s best bet for a complex game. The closed centre positions lead to unbalanced play with chances for both sides. I still have a preference for 421

422

Sample games

 

Sample games Game 16 CK 1.7 (B10) Rauf Mamedov 2660 Luka Paichadze

as the bishop is ideally placed on b2. 9…♕xe3+ 10.fxe3 ♗f5 11.♗b2 ♘d7 12.0-0-0 White of course does not have much here, but his position is easier to play and improve. Gaining space with pawns is not easily done for Black, while White may push his central pawns or the kingside. 12…e6 13.♗e2 ♗e7 14.♖hg1 ♗f6

2535 Nakhchivan 2014 (6)

1.e4 c6 2.♘c3 d5 3.♘f3 dxe4 4.♘xe4 ♘f6 5.♕e2 This move is trickier for Black to face than the more natural 5.♘xf6+. After 5…exf6 Black’s position seems to be quite solid these days. 5…♘xe4 6.♕xe4 ♕d5 7.♕f4 ♕f5 8.♕e3 ♕e6

Not the kind of opening line you show to beginners. ‘Don’t move the queen multiple times in the opening!’ – unless you know what you are doing, you might add. 9.b3 Mamedov had played 9.♗e2 in an earlier game. His novelty is very logical 423

15.d4 There is some logic behind this move. White has more space and thus wants to avoid piece trades. However here it is more important to actually achieve g2g4 as intended, so White should have played it right away. After 15.g4 ♗xb2+ 16.♔xb2 ♗g6 17.h4 White keeps a pull. It is obviously not a very forced varition, so a concrete, deep analysis makes little sense. However some ideas should be explored to get a better idea about the concepts: A) After 17…f6 18.h5 ♗f7 19.d4 White gets in c2-c4 and ♗d3. It is easier for White to gain space; B) 17…h5 18.g5 0-0-0 19.d3. 424

Sample games

analysis diagram

In positions like these, with Black having played …h7-h5, White has some ideas to improve. One key point is the knight going to g3 or f4, putting pressure on the bishop on g6 and the h5-pawn. In general the bishop on g6 is quite unimpressive. 15…h5! 16.♘d2 This allows Black to equalize completely. 16.♗d3 ♗xd3 17.♖xd3 and e3-e4 next still gave White the slightly easier game. 16…♗g5 17.♘c4 ♘b6 Black forces some pieces o and simpli es his defence considerably. 18.♘d6+ Mamedov does not want to make it too easy for Black. 18.♖ge1 ♘xc4 19.♗xc4 ♗e4 is completely equal. 18…♔d7

425

19.♘xf7 White is not running a great risk with this exchange sac as we’ll see. 19…♗xe3+ 20.♔b1 ♗xg1 21.♖xg1

21…♘d5 A very pragmatic decision by Paichadze, simply returning the material. We need to check what happens if Black clings on to the material: 21… ♖hg8 22.♗xh5 ♘d5 23.h4

426

Sample games

analysis diagram

is a logical enough sequence. White has enough for the exchange, with a pawn and the bishops. The knight has good squares on e5 and g5. In this particular position the concrete line 23…♘e3 24.♔c1 ♘xc2 25.g4 g6 26.gxf5 gxh5 27.fxe6+ ♔xe6 28.♘g5+ ♖xg5 29.♖xg5 ♘b4 will end in a draw. 22.♘xh8 ♖xh8 This position is much easier for Black to play than the ones with the extra exchange. It is important to note that Black threatens/intends both …♘e3 and …♗g4. 23.♔c1 ♗g4 And the bishop pair is gone. 24.♔d2 ♗xe2 25.♔xe2

427

Black has defended well, but it is not a clean draw yet. The bishop is still the potentially better piece, but it should not matter all that much if Black is precise. 25…♘f4+ 26.♔d2 ♖f8 27.c4 g5 28.♔e3 ♖f7 29.♗c3 ♘g6 30.♗d2 g4 31.a4 ♘h4 32.♗e1 ♘f5+ 33.♔d3 Black has found a good, harmonious set-up. 33…♔e8 34.♔e4 ♘d6+ 35.♔e3 ♘f5+ 36.♔e4 ♘d6+ 37.♔d3 ♘f5 38.♗f2 ♘e7 39.♗e3 ♘f5 40.♗f2 ♘e7 41.♗g3 ♔d7 42.♔e2 ♘g6 43.♗b8 a6 44.h3 ♖f8 45.♗h2

Not much has happened, but now Black needs to be alert. 45…♘h4? Best is 45…e5 46.hxg4 hxg4 47.dxe5 ♔e6 and this should draw for Black. 46.hxg4 hxg4 47.♗e5?! 47.♗g3! was best here, gaining control of the h- le. After 47…♘f5 48.♗e5 Black is in trouble, as 48…♘h4 49.♖h1 ♘xg2 50.♖h7+ ♔e8 51.♖g7 ♘h4 52.♖xg4 ♘f5 53.a5 is a thankless defensive task. 47…♖f7 48.a5 ♘g6 49.♗g3 e5?! 428

Sample games

He had to try 54…♔xe5. Now White is better after either 55.♔e3 or the simpler 55.♖xg4, but it is not a ‘mathematical’ win yet. 55.♖e4 ♔f5 56.♖f4+? 56.♔e3 was winning, as later played in the game. 56…♔g5? 56…♔xe5 is better, as analysed above. 57.♖e4 ♔f5 Understandably, Black is tired of sitting passively. However, probably he should have continued to wait instead, as we will see. 50.dxe5 50.♔d3! was very strong. It is impossible for Black to avoid capturing on d4 in the long run – and this invites the white king to invade. 50…exd4 51.♔xd4 ♔c8 52.♖e1 is a good line to see the problems. g4 is still weak and White dominates the board. Mamedov’s move keeps an advantage, but it is not as convincing. 50…♘xe5 50…♔e6 51.♖d1 ♘xe5 52.♖d4 leads to the game. 51.♖d1+ Best, as 51.♗xe5 ♖e7 52.♖d1+ ♔c8 53.♔f2 ♖xe5 54.♔g3 ♖xa5 55.♔xg4 ♖a3 will end in a draw. 51…♔e6 52.♖d4 ♖e7 53.♗xe5 ♔f5 54.♖f4+ Very likely time trouble played an important role in the coming moves. 54…♔g5? Black allows the transition into a winning king and pawn ending.

58.♔e3! Now Mamedov plays the winning move. It is not at all easy to gure out that it wins as all depends on one tempo. 58…♖xe5 59.♖xe5+ ♔xe5 60.c5 ♔d5

61.♔f4 ♔xc5 62.♔xg4 b5 429

430

Sample games

62…♔b4 doesn’t save it either: 63.♔f5 ♔xb3 64.g4 c5 65.g5 c4 66.g6 c3 67.g7 c2 68.g8♕+ with a check! 63.axb6! ♔xb6 64.♔f4 c5 65.g4 a5 66.g5 c4 67.g6 It’s never too late to spoil it. Here 67.bxc4?? a4 is a draw. 67…cxb3 68.g7 And Black resigned in view of 68…b2 69 g8♕ b1♕ 70 ♕b8+. The position after the early queen exchange does not look like much at rst, but White is the only one being able to play for anything constructive. Black is solid, but has no active ideas. This is an excellent scenario, where Black may believe he is not facing any dangers, but in fact gets pushed back slowly. In the game Black defended well, but ultimately faltered in the endgame. Game 17 CK 1.3 (B10) Maxime Vachier-Lagrave 2804 Alexander Riazantsev 2671 Doha rapid 2016 (10)

1.e4 c6 2.♘f3 d5 3.♘c3 ♗g4 This is the traditional main line, but it has lost some ground in recent years. The more fashionable lines are 3…dxe4 4.♘xe4 ♘f6 and in particular 3…♘f6. One of the reasons for this shift is the frequently seen position type that we will examine in this game. It is very solid for Black, obviously, but also somewhat passive. 4.h3 ♗xf3 5.♕xf3 ♘f6 6.♗e2 e6 7.0-0 ♘bd7 431

As discussed in the theoretical section, I think that 7…♗e7 is more interesting for Black. White does not get in d2-d4 so easily now. 8.d4 dxe4 9.♘xe4 ♘xe4 10.♕xe4

This is the type of position you’ll get quite often in the Two Knights Caro. White has the pair of bishops and more space, but Black has no pawn weakness, and with two minor pieces already traded, his position is not all that cramped. The white strategy is to slowly expand and while doing so keep and eye on Black’s pawn breaks with the c- or e-pawn. Those pawn breaks are the only real plan that Black has in terms of changing the pawn structure. If we make sure they are not possible at all, or at least unattractive, our play will be much easier. Maxime VachierLagrave does an excellent job in this game to demonstrate White’s possibilities. 10…♘f6 11.♕d3 ♗d6 A natural developing move, but there is more to it. Black would love to trade this bishop, getting rid of the longterm concern of facing the bishop pair. 432

Sample games

One way to trade is via f4, so Black intends to go …♕c7 and …♗f4. 12.♖d1 I like MVL’s concept of dealing with the …♗d6-f4 idea. A di erent approach is 12.g3 0-0 13.♗f3, but I prefer to have the bishop on c4 later, as we will see in the game. 12…♕c7 13.♗d2 Of course, 13.♕f3 was ne as well, preventing the trade on f4 di erently. This is a bit more natural than the game continuation, and probably more precise. 13…♗h2+ 14.♔h1 ♗f4

15.♗e1 I know it might look odd, but placing bishops on the back rank for later use is not rare at all. This is mostly applied in closed positions that are later opened up. 15…0-0 16.a4 This is often useful to gain some space and possibly provoke a concession on the queenside. I am doubtful of the timing though. I probably would have gone for 16.g3 rst, clarifying the situation with the bishop on f4: 16…♗d6 17.♔g2 ♖fe8 433

18.♕f3 (this is very similar to the game, just without a2-a4 for the moment. This is important in some lines) 18…c5 (the position type after the superior 18…e5 is examined in the note to move 16) 19.dxc5 ♗xc5 20.b4 ♗d6 21.c4 and White manages to get his pawn majority rolling. A moving majority combined with the bishop pair can be a powerful asset. 16…♖ad8 We need to check the very principled 16…e5, transforming the pawn structure to complete symmetry. One important aspect of this move: Black opens up the diagonal towards f7, which is an argument for keeping the white kings’s bishop exible (see the note to move 12). 17.dxe5 ♗xe5 18.c3 ♖fe8 19.♕f3.

analysis diagram

I think it is equal here. White has lost a bit of time in the last moves, and it is not clear that having played a2-a4 actually helps. The bishop on f1 likes to be on the a2-g8 diagonal, and often has a good, protected spot on b3; now this is not a stable square anymore. Still, the 434

Sample games

game is not a draw, of course, as White may still push his queenside pawns later. 17.♕f3 ♖fe8 Black has centralized and all his pieces are developed, but what is the plan to make any progress? Pushing …e6-e5 was better timed the move before. 18.a5 a6 That’s the almost automatic response. If Black allows the a-pawn further with a move like 18…♗d6, we happily oblige: 19.a6 b6 20.g3 – a6 will be an asset in any endgame and c6 looks like a weakness Black should rather avoid. It is nothing terribly bad for Black, but the game move is natural and better. 19.♗c4 ♘d5 Black from now on adopts a sit-andwait strategy. I am very sceptical that this is a good solution, but maybe he evaluated …e6-e5 as just too unattractive? As Black you’ll absolutely never win after the e-pawn push, while the game is worse for Black but it remains more complicated. 19…♗d6 20.g3 e5 was the way to go for the ‘central clearance’. 20.♗b3 ♗g5

21.c4 Forward! Still, pushing c2-c4 is a choice that needs to be very carefully considered. There is one drawback: White weakens the d4-pawn quite seriously, and we need to check that we have enough support for it, or may advance it further. Note that once c2c4 is played, any black pawn push …c6c5 or …e6-e5 might be answered with d4-d5. 21…♘e7 22.♗c2 ♗f6 23.♗c3 ♘c8 24.♖d2 ♘d6 25.b3 ♖d7 26.♖ad1 By now the pawn pushes will never happen as White has too much control. Black is in full waiting mode now. 26…♕d8 27.♗b4 g6 28.g3 ♗g7 29.♔g2 ♕c7

30.h4 Bent Larsen once said something to the e ect that you should push the rook pawns if you don’t know what else to do. In fact here it is very sound strategy as Black either has to allow the potential opening of the h- le or weaken himself. 30…♖ed8?

435

436

Sample games

I think he should have played 30…h5, making it more di cult for White to nd a straightforward plan. It’s not easy to play the g-pawn and open up the kingside. White should manoeuvre slowly and probably try to get a later d4-d5 push going. While Black is passive, his position is still a tough nut to crack. 31.h5 ♘e8 32.♗c3 ♘d6 33.g4 ♘e8 34.♕e3 ♕d6 35.hxg6 hxg6

The rst open le on move 36. 36.♖h1 ♕e7 37.♖dd1 White needs both rooks on the h- le. We need to keep in mind that this was a rapid game, and now some inaccurate moves creep in. 37…♕f6? This allows a strong continuation for White. Black had to wait with a di erent move. His situation is not very enviable, but there is no clear win for White in sight yet.

437

38.♖h3? Here 38.♗b4! was strong, controlling the f8-square! Now Black does not have anything better than 38…♖xd4 39.♕h3 ♖4d6, giving up the exchange with a large, probably winning advantage for White. (If Black tries to avoid this and goes 39…♘d6 he runs into problems on the d- le. 40.♖xd4 ♕xd4 41.♖d1 ♕e5 42.♕g3! (this is very strong) 42…♕xg3+ 43.fxg3 c5 44.♗xc5 ♗f8 45.♗b6 ♖a8 46.♗c7 ♘e8 47.♗f4 and White is winning.) 38…♕e7 39.♖hh1! Being able to correct your mistaken decisions is very important. Black threatened to play …e6-e5, so d1 needed cover. Of course MVL checked the natural 39.♖dh1, but Black has one defence. Only one, but it is enough! 39…♖xd4! 40.♗xd4 ♖xd4 and here I start to like Black’s chances. He is not going to get mated anymore, and …e6-e5 and a knight tour to e6 looks scary. White can prevent that, but the game has turned around. 39…♕f6?

438

Sample games

Trying to repeat, which is very understandable. 40.g5? We checked 40.♗b4! already on move 38. 40…♕e7 41.f4

That’s the point of MVL’s play. He tries to establish a full pawn bind, preventing all breaks. 41…♘d6? Here Black absolutely needed to play 41…c5!, all other moves lose! It is not a trivial calculation at all though – in a rapid game this is very hard to do when you are low on time. 42.dxc5 ♖xd1 43.♗xd1 ♗xc3 44.♕xc3 and this is the moment where you will probably stop in your calculation. White threatens mate and after 44…♘g7 45.b4 keeps the pawn. However… 45…e5! saves the day for Black! 46.♕xe5 ♖d2+ 47.♔f3 ♕d7 will end in a perpetual! Easy for engines, but close to impossible to nd and evaluate correctly with short time available. 42.♕h3 Now there is no rescue anymore. The threat is d4-d5, looking all the way to 439

the h8 mating square. 42…♘f5 42…f5 43.♕h7+ ♔f7 44.♖h6 doesn’t help either. 43.♗xf5 exf5 44.♖de1 ♕a3

45.d5! Finally this bishop has its say! 45…♕a2+ 46.♔f1 And Black resigned, being unable to prevent the mate. This position type is a frequent guest in this line of the Caro-Kann. White always has a slight pull, and needs to patiently improve his position. It seems that Black at best manages to simplify to a symmetrical position where he is claiming that the bishop pair is not enough to win. It’s all rather comfortable for White, who is in no risk at all. Game 18 CK 1.6 (B10) Michael Adams 2758 Alexander Motylev 2682 Germany Bundesliga 2014/15 (12)

1.e4 c6 2.♘c3 d5 3.♘f3 ♘f6 440

Sample games

This is currently the most fashionable move and probably the main line of the future. It leads to more combative positions than the traditional 3…♗g4 and is therefore often preferred by modern Caro-Kann players. 4.e5 ♘e4 5.♘e2 ♕b6 6.d4 e6 7.♘g3 The sharpest move is 7.♘fg1, but Michael Adams is a very classical player. His repertoire is always worth checking if you look for a good, common-sense approach to the opening. He rarely goes for ‘crazy’ lines, but rather ‘keeps it simple’. 7…c5 8.c3 ♘c6 9.♗d3 ♘xg3 10.hxg3 cxd4 11.cxd4 ♗d7

This sequence is very likely to happen after 7.♘g3. For deviations along the way please refer to the theoretical section. 12.♔f1 A very important trick that covers the d-pawn by tactical means. 12…♗e7 If 12…♘xd4? 13.♗e3 ♗c5 14.♖c1 wins decisive material. Note that not only 15.♖xc5 is a threat, but also 15.♖h4!. 441

13.g4 I recommend 13.a3 in the theoretical section, preventing 13…♘b4. Unfortunately the games played with this move quite quickly saw White going wrong, so I decided to show this high-quality game with a di erent approach on move 13. It is nevertheless an excellent showcase of the type of position this line leads to. The move played in the game has a couple of ideas. It prepares g2-g3 and ♔g2 and allows ♖h3 in many cases, using the rook on the third rank. 13…♘b4

That must be the critical move. Black is happy to activate his pieces on the queenside, in particular the bishop coming to b5 is desirable. 14.♗b1 The consequences of 14.♖xh7 are not entirely clear to me. I would not recommend to play the move, because the ensuing positions are not easy to handle. The bishop is going to be trapped and I don’t see how to extract it ever. On the other hand I don’t see how Black is supposed to win it either. A sample line: 14…♖xh7 15.♗xh7 g6 442

Sample games

16.♔g1 (White needs to get out of the checks and …♘d3 based tactics) 16…00-0 17.♗g5 ♘c6 18.♗xe7 ♘xe7 19.♕d2 ♘c6. Engines give White a small advantage here, but how are you going to get the bishop out? 14…♗b5+ The Caro player wants this bishop out of the chain or traded, so doing it with tempo is an obvious choice. 15.♔g1 ♕a6 I think 15…♘c6 was a good alternative at this particular moment. Black gets the knight back in order to increase pressure on d4. This prevents the ♗g5 idea seen in the game. The game continuation is not terrible for Black, but I rather want to keep my darksquared bishop in order to provide more safety for the king. You need to be aware that Black will often have his king in the centre in this line. Going kingside is never possible, so there is only this option or the queenside. White should reply 16.♗e3 with a complicated game.

White should grab this opportunity. As mentioned above, this bishop is an important defender. 16…h6 Now we know the address of Black’s king – the centre. It looks tempting to play 16…♗e2, but after 17.♕d2 ♘c6 18.♗xe7 ♘xe7 19.♖h3 White is better. Note that the white rook might go all the way to a3 to attack the queen. There is also 16…♘c6 17.♗xe7 ♘xe7 18.♖h3 with very similar ideas, or even a transposition to the line above. 17.♗xe7 ♔xe7 18.♕d2 ♘c6 19.g5?! This is not well timed. Of course this push is on White’s agenda, but he needed to prepare it. It was better to go 19.♗c2 ♕a5 20.♕e3 ♕b4 21.♗b3 and only later go for the push. White is at least slightly better here. 19…♗e2? The issue with 19.g5?! was 19…♕a5!. Now there is no satisfactory reply. You don’t want to trade queens with Black’s king a bit unsafe, but 20.♕e3 ♕b4 is annoying. 20.gxh6 gxh6

16.♗g5!

443

444

Sample games

21.♖h3 Here Stock sh suggests 21.♘e1 ♖ag8 22.♖h4 h5 23.♗d3 ♗xd3 24.♘xd3 with a nice advantage for White. King safety and better pawn structure count for a lot. Adams’ move keeps a pull as well. 21…♗xf3 Probably Motylev was worried about a line like 21…♖ag8 22.♘h2, which leaves the bishop a bit stranded. 22.♖xf3 ♕a4 After 22…♕a5, again White needs to keep the queens on: 23.♕d1! ♕b6 24.♖f4 ♖ag8. Here engines suggest the remarkable 25.b4!. White will basically play ♗d3 or ♗c2 and ♖b1, no matter what Black does! 25…♘xb4 26.♗d3 ♘c6 27.♖b1 ♕c7 28.♕b3 and the attack is extremely dangerous. 23.♖d3 ♕a5 Against 23…♕b4, 24.♕d1! is strongest, and even better than just trading on b4 and going ♖b3. White has an excellent attack here: 24…♕xb2 25.♗c2 ♘b4 26.♖d2 and Black is in trouble, with combinations of a2-a3, ♗a4 and ♖b1 in the air. 24.♕e3 ♖ac8

25.a3 25.♖b3! is pointed out by the engines as even stronger. 25…♖c7 26.♗d3 ♔d7 27.♗b5 is a possible line. Black is stuck with weaknesses and his king does not nd a safe place. 25…♔d7 26.♖b3 ♖c7 27.♗d3 ♖hc8? This is a decisive mistake. Black needed to play 27…♔c8 in order to have 28.♗b5 ♘e7 available. As we will see in the game, Black is quite helpless after the trade of the minor pieces, so he needs to avoid it. 28.♗b5 a6 29.♗xc6+ ♔xc6 30.♕xh6 First 30.♕c1+ ♔d7 31.♕xh6 was more convincing, but most likely this was a time trouble situation for some moves already. 30…b6 31.♖c1+ ♔b7 32.♖xc7+ ♖xc7 33.♖b4 A funny way to dominate the queen. Black needs to get it back into play. 33…b5 34.g3 ♖c2 35.♔g2 ♕c7

36.a4? White had lost some of his advantage already, but this should have serious consequences. 36.♕f4 was still a huge advantage. 445

446

Sample games

36…♕a5 37.♕f8 ♕c7? 37…♖c8! was leading to a clean draw: 38.♕xf7+ ♖c7 39.♕f8 ♖c8 with a threefold repetition to come.

Black is running out of checks, and resigned here. A tough game that had some timetrouble-in uenced swings at the end. However in general we can conclude that this line of the Two Knights o ers interesting play and good chances to attack the black king as it may be stuck in the centre. I suggest an alternative on move 13 as my main recommendation. It’s good to see that there are multiple options for White. You can choose according to taste and are able to be less predictable in case your opponent knows you repertoire.

38.axb5? 38.g4! wins, but a position like this is impossible to play perfectly in time trouble. 38…♖xf2+? This does not lead to a perpetual, but 38…a5! probably would have: 39.b6 (39.♖b3 ♖xf2+ 40.♔xf2 ♕c2+ is a perpetual straight away) 39…♖xf2+ 40.♔xf2 ♕c2+ 41.♔e3 axb4 and this will end in a perpetual as well. 39.♔xf2 ♕c2+ 40.♔e3 ♕c1+

41.♔d3 447

448

Introduction

PART IV

French Defence: 1.e4 e6 Introduction

1.e4 e6 2.d4 d5

3.exd5 Really? Yes, I am recommending the French Exchange! Isn’t that line just very boring and drawish? This is the usual objection that people have when learning that I advocate trading on d5. Let me explain my rationale behind this choice. The French Defence usually 449

features a pawn structure with White advancing the pawn to e5, either now on move 3 (‘Advance French’), or soon after, for example after 3.♘c3 ♘f6 4.e5 or 3.♘d2 ♘f6 4.e5. The diagram shows the structure after 3.e5.

This pawn structure is usually even called ‘French structure’. The black player will get this structure in most of his French Defence games and will know the basic plans quite well, having this type of game over and over again. He will break with …c7-c5 and/or …f7f6, plans like that are well known. The white player, however, does not face the French as often and will be less experienced in the arising positions. The French player aims for this structure with his choice of opening, so with the Exchange French we are going for the spoilsport option, preventing a closed game with typical manoeuvres that Black will most likely know better than we do. Before we discuss the Exchange French in detail, I need to mention one important point. Due to the often closed nature of the positions after the other main lines (3.♘c3, 3.♘d2, 3.e5) 450

Introduction

the game is not very forcing, but rather slow. This means that Black has a huge choice of lines to go for, which makes it di cult to prepare for. Of course, within the main lines, like 3.♘d2, there are ways to cut down on the theoretical workload, but these ‘second-rate’ versions of the main lines promise no more opening advantage than the Exchange in my opinion. Also, don’t underestimate the nuisance value of playing the Exchange – oftentimes black players will be annoyed, because they don’t get to play the closed type of game they planned and maybe learned in detail. They will also often believe you are just playing for a draw, which is not the case at all! This sometimes tempts black players to underestimate White’s game and encourages them to take unwarranted risks. So what we are doing is: we play for a very slight edge and accept equality if Black does a good job, but we will punish overambitious play and pose Black some problems that he often will not expect to happen in the rst place – isn’t White just playing for a draw? No, we are not! 3…exd5 4.♘f3 This is already important: we need to play 4.♘f3, it is the only move to pose some problems. Why is that? On a very general level you could claim that f3 is the best square for the knight anyway, so we should do this now and only later decide what to do with the bishop on f1, for example. This is true to some extent, but there are concrete considerations as well. The main

alternatives are 4.♗d3 and 4.c4, aiming for IQP play. The drawback of the immediate 4.c4 is that Black may opt for an active set-up involving an early …♗b4, for example after 4.c4 ♘f6 5.♘c3 ♗b4!, which is ne for him. We play 4.♘f3 and keep the option of playing c2-c4 at a better moment, in particular after 4…♗d6 we go 5.c4 – this way …♗b4 is not an attractive option anymore. Let’s have a look at 4.♗d3. This is a common move, but actually imprecise. Black has the interesting option of playing 4…c5 (note the similarities to our idea to play 4.♘f3 ♗d6 5.c4) or the simpler 4…♘c6, after which the d4pawn is attacked. Now 5.c3 is very passive, and 5.♘f3 allows the pin with 5…♗g4, which is a lot stronger now than after 4.♘f3 ♗g4, as we will examine later. In general, a black set-up with …♘c6, …♗d6, …♘ge7 and …♗f5 or …♗g4 will be OK for Black. We should always try to prevent such a setup in order to pose Black more problems.

451

452

Introduction

After 4.♘f3 Black has many moves available. Note that it pays to learn the details for each, but if you forget some detail your position will still be very sound and solid. It is hard to get into trouble with white early on in the Exchange French. The coverage of the French Defence is split into four chapters. Chapter 21 deals with early deviations by Black and brie y covers rare moves after 4.♘f3. The main chapters are: Chapter 22: 4…♗d6

Very natural play, but Black gives up on the mentioned idea to play a later … ♘e7 and …♗f5. In this symmetrical line Black stays very solid, but we can still pose problems. Please see Chapter 23 for details. Chapter 24: 4…♘c6/4…♗g4 Black has two ambitious moves, often used to unbalance the game. These are 4…♘c6

This is one of Black’s tries to reach the …♘e7/…♗f5 trade. Against …♗d6 we go for the IQP with 5.c4. This is analysed in Chapter 22. Chapter 23: 4…♘f6 and 4…♗g4.

453

454

Chapter 21: Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

CHAPTER 21

Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines 1.e4 e6 2.d4

We need to know some lines against them, but studying Chapter 24 will give you all the information you need to face them with con dence. In fact, Black struggles here; in particular after 4…♗g4 White is better.

Black sometimes leaves the French territory with 2…c5. This is a quite rare move that invites White to play 3.d5, leading to a Benoni type of position. The main theoretical objection against 2…c5 is this transposition into a Benoni position where White is slightly better due to the space advantage. If you are familiar with Benoni structures, you should probably play 3.d5, but if not, it does not make sense to learn all this for a line you will maybe never face. I suggest to play 3.c3. This is the most practical choice. We just transpose into our line against the Sicilian starting with 1 e4 c5 2 ♘f3 e6 3 c3, something that we should know well anyway as it will be on the board quite frequently. Another rarity is 2…b6. This transposes into the Owen’s Defence, 455

456

Chapter 21: Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

usually starting with 1….b6 and covered via this move order: 3.♗d3 ♗b7 4.♘f3 and we have transposed to the mentioned line starting with 1… b6. 2…d5 3.exd5 exd5 3…♕xd5 is a Scandinavian with a pawn on e6, forfeiting the option of developing this piece outside of the pawn chain. It is a very rare move, but one thing is good to know: going ♘c3 immediately is less promising than delaying this move. 4.♘f3 ♘f6 5.c4 ♕d8 6.♘c3 is the best way of handling the situation. The knight is better placed behind the c-pawn than in front of it. This way the pawn on c4 controls important squares. 4.♘f3 The most common moves now are examined in the following three chapters. Here we are going to check the rare lines. A) 4…♗f5

This is not very popular, but certainly a quite logical-looking developing move. 457

There is one drawback though: Black’s kingside development is lagging and this, surprisingly, may become a serious issue soon. 5.♗d3! It is important to remember that, instead, 5.c4 is no good because of the very direct reply 5…♗xb1+! 6 ♖xb1 ♗b4+ and Black manages to trade some minor pieces favourably. Black is not better, but he equalizes comfortably. 5…♗xd3 6.♕xd3 c6

The simplest way to address the 7.♕b5+ threat. 7.♕e2+! One of the surprising revelations studying the French Exchange was to learn how often this check irritates Black. 7…♕e7 This is probably best, but it clogs up Black’s development a bit. After 7…♗e7, 8.0-0 ♘f6 9.♖e1 poses a problem. Black cannot castle for the time being. He needs to regroup: 9… ♘bd7 10.♘bd2 ♘f8 11.♘b3 ♘g6 12.♘e5 and White is better. Black still cannot easily castle. 458

Chapter 21: Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

8.♗e3 ♕e6 The bishop move to d6 needs to be prepared ASAP. 9.0-0 ♗d6 10.♘g5

Black is probably equal with best play, but he needs to nd the right moves rst. White’s position is de nitely easier to handle here. Keeping Black busy. He must solve concrete problems now – where to move the queen? 1) 10…♕e7?! is the most cautiouslooking move, but it leads to problems: 11.♖e1 h6 (11…♘f6 12.♗f4! is very awkward for Black to face) 12.♘f3 ♘f6 13.♗f4 ♕xe2 14.♖xe2+ ♔d7 15.♗e5 (White has an annoying initiative. Black should prevent 16.♗xf6, so he needs to go back with his knight) 15… ♘e8 16.♘bd2 f6 17.♗xd6 ♘xd6 18.♖ae1. White is still ahead here, keeping the e- le and having weak squares on the kingside to exploit (♘h4-f5, for example); 2) 10…♕f5 is a bit cold-blooded, but Black has to allow the discovered check: 11.♗f4+ ♗e7 12.♕d2.

459

B) 4…c5?

This move is clearly mistimed. Black opens the position while being behind in development, in particular on the kingside. You may also compare the position to 1.e4 c5 2.♘f3 e6 3.c3 d5 4.exd5 exd5 5.d4 – here White has ♗b5+ instead of c2-c3, which makes a big di erence in a wide open centre position. I will not analyse this line in huge detail as Black has many di erent moves available, so looking at all of them will only make things messy. You 460

Chapter 21: Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

need simply to develop quickly here, using your momentum. Note that often ♘c3 is a lot better here than some slow build-up with pawn to c3. Typical moves are ♖e1+, ♗g5, ♘e5 etc., immediately increasing the pressure. 5.♗b5+ ♘c6 Instead, 5…♗d7 is even worse: 6.♗xd7+ ♕xd7 7.0-0 ♘f6 8.♖e1+ ♗e7 9.dxc5 0-0 10.♘c3 ♗xc5 11.♗g5 and White has a huge advantage; he will win d5 for no compensation. 6.0-0 ♘f6 7.♘c3 This is not played all that often, but it is strong, putting immediate pressure on d5. 7…♗e7 8.♘e5 ♗d7 9.♘xd7 ♕xd7 10.dxc5 ♗xc5 11.♗g5

And Black is in grave trouble again.

5.♗d3 either. Black will very likely play …♗d6/…♘e7/…♗f5, but White may go with a setup involving b2-b3 and c2c4 later, to grab some space. It is a good alternative if you like to vary your play.

5…♘f6 Some possible transpositions: 1) 5…♗d6 6.♘c3 transposes to 4… ♗d6 5.c4 c6 6.♘c3, examined in Chapter 22; 2) 5…♗b4+ 6.♘c3 ♘f6 7.♗d3 – this we examine via 5…♘f6 6.♘c3 ♗b4 7.♗d3 below. 6.♘c3 ♗e7 Or 6…♗b4 7.♗d3 0-0 8.0-0. I think this is a slightly inferior version of the 6…♗e7 line for Black. I see no upside of having the bishop on b4 as taking on c3 is rarely a serious option. 7.♗d3 0-0 8.0-0

C) 4…c6 This is not frequently played, but quite solid. Black supports the centre and is still exible regarding kingside development. 5.c4 This is most ambitious, but there is nothing particularly wrong with 461

462

Chapter 21: Exchange Variation: 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

11.h3 ♗h5 12.g4 ♗g6 13.♘e5

Now Black captures on c4 most of the time. We don’t mind if Black delays it, as we have many useful moves available (♖e1, h2-h3, ♗f4, etc.). 8…dxc4 9.♗xc4 This position is comparable with lines after 4…♗d6 5.c4. One key di erence is that the bishop is on e7, thus making a pin with ♗g5 less attractive. It is important to note that Black has committed to …c7-c6, so he is no longer able to choose the most active set-up with …♘c6. White has a comfortable game, while Black is very solid. Let’s check two possible continuations. C1) 9…♗g4 10.♖e1 This is more than a routine move. White looks at the e7-bishop, so that, after …♗xf3, he will be able to recapture with the queen. 10…♘bd7 The forced line 10…♗xf3 11.♕xf3 ♕xd4 12.♖xe7 ♕xc4 13.♖xb7 is slightly better for White. He enjoys the better pawn structure and will also bene t from having bishop vs knight in the endgame. 463

A typical manoeuvre to seize the initiative. White is better here, intending f4-f5, as we have seen numerous times in these lines. C2) 9…♘bd7 10.♖e1 ♘b6 11.♗b3 ♘bd5 12.♗g5 ♗e6 13.♘e5 With a fairly standard IQP position. White enjoys more space, but Black is obviously quite solid. Conclusion The early deviations in the French and sidelines versus the Exchange Variation are not worrisome at all. They are either questionable or lead to positions that are very similar to the following chapters. This is true in particular about 3.exd5 exd5 4.♘f3 c6, which is very similar to or even transposes to Chapter 22.

464

Chapter 22: Exchange Variation with 4…♗d6

CHAPTER 22

Exchange Variation with 4… ♗d6 1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3 ♗d6 This signals Black’s intention to later play …♘ge7 and …♗f5, or sometimes …♗g4. This set-up we like to avoid, so our choice is to change the structure immediately. 5.c4

Now that Black has moved the bishop, a …♗b4 set-up is no longer a worry. A) 5…♘e7?? This has been played 32 times in Megabase 2018. 6.c5 Amazingly, only 27 out of 32 games saw this reply. B) 5…c6 This is played to prepare …♘e7. 6.♘c3 ♘e7 Instead, 6…♘f6 would have transposed to 5…♘f6. 465

7.♗d3 I like this natural move, but instead 7.c5 is an interesting alternative that leads to completely di erent play. 7…0-0 8.0-0 dxc4 Moving the bishop on c8 is no good: 1) 8…♗g4? 9.♗xh7+ – oops!; 2) 8…♗f5?! is ill-timed in view of 9.♗xf5 ♘xf5 10.cxd5 cxd5 11.♕b3 and White wins a pawn. 9.♗xc4

This position is easy to play for White. He enjoys good development and Black does not have any pressure against the IQP. That being said, Black is quite solid, of course. 9…♘d7 Quite typically, 9…♗g4 10.h3 ♗h5 should be answered by 11.g4. You may also play a more conservative move, but this puts Black under pressure. This g2-g4/♘e5/f2-f4 set-up is always worth considering, and is often just good for White: 11…♗g6 12.♘e5 ♘d7 13.f4 (now f4-f5 is a serious threat, limiting Black’s options) 13…b5 14.♗b3 b4 15.♘xg6 ♘xg6 16.♘e4 and White is better with more space and 466

Chapter 22: Exchange Variation with 4…♗d6

the pair of bishops. Of course, you need to be careful with the open state of White’s king. 10.♗g5 ♘b6 11.♗b3 ♔h8

This is played most of the time. It looks a little odd, but the pin is annoying, and this way …f7-f6 is prepared. 12.♗xe7 A concrete approach, trading the bishop pair for time. Other normal, active moves, like ♖e1, for instance, are also ne, of course. 12…♕xe7 After 12…♗xe7 13.♘e5 ♘d5 14.♗xd5 cxd5 15.♕b3 ♗e6 16.♕xb7 White’s initiative has netted a pawn. Black’s compensation seems insu cient. 13.♖e1 ♕c7 14.♕d3 This prevents 14…♗f5 and means that 14…♗g4 can be answered with 15.♘e5. White is slightly better here.

This position is reached quite often via a totally di erent move order, starting with the Queen’s Gambit Accepted! 1.d4 d5 2.c4 dxc4 3.e3 e5 4.♗xc4 exd4 5.exd4 ♘f6 6.♘f3 ♗d6 are the concrete moves. 7.♕e2+! This is a very important idea that makes this line more attractive for White. There is nothing wrong with just developing with 7.♘c3 (and in fact this is part of our repertoire after 5.c4 ♘f6 6.♘c3 dxc4 7.♗xc4), but the queen check is tougher to meet for Black. C1) 7…♗e7 Black moves the bishop again. Nothing to worry about, but it is a sound choice. 8.0-0 0-0 9.♘c3

C) 5…dxc4 A small concession, but it is not terrible for Black. 6.♗xc4 ♘f6 It should be mentioned that 6…♘e7?? is not possible at all. After 7.♘g5 0-0 8.♕h5 Black collapses on f7/h7. 467

468

Chapter 22: Exchange Variation with 4…♗d6

Very natural play by White. In general, always try to play the most active moves, going for piece play. This sounds like good advice in general terms, but it applies in particular to IQP positions. C11) 9…♗g4 10.h3 As usual we directly attack the pinning bishop. Black now only has the choice between being tactically or strategically worse. C111) 10…♗h5?!

Black is already in grave danger. Here is some evidence: 1) 12…c6? A lame move like this does not cut it: 13.h4! and Black is already lost. The o side bishop is the problem; 2) 12…♕xd4? loses straight away as well. It is a long line, but simple to remember or nd over the board: 13.♘xg6 hxg6 14.♖d1 ♕c5 15.g5 ♘h5 16.♖d5 ♕c6 17.♗b5 ♕e6 18.♖e5 ♕xh3 19.♖xe7; 3) 12…♘c6 13.♘xc6 bxc6 14.f4! h6 15.f5 ♗h7 16.♖d1 and Black is in terrible shape with the dead piece on h7. C112) 10…♗xf3 The strategic concession mentioned above. 11.♕xf3 ♕xd4 After 11…♘c6 12.♗e3 ♘xd4 13.♕xb7 White is better with the long-term pluses of the pair of bishops and the better pawn structure. 12.♕xb7 ♘bd7 13.♗b3 And again White is better with the bishops and better pawns. C12) 9…♘c6 10.♖d1 ♗f5

This gets punished by typical, direct play. 11.g4 ♗g6 12.♘e5

469

470

Chapter 22: Exchange Variation with 4…♗d6

11.♗g5 The engine suggestion to improve on previous games. Black is under some pressure from the active white forces. 11…h6 12.♗h4 ♘d7 13.♗g3 ♗d6 14.♘e5

And Black’s defensive task is not easy at all.

arising position is not so well coordinated for Black, and tough to organize. Looking at some moves will show the issues that Black faces: C21) 8…♔xe7 9.0-0 ♗e6 This is played almost all of the time, but after it White is on top for sure. The engines like 9…♘c6 best, but it is not particularly fun for Black, either: 10.♖e1+ ♔f8 11.♘c3 (how does Black organize his forces now?) 11…♗g4 12.♘g5 ♗h5 13.♗e3. This is just much easier to play for White. The computer gives White just a slight edge, but in practice Black is su ering here. 10.♖e1 ♔d7 11.♗xe6+ This is the simple solution, but keeping the tension is also quite promising. 11…fxe6 12.♘c3 ♖e8 13.♘e5+ ♔e7 14.♗g5

C2) 7…♕e7 This is more natural compared to 7… ♗e7, but White retains an unpleasant pull.

White has a nice pull here with superior activity.

8.♕xe7+ Why does White trade pieces when having the IQP? Isn’t that a weird way to play? The reasoning here is that the 471

C22) 8…♗xe7 This way Black keeps castling rights, but he has lost time with the bishop and will do so even more. 9.0-0 0-0 472

Chapter 22: Exchange Variation with 4…♗d6

Stock sh suggests 9…♗e6, giving White just a minimal edge: 10.♗xe6 fxe6 11.♘g5! (this is important; the obvious alternative, 11.♖e1, is strongly answered by 11…♘c6! 12.♖xe6 ♖d8 and Black is OK) 11… ♔d7 (the only move to keep the pawn) 12.♘f7 ♖e8 13.♘e5+ ♔c8 14.♘c3. Black is quite solid, but still I’d rather be White here. Black is not fully coordinated and e6 still feels like a more serious weakness than d4. 10.♖e1 ♗d6

12.♘b5 instead would net the bishop pair and is a good choice as well. I prefer 12.♘e5 though, to keep more pressure. 12…♘bd7 13.♗f4 ♘b6 14.♗b3 ♗f5 15.♘b5 White has a slight pull here, for sure, but Black is quite solid. D) 5…♘f6 Maybe Black’s best move available, keeping the tension. 6.♘c3

11.♘c3 11.♗g5 is a very good alternative. It was played in our sample game 20, Ponkratov-Magalashvili, Amol 2018. 11…h6 This prevents ♗g5 by White. Stopping 12.♘b5 with 11…a6 is also possible, but allows very uent development for White: 12.♗g5 ♘bd7 13.♘e5 ♘b6 14.♗b3. White is in good shape here and has easy moves to play. The pressure on f7 is very annoying to handle for Black. 12.♘e5

D1) 6…c6 A solid choice, simply covering the d5pawn. 7.♗d3 0-0 8.0-0 dxc4 9.♗xc4 Compared to 5.c4 dxc4 or 5…♘f6 6.♘c3 dxc4 Black is a full tempo up, but he has spent it on …c7-c6, which is quite passive in this set-up. 9…♘bd7 As usual, 9…♗g4 is not advisable if Black does not have enough control over e5: 10.h3 ♗h5 11.g4 ♗g6 12.♘e5. With f4-f5 ideas. Black is struggling here already. 10.♗g5 ♘b6 11.♗b3 ♗f5 12.♖e1

473

474

Chapter 22: Exchange Variation with 4…♗d6

I prefer White here, with his active pieces and annoying pressure on f7, combined with the pin of the f6knight. D2) 6…dxc4 7.♗xc4 This position is very similar to 5.c4 dxc4 6.♗xc4 ♘f6 when I recommend 7.♕e2+!. If White plays 7.♘c3 instead we would reach exactly the same situation here. So in some way 5…♘f6 6.♘c3 is a more precise move order by Black. That being said, White still has chances to play for an advantage. 7…0-0 8.0-0

8…♘c6! A very important move for Black! It sounds a bit weird, maybe, but other set-ups seem to be problematic for him: 1) 8…♗g4?! 9.h3 ♗h5?! (taking on f3 is just handing over the pair of bishops for nothing, but this is even worse) 10.g4 ♗g6 11.♘e5 (this is a recurring theme in these IQP positions; Black is in severe trouble due to the threat of f2-f4-f5) 11…♘fd7 (this is relatively best) 12.f4 ♘b6 13.♗b3 ♗xe5 14.dxe5 with a clear advantage for White; 475

2) 8…♘bd7 9.♗b3 ♘b6 10.♖e1 c6 11.♗g5 and White is much more comfortable here. ♘e5 is a serious idea, putting a lot of pressure on Black. 9.h3 This is a very useful move, preventing 9…♗g4, now that the mechanism of h2-h3/g2-g4/♘e5 no longer works due to Black’s increased control over e5, after he played 8…♘c6. D21) 9…♗f5 This is rare, but some people really don’t care about prophylaxis at all. 10.♗g5 ♗e7 After 10…h6 11.♗h4 g5 12.♗g3 ♗xg3 13.fxg3 Black will su er with his weakened king position in the long run. 11.d5 ♘b8 12.♘d4 ♗g6 13.♕b3 And White is clearly better. D22) 9…h6 Black also fears 10.♗g5; he answers in symmetrical fashion. 10.♕c2 I think this is interesting because of the attacking ideas shown below. The lines after 10.♕c2 go exceptionally deep for this repertoire, but they were interesting to explore, so I decided to include them. Again, if you don’t feel like going there: 10.♖e1 or any other active move is ne. 10…♘b4 Black also has 10…♘a5; a good move, pushing the bishop on c4 back to a less active square: 11.♗d3 ♗e6 12.♖e1 ♘c6 13.a3. This position is roughly equal, but with lots of play left. White will centralize (♗e3, ♖ad1) and then 476

Chapter 22: Exchange Variation with 4…♗d6

choose a plan. I like the idea to go ♘c3a4-c5, but there are also scenarios for ♗e3/♕c1 and ♗xh6 ideas. 11.♕b1

21.♖e4 (an important double attack; White gets the piece back) 21…♗f6 22.♖xb4. This should be somewhat better for White, but it is di cult to play this position perfectly. At least White has a clear idea: 3 connected passers! 17.♕f4 ♗xe4 18.♕xe4

D221) 11…c6 This normal-looking move allows White to realize his attacking idea connected to ♕c2. 12.♗xh6 Computers evaluate this as equal at the end, but White has all the fun. 12…gxh6 13.♕g6+ ♔h8 14.♕xh6+ ♘h7 Black needs to avoid 14…♔g8?? 15.♘e5. Note the ideas ♘g6 and ♖e1e3-g3: 15…♗xe5 16.dxe5 ♘h7 17.♘e4 ♗f5 18.♖ae1 and White has a winning attack. 15.♘e4 ♗e7 16.♘e5 Black is under huge pressure. Engines nd their way around this, but it is very unpleasant for humans to play Black. 16…♗f5 After 16…♕xd4 we get to an endgame almost by force: 17.♘xf7+ ♖xf7 18.♗xf7 ♕xe4 19.♖ae1 (White needs to avoid 19.♗g6? ♕h4!, which defends for Black) 19…♕h4 20.♕xh4 ♗xh4

It is always tough to choose how long a line should be analysed in a publication like this. I have checked this position with the help of engines, of course, and they evaluate it as equal, so White is not in trouble. In fact, I think it is much easier to play White here. I will just give two sample lines that look logical to me: 1) 18…♕d6 19.♗xf7 ♘g5 (White was threatening 20.♘g6+, so this is mandatory) 20.♕g6 ♕f6 21.♕h5+ ♔g7 22.♗b3 ♕f5 23.f4 ♘e4 24.♕xf5 ♖xf5 25.g4 ♖ 8 26.♖ae1 and while the computer gives 0.00 I don’t think that many humans would choose Black here; 2) 18…♘d5 19.♕f5 ♗d6 20.♘xf7+ ♔g7 21.♘xd8 ♖xf5 22.♘xb7 ♗b4 23.a3 ♖b8 24.♗a6 ♗e7. White has

477

478

Chapter 22: Exchange Variation with 4…♗d6

four pawns for the piece by now, but Black’s pieces are active. Computers see until 0.00, but for humans it is interesting to play. D222) 11…♗e6 This is the drastic way of opposing the strong bishop on b3. 12.♗xe6 fxe6 13.♖e1 The e6-pawn is a weakness now, as is the square g6. Black has a good square on d5 for his knights as compensation. I think that White is only very slightly better. 13…♕d7 14.a3 ♘bd5 15.♘e5 This is a forcing approach. As Black won’t allow ♘g6, this knight has to be taken. 15…♗xe5 16.dxe5 ♘xc3 17.bxc3 ♘d5 18.c4 ♘b6 19.♗e3 And White has a small advantage. D3) 6…0-0

A clever move order by Black. Now going 7.♗d3 leads to line D2 with a tempo down, so White is more or less forced to take on d5. 7.c5 is playable, by the way, but I don’t like to advance while still being 479

underdeveloped. 7.cxd5 This move also does not develop immediately, but it frees c4 for the bishop and Black will have to spend time getting the pawn back. We will try to hang onto this pawn! This is not so much due to the idea that an extra pawn is useful, but more because on d4 and d5 the pawns control some good squares. D31) 7…♖e8+ 8.♗e2 Instead, 8.♗e3 ♘g4 is annoying to say the least. 8…♘bd7 Black’s other sensible option is to play 8…h6 to prevent a ♗g5 pin that would help to keep the pawn on d5. It is slow, though, and I think White is somewhat better now: 9.0-0 ♘bd7 10.♗c4 (the key point. The bishop changes its position and helps to keep the pawn on d5) 10…♘b6 11.♗b3 a5

12.♘e5! (an important idea to remember) 12…♗xe5 13.dxe5 ♖xe5 14.a4 (after the coming ♗f4, White will be better due to his bishop pair and space advantage. Note that ♕d4 is an 480

Chapter 22: Exchange Variation with 4…♗d6

important idea as well) 14…♘bxd5? (of course, we have to check this) 15.♘xd5 ♘xd5 16.f4 ♖f5 17.g4 and White wins. 9.0-0 ♘b6 Now we don’t want to simply return the pawn, so we need to pin the knight on f6. 10.♗g5 ♗e7 This is sensible. How else is Black going to get the pawn back? 11.♗b5 ♗d7 12.♗xf6 ♗xf6 13.♕b3 And White keeps the pawn. Note that as long as the pawn is on d5, Black’s c7-pawn is a target. D32) 7…♘bd7 8.♗g5 This is played in accordance with the ‘Keep It Simple’ approach. Instead, 8.♗c4 is interesting to check as well. After 8.♗c4 (the move we would like to play for sure!) Black goes 8…♖e8+ 9.♗e3 ♘g4 and these complications are unexplored territory that you might want to analyse. I could not nd an advantage, but it’s a good alternative to check. 8…h6 9.♗h4 ♘b6 Here 9…♖e8+ 10.♗e2 would transpose to 7…♖e8+. 10.♗e2 ♗e7 We have already analysed this position with the additional moves …♖e8 and 0-0. The di erence is that ♗b5 is useless here, as there is no rook to attack on e8. Nevertheless, Black needs to unpin here, in order to try to get d5 back. 11.♗xf6 ♗xf6 12.♕b3

481

We’ve kept the pawn! But, of course, it is a doubled pawn and Black has got the bishop pair, so we are not simply on the way to converting extra material. Let’s look a bit further. 12…a5

Trying to provoke 13.a4 probably makes sense for Black. 13.0-0 Note that instead 13.a3?! a4 14.♕a2 is bad due to the strong reply 14…♖a5!, as happened to me in online blitz before I had looked at this in detail. Better is 13.a4, as already played in two master games, but not moving the apawn at all seems best to me. 13…a4 14.♕b5 ♗d7 15.♕d3 ♖a5 16.b4 axb3 17.axb3 ♖xa1 18.♖xa1 Black is probably OK here, but he is still a pawn down and needs to prove it. Conclusion Black’s 4…♗d6 is a ‘middle of the road’ approach. It is more combative than the symmetrical 4…♘f6, but not as uncompromising as the lines of Chapter 24. With 5.c4 we will get an interesting, unbalanced game. The best defences 482

Chapter 23: Exchange Variation with 4…♘f6

should get Black equality or close to that. However there are many other lines that are frequently played that lead to trouble for Black. For example, the 7.♕e2+ line is not easy for Black to handle. Another common scenario sees Black playing an untimely …♗g4, which is usually punished by h2h3/g2-g4 etc.

CHAPTER 23

Exchange Variation with 4… ♘f6 1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3 ♘f6

Black just stays symmetrical. Unsurprisingly, this move is the most solid and reliable choice that Black has. I always thought this simple approach was the main argument against the Exchange French, as White gets nowhere. However, it turned out to be not as simple, after a closer look. White has some ideas that Black needs to answer precisely – let’s check some details. 5.♗d3 Here 5.c4 is again answered by 5… ♗b4+, which is ne for Black. 5.♗d3 is best. Now Black needs to choose if symmetry should be maintained or not. A) 5…c5

483

484

Chapter 23: Exchange Variation with 4…♘f6

Black plays very actively, directly attacking the d4-pawn. White needs to be precise to retain a slight pull. 6.0-0! In general we would like to take on c5, but not when Black is able to recapture in one go. 6…c4 This is the most ambitious move, giving this line independent value. 1) With 6…♗e7 Black keeps the tension, allowing the IQP structure under more favourable circumstances for White. This will very likely transpose to a position that we examine via 5…♗e7 6.0-0 0-0 7.♗g5 c5 7.dxc5 ♗xc5 8.♖e1+ ♗e6 9.♗g5 (it is important to avoid 9.♘g5 0-0!, which gives Black good play!) 9…0-0 10.♘c3 ♘c6 11.♕d2 and we have transposed to the mentioned line usually starting with 5…♗e7 (line C in this chapter); 2) 6…♘c6 7.♖e1+ ♗e7 8.dxc5 0-0 9.♗g5 ♗xc5 10.♘c3 ♗e6 11.♕d2 and again we get the position usually reached via 5….♗e7 6.0-0 0-0 7.♗g5 c5. 7.♖e1+ ♗e7 8.♗f1 0-0

9.b3 Black has gained some space, but we are ready to attack the advanced pawn, gaining the better pawn structure as Black is forced to capture. 9.♗g5 was played by Kasparov in this position, and is a ne alternative, also leading to a small edge. 9…cxb3 10.axb3 ♘c6 11.♘e5 A very straightforward idea. White intends to take on c6, followed by ♗a3 most often. This gives us the slightly better pawn structure. 11…♗f5 Aagaard and Ntirlis mention 11…♖e8 in their book on the French, but after 12.♘c3 I still would prefer White slightly. 12.♘xc6 bxc6 13.♗a3 ♗xa3 14.♖xa3 We have a small edge here, trying to exploit the better structure. Note that Black has pawns on light squares and a light-squared bishop. We would like to x those pawns long-term. There are some similarities between this position and our play in the Scotch Four Knights.

485

486

B) 5…♗g4

Chapter 23: Exchange Variation with 4…♘f6

This has little independent value – it will transpose to lines usually reached via 5…♗e7. 6.0-0 ♗e7 7.♗g5 This we examine below under C).

against, when trades of minor pieces help. 6.0-0 0-0

C) 5…♗e7

Placing the bishop on e7 means that any unpleasant checks on the e- le are avoided, but quite often Black will move the piece again to d6 after all. Against these ♗e7 set-ups I recommend playing with a standard set-up of 0-0/♗g5/♘bd2/c2-c3 and ♕c2, in most cases. It is important to mention that White should play these moves rst and avoid an early ♖e1. We will see in the main line why this natural move should be played only later. Another key point is that Black will play …♗g4, after which we should not play h2-h3, as …♗h5 is useful for Black since trading the bishops with … ♗g6 is part of his agenda. This applies in particular if Black still has the pawn on c6, when the bishop is the potentially ‘bad bishop’. After …c7-c5 by Black we don’t mind trading so much, as we have the IQP to play 487

Black may also play 6…♗g4 rst. It has no independent value compared to 6… 0-0. We still play ♗g5/♘bd2 etc. like in the main line. Black will most likely castle next move and transpose anyway: 7.♗g5 0-0 8.♘bd2 and we have transposed to 6…0-0 7.♗g5 ♗g4, the main line below. 7.♗g5 This should be played rst, waiting with ♖e1 until it is better timed. We’ll see some moves later why saving this move is important. I would also like to mention that 7.h3 is a very decent alternative, but Black may also start with 5…♗g4 6.0-0 ♗e7 when we would need to transpose to 7.♗g5 anyway. So we go for a line that is usable against all ♗e7 set-ups, meaning that we don’t need to learn two separate lines. 7…♗g4 Very natural, and the most popular by far.

488

Chapter 23: Exchange Variation with 4…♘f6

Black also has a completely di erent idea with 7…c5. Black goes for an IQP set-up to get free piece play, but of course White has a target now – the isolated pawn. It is interesting to compare this line to the IQP line that we play after 3.exd5 exd5 4.♘f3 ♗d6 5.c4. Black needs two moves to take on c5, and we have White here, so it’s two extra tempi, which gives us a nice position: 8.dxc5 ♗xc5

Instead, I suggest to just play 11.♗h4 and be better with the long-term advantages. 8.♘bd2

9.♘c3!. This active move is a lot better than c2-c3/♘bd2 based set-ups. The direct threat is 10.♘xd5!, winning the centre pawn: 9…♗e6 (this is necessary to cover d5, and a sign that Black is not all that comfortable here – it is solid, but passive. 9…♘c6? 10.♘xd5! ♕xd5 11.♗xf6 and we have won a pawn, as Black cannot take 11…gxf6?? 12.♗xh7+) 10.♕d2 (we aim for full development with ♖ad1, ♖fe1, and put pressure on Black’s central IQP. White is slightly better here) 10…h6 11.♗h4. This is simple and good. I couldn’t make 11.♗xh6 work fully, but it is quite interesting to analyse. White has an attack with a draw in hand.

8…♘bd7 Again the main alternative is 8…c5. Now White already has the knight on d2, so this looks a bit more attractive, but there is a slight issue with combining …♗g4 and …c7-c5: 9.dxc5 ♗xc5 10.c3 ♘bd7 (not the ideal square at all, but 10…♘c6?! instead does not work out all that well: 11.♕b3 is awkward now for Black as ♗xf6, and the combined attack on b7 and d5 by the queen, is hard to answer. White is better) 11.♕c2 h6 12.♗h4. I prefer White here, as this IQP position is not very active for Black. He is quite OK, having free piece development, but it is still a static weakness to play against. 9.c3 c6 You probably already guessed that 9… c5 needs to be checked again. Against this …c7-c5 approach I like to just keep the tension, as 10.dxc5 ♘xc5 does not look ideal: 10.h3 ♗h5. This position is reached in Ntirlis and Aagaard’s book.

489

490

Chapter 23: Exchange Variation with 4…♘f6

They only mention 11.♕c2 now, but 11.♖e1 is stronger. This deceptively simple position is not so easy for Black to handle: 1) 11…♖e8 12.♗b5 is awkward for Black; 2) 11…♗d6 12.♕c2 is no improvement either. White is slightly better; 3) 11…cxd4 is the simplest approach: 12.cxd4 (White’s idea is ♕b3, and this is surprisingly di cult to meet) 12… ♗d6 (this removes the bishop from the attack by the e1-rook, thus making the black queen mobile enough to answer ♕b3 with …♕b6) 13.♕b3 ♕b6 14.♕xb6 ♘xb6 15.♘h4 ♗g6 (probably needed to defang ♘f5, so Black needs to give up the pair of bishops) 16.♘xg6 hxg6 17.♖ac1. White is better for the rest of the game with the bishop pair and the better mobilized pieces. 10.♕c2

copying White’s set-up – we have the capture on h7. Now we have a nal split of lines: C1) 10…♖e8 11.♖fe1 Inserting the rook move has not helped Black. There is still no obviously constructive move, as 11…♗d6 and 11…♕c7 fail to 12.♗xh7+. 11…h6 12.♗f4 ♘h5 Otherwise, White goes h2-h3 and ♖e3, doubling on the e- le with nagging pressure. 13.♗e5

C11) 13…♘xe5 14.♘xe5 This is a worse version of 10…h6 11.♗f4 ♘h5 12.♗e5 ♘xe5 13.♘xe5, as we will see. 14…♗e6? This fails now. Better is 14…♘f6 15.c4, keeping White on a small advantage. 15.♗h7+ ♔f8 16.♘g6+ fxg6 17.♖xe6 And White wins. This is the position that shows why it is useful to delay ♖e1 in favour of quick queenside development. Now Black is not able to play …♗d6 or …♕c7,

491

C12) 13…♗g5

492

Chapter 23: Exchange Variation with 4…♘f6

D) 5…♗d6

This was played in Vishy Anand-Hou Yifan, Douglas 2017. White obtained a typical edge and managed to win a ne game. This game is our sample game 19. C2) 10…h6 This is probably best. The alternatives give White more hope for a small edge. 11.♗f4 ♘h5 White had h2-h3 followed by ♖e3 on the agenda, so this is best. 12.♗e5 ♘xe5 13.♘xe5 ♗e6 14.♖fe1 ♗d6 15.♘df3

This symmetrical move is certainly the most principled one. Black claims that copying White’s moves will equalize, as there is little tension in the position. Logic tells you that he is probably correct in this assumption, but we will see that there are some problems for Black to solve. I think that White can play for an edge in some cases, in others it’s just very close to equal with almost zero losing chances. 6.♕e2+! This move poses the most problems for Black. Without this check White’s opening would be rather toothless, as Black would equalize by simply copying everything.

White has easy moves to play, like g2g3, and the doubling of rooks with ♖e3 and ♖ae1. It is not much, but I’d prefer White here.

D1) 6…♗e7 This is de nitely better than 6…♕e7. It looks like a concession, moving the bishop again, but we need to keep in mind that the white queen will probably need to move again as well, when …♖e8 is played at some point. 7.0-0 0-0 Black needs to play this now, otherwise 8.♖e1 will keep the king in the centre.

493

494

Chapter 23: Exchange Variation with 4…♘f6

8.♖e1 ♖e8 After 8…♗d6, 9.♗g5 is very annoying. Black has no good way to break the pin and 9…♖e8 is not possible either. 9.♗g5 Very concrete play. We threaten 10.♗xf6. 9…♗e6 Very reasonable, blocking the e- le. 10.c3 Clearing the way for the queen to go to c2. 10…♘bd7 11.♕c2 h6 12.♗h4 c6 13.♘bd2 If anything, White still has a slight pull. But it is not much, obviously. There are still some interesting possibilities. Let’s check a possible line. 13…♖c8

14.♖xe6!? A remarkable exchange sac, suggested by Stock sh. We can’t ask AlphaZero what it thinks about it, but it seems very valid. 14…fxe6 15.♗g6 ♖f8 16.♖e1 ♘e8 This seems to be best. 17.♗g3 ♖f6 18.♘e5 With good compensation. I’ll stop here, as it gets a bit deep. But feel free to 495

check this for yourself if you are curious. This exchange sacri ce shows that even simple-looking, symmetrical positions can have remarkable resources. D2) 6…♗e6

Also better than 6…♕e7 and the choice of Fabiano Caruana and Alexander Grischuk when they faced 6.♕e2+, as played by Magnus Carlsen against both of them. 7.c4 7.♘g5 is also possible, but I won’t argue with Carlsen’s choice. 7…dxc4 I also checked the yet untested 7…0-0 8.c5 (we have to play this, otherwise Black would take on c4 twice and get a good anti-IQP position, because our important light-squared bishop is traded) 8…♗e7 9.0-0 b6 10.♗e3 with a complicated position. As mentioned, this is all untested, so it is unclear how deep you will want to analyse it. I’ll stop here, being content with a ‘let’s play chess’ position. 8.♗xc4 ♕e7 9.0-0

496

Chapter 23: Exchange Variation with 4…♘f6

Trading on e6 does not lead to much. This would be a somewhat inferior version of a similar line arising from 4.♘f3 ♗d6 5.c4 dxc4 6.♗xc4 ♘f6 7.♕e2+ ♕e7 – please compare. 9…0-0 10.♘c3

With a complicated position that is easier for White to play.

D21) 10…h6 This was Fabiano Caruana’s choice in a classical time control game against Carlsen in Baden-Baden 2017. 11.♖e1 Carlsen played 11.d5 instead, going for the endgame. 11.♖e1 tries to exploit White’s better development, with queens on the board. 11…♗xc4 12.♕xc4 ♕d7 After the other retreat 12…♕d8 13.♘e5 White has an initiative based on the more active pieces. Moves like ♕b3, ♗f4 and ♖ad1 come next. This is not terrible for Black, but he has to be careful. 13.♘e5 ♕f5 14.h3 This is a good idea, intending g2-g4 to push the black queen to an unattractive square. 14…♘bd7 15.g4 ♕h7 16.♗f4

497

D22) 10…♖e8 This was played by Grischuk in a blitz game against the World Champion in late 2017. 11.♗g5 h6 12.d5 Okay, I (and Stock sh) try to improve on Carlsen’s play. The World Champion played the natural 12.♗h4 instead – remember, it was a blitz game. 12…♗g4 13.♗xf6 Leading by force to an interesting endgame. 13…♕xe2 14.♗xe2 gxf6

15.♗d1

498

Chapter 23: Exchange Variation with 4…♘f6

White prepares 16.♘h4, trying to exploit the weak f5-square. Let’s check some of Black’s possibilities at this junction: 1) 15…♗f5 16.♘d4 ♗d3 17.♗a4 and White is slightly better in this endgame; 2) 15…♗xf3 16.♗xf3 is also slightly better for White. Ideas like ♖c1 and ♘b5 are possible, with ♗g4 being very annoying in some positions – …♖c8, for instance, is prevented that way; 3) 15…♘d7 is not great, but I like to show the idea White has planned: 16.♘h4 ♗xd1 17.♖fxd1 and ♘f5 will be very annoying. D3) 6…♕e7

This is the most popular move, but it is actually the easiest to play against. White keeps a nagging edge. 7.♕xe7+

moves to increase the pressure. It is not necessary to analyse a maze of lines here. Let’s look at a good example game for some moves: 10…♖e8

11.♘bd2 This is the strongest move here. I have played 11.h3 in a recent league game, which also led to a nice position, but … ♗g4 is not worth preventing. 11…♗g4 12.♘e5 ♗e6 After 12…♗h5 13.h3, intending 14.g4 and going after the black bishop on g6, White is slightly better for sure. 13.♘b3 a5 14.♘c5 ♗xc5 15.dxc5 ♘bd7 16.♘xd7 ♗xd7 17.f3 And White was clearly better in Magnus Carlsen-Georg Meier, Pro Chess League rapid 2017;

D31) 7…♗xe7 8.♗f4! This is very awkward to meet. 8…c6 9.0-0 0-0 10.♖e1 This position is just very uncomfortable for Black to play. White is quicker on the e- le and has easy

D32) 7…♔xe7 This does not equalize either; we are quicker on the e- le. 8.0-0 ♖e8 9.♖e1+ ♔f8 10.♖xe8+ ♔xe8 11.♘c3! This is very important. Usually, ♘c3 does not make that much sense in the Exchange French if Black has …c7-c6 available. Here it is important to gain

499

500

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

this tempo to get on the e- le more quickly. 11…c6 12.♗d2 h6 13.♖e1+ ♔f8 14.♘e5 ♘bd7 15.f4

CHAPTER 24

Exchange Variation with 4… ♗g4/4…♘c6 1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3

This position was reached in CarlsenAkobian, Pro Chess League rapid 2017, which is our sample game 21. Black faces a thankless task defending this position. Conclusion Black goes for symmetrical play in this chapter. It’s clear that White needs to be precise to pose any problems at all. In fact there are some ideas that make the symmetry more interesting, in particular 6.♕e2+ is a line that will score heavily. Quite simply, many opponents won’t know it and will just go 6…♕e7, trading into an uncomfortable ending. It is also important to note the move order subtleties of 5…♗e7. If you know your move orders, as explained here, you will have a slight pull – if you play a harmless move order Black will get completely stale equality.

501

Black’s probably most ambitious moves are examined in this chapter. Both aim for the …♘c6/…♗d6/…♘ge7/… ♗g4(f5) set-up that we like to avoid. A) 4…♗g4

This is probably Black’s most ambitious line on move 4. He pins the white knight immediately and has a set-up 502

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

with …♘c6/…♕d7, and even …0-0-0, in mind. The good news is that White will be better with precise and (very important!) concrete play. Just playing schematic moves leads to nothing. 5.h3 ♗h5

The only logical option. Taking on f3 just gives up the bishop pair, and there is no other decent square to retreat to. 6.♕e2+! This is what I referred to as ‘concrete play’. This disruptive check is the reason why 4…♗g4 is considered risky for Black. A1) 6…♘e7?! This leads to big trouble quickly. 7.g4! Please avoid 7.♕b5+? ♘bc6 8.♕xb7 ♖b8 etc., giving Black a clear advantage! 7…♗g6 8.♘c3 Black’s development is completely clogged up now. White, in the meantime, has easy moves with ♗f4, 0-0-0, ♘e5, h2-h4 etc.

This essentially gambits a pawn for some compensation. It is not enough, but not terrible either. 7.♕b5+

A21) 7…♘c6? This does not turn out well. 8.♕xb7 ♘b4 Against 8…♗b4+ 9.c3 ♘ge7 White needs to avoid pitfalls: 10.♗b5! (instead, 10.cxb4 ♗xf3 11.gxf3 a6! is to be avoided) 10…0-0 11.♘bd2! ♖b8 12.♕a6 and White keeps the pawn without any problem. 9.♕b5+ ♔f8 10.♕a4 Black does not have enough for the pawn. A22) 7…♘d7 8.♕xd5 ♘gf6 9.♕b3 So we have moved our queen four times in the last four moves. Black has very nice development now, but he has also given up the important central pawn on d5. I don’t think there is enough compensation, but don’t underestimate Black’s uent piece play! 9…c5 10.♗e2

A2) 6…♗e7 503

504

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

I think White is better here. Note that we are behind in development, but we have no weaknesses. This is very important, as without any weaknesses to attack the black activity might peter out. A3) 6…♕e7 Not the most natural-looking move, but the alternatives are worse. 7.♗e3 This avoids the queen trade and prepares 0-0-0. White is better here due to his edge on the kingside, with g2-g4 to come. 7…♘c6 8.♘c3 0-0-0 Of course, 8…♗xf3? 9.♕xf3 ♘xd4 10.♕d1 is excellent for White, regaining the pawn on d5 with a huge advantage. 9.g4 ♗g6 10.0-0-0 This is the position that arises most often after 4…♗g4, if White goes for the most critical line, like we do.

Now Black has tried some moves that we need to check: A31) 10…♘b4 11.♖d2 The move 11.a3, trying to trap the knight, is also possible, but terribly complicated. 11.♖d2 just keeps a comfortable advantage with not much to learn beforehand. 11…f6 12.a3 ♘c6 With similar play as in the main line 10…f6. A32) 10…♘f6?! This simply ignores White’s idea to play 11.♘e5. 11.♘e5 ♕e6 After 11…♘xe5? 12.dxe5 ♕xe5 13.f4 ♕e8 14.f5 White wins a piece and the game. 12.♗g2 We complete development rst and cover e4 some more. f2-f4 is on the agenda next. 12…♘e4 13.♘xe4 ♗xe4 14.f3 ♗g6 15.f4 And White has a nice initiative with easy play. A33) 10…♕e8

505

506

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

Preparing 11…♗b4. The queen on e7 was blocking Black’s development. 11.♘e5 ♗b4

lines clearly show the dangers of a knight coming to e5. 11.a3 This is a useful move in the long run. Black might have …♘b4 or …♗b4 available after a later queen move. White also intends to go ♘d2, after which …♘b4 cannot be answered with ♖d2 anymore. 11…♕d7 12.♘d2 This prepares the push of the f-pawn and the knight heads towards b3, and possibly c5 later. 12…f5

12.♘b5! This is stronger than the usually played 12.♘xg6. 12…♔b8 The pawn on a7 was under re, so this makes a lot of sense. 13.c3 ♗e7 14.♗f4 And Black is under severe pressure. A34) 10…f6

This is the most popular move, covering e5 and opening some squares for the bishop on g6. The alternative

507

This was played in the stem game of this line, Kasparov-Short, Tilburg 1991. Black has other options, but White will play with ♗g2, f2-f4, ♘b3 in all cases. He is better because of his kingside space advantage and better mobilization, as Black needs some moves to get the kingside untangled. 13.♘f3! This engine suggestion improves White’s play considerably. Kasparov played 13.♘b3 and obtained a better game nevertheless, but heading back with the knight to use the weakened 508

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

e5-square is a lot more forceful. Black is in trouble: 1) 13…♘f6 14.♘e5 ♕e6 15.♘b5! ♔b8 (15…♘xe5? already loses, the following line being the main point: 16.dxe5 ♕xe5 17.♘xa7+ ♔b8 18.♘c6+ bxc6 19.♗a7+) 16.♗f4! ♘e8 17.♗g5 ♘f6 18.♗g2 and Black is facing a huge white initiative; 2) The engine likes 13…a6 relatively best, but it’s not an easy choice to make over the board: 14.♘e5 ♕e6 15.♗g2 ♘f6 16.♗g5 with a very nice advantage for White.

The most natural move, sticking to the initial plan of …♗d6/…♘ge7. We need to check two rare moves though: 1) 5…a6. Black invites White to damage the pawn structure in exchange for the bishop pair. I see as the main problem for Black his lack of development: 6.♗xc6+ bxc6 7.♕e2+! (yet another case of this annoying queen check) 7…♕e7 (this is the only move, basically. Against …♘e7 or … ♗e7 just 0-0 and ♖e1 will be extremely annoying) 8.♗e3.

B) 4…♘c6

Black intends the typical …♘c6/… ♗d6/…♘ge7 set-up, but wants to have …♗b4+ still available in case White pushes 5.c4. 5.♗b5 This is the only testing move. White needs to have something concrete against …♗d6/…♘ge7. A lame move like 5.♗d3 leads nowhere after 5… ♗g4. 5…♗d6

White follows up with 0-0, ♘bd2 and c2-c4. White is at least slightly better here; 2) 5…♘ge7 has been played by some strong players, but it clogs up Black’s development quite substantially: 6.0-0 a6 7.♗a4 b5 8.♗b3 g6 9.c3. This is just a sample sequence for this rare line. White is better with just normal developing moves as Black’s structure is weakened and he is still not developed harmoniously. 6.c4!

509

510

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

This is one point of the 5.♗b5 sortie. This move initiates a sequence of moves that leads to a very unconventional pawn structure. 6…dxc4 7.d5 a6 8.♗a4 b5 9.dxc6 bxa4 10.0-0 ♘e7

This sequence is almost forced after 6.c4 dxc4. It’s a rather uncommon pawn formation with Black’s queenside completely shattered – he’s got an extra pawn, though, that White still needs to recover, and the bishop pair. This, however, is quite often just a shortlived point as ♘b1-d2-c4xd6 is a common idea. 11.♘bd2 This is not yet very popular, but it is better than 11.♕a4. I think that White retains a small edge after this move. It is nothing special, but it’s the kind of position you will win sometimes and should never lose. After already having completed my analysis I noticed that Axel Smith recommends the same approach in his e3 Poison book – an additional stamp of approval. B1) 11…c3

511

A frequently seen idea is to return the pawn while damaging the opponent’s structure. White has a stronger move than taking, though. 12.♘e4! An important point, and a key di erence to 11.♕a4, when White needs to take on c3 with the pawn in similar scenarios. 12…0-0 Here 12…cxb2? is too risky: 13.♗xb2 0-0 14.♘xd6 (instead, 14.♕d4 is strong as well, but taking on d6 is very onesided. It’s your call: to be better in an endgame or in a middlegame) 14… ♕xd6 15.♕xd6 cxd6 16.♘d4 White is a lot better due to the dangerous cpawn. 13.♕xa4 ♕e8 This was played in the only sample game, Meier-Blübaum, Dresden 2017. White kept a nagging edge in this game. The capture 13…cxb2 is better now than on move 12, but still dangerous: 14.♗xb2 ♖b8 (this is the only playable move, surprisingly. Black needs the tempo against the b2-bishop. Against other moves, ♖ad1 is very 512

Chapter 24: Exchange Variation with 4… ♗g4/4…♘c6

problematic) 15.♘xd6 ♖xb2 16.♘xc8 ♕xc8 17.♖fd1. Black still needs to be precise here as the c6-pawn is dangerous and hampers Black’s coordination. 14.♘xd6 cxd6 15.♘d4 Of course – we need to keep c6 as long as possible. 15…♗b7 16.♖e1 ♗xc6 17.♕a3 ♕d7 18.♕xc3 White is slightly better again. We have the better structure and Black will most likely not be able to keep the bishop pair.

This happened in the mentioned MeierBlübaum game. White is slightly better. He’s got the better pawn structure and will get a very nice bishop on c3/d4. B2) 11…0-0 12.♘xc4 ♘xc6 13.♕xa4 ♗b7 14.♖d1 This natural move is a novelty. White simply intends to take on d6, with a comfortable pull. 14…♘e7 15.♘d4 We need to avoid 15.♘xd6?! in view of 15…♗xf3. 15…♗c5 This is the cold-blooded engine move, not worried about knight discoveries. 16.♗e3

513

Conclusion Black is ambitious in the lines examined here. Quite frequently, black players are used to the scenario that lower-rated opponents like to play for a draw with the Exchange – so they learn a line that tries to break the symmetry early on. However, as we have seen in the analysis, Black is not equalizing at all with 4…♗g4 – rather the contrary. If you remember the lines with 5.h3 and 6.♕e2+, you will be better. 4…♘c6 is better in comparison, but I still feel that rather White is calling the shots here. Make sure to study this line until the end. It is usually chosen by theoretically prepared people who will know more than just up to move 11.

514

Sample games

 

Sample games Game 19 FR 1.4 (C42) Viswanathan Anand 2794 Hou Yifan 2670 Douglas 2017 (9)

1.e4 e5

Wait, isn’t this supposed to be a game to illustrate the French Defence? Well, it is… just see what happens! 2.♘f3 ♘f6 3.♘xe5 d6 4.♘f3 ♘xe4 5.d3 ♘f6 6.d4 d5 And we have reached the French Exchange! This position is identical to the one after 1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3 ♘f6, just with an added move count (it took 6 moves instead of 4). Against the Petro I suggest to play 3.♘c3, but the line chosen by Anand in this game is a good alternative that ts with our repertoire. There is no 515

worrying deviation along the way. Black could avoid the very natural 6… d5 and play something like 6…♗e7 instead, but this is answered by natural development like 7.♗d3, followed by 00, ♗g5 etc., quite similar to the scheme used in the game. Quite often a later … d6-d5 will transpose to the French Exchange after all. 7.♗d3 ♗e7 8.0-0 0-0 9.♖e1 As mentioned in the theoretical section, I think that this very naturallooking move is not the most precise one. I recommend to play ♗g5/♘bd2 rst to avoid the Black set-up shown in the note to move 11. 9…♗g4 10.♗g5 ♘bd7 11.♘bd2

11…♖e8 I think Black should play 11…c6! rst in order to get in …♗d6 and …♕c7. After 12.c3 ♗d6 13.♕c2 ♕c7 White now has the extra move ♖e1 in a symmetrical position. Unfortunately it is not possible to make anything out of it. There is no control over e5 in particular. Most likely there will be trades on the e- le, and nothing much happens in a very equal position. 516

Sample games

12.c3 c6 13.♕c2 h6 It is too late now for …♕c7 or …♗d6 due to the pressure on h7. One important small detail to mention: playing …h7-h6 has a lot of relevance for the centre as …f7-f6 is hardly possible anymore now. A knight appearing on e5 will be very annoying this way, as there is no good way to force it to back out. This motif we will witness in the game. 14.♗f4 No …♕c7/…♗d6, please! 14…♘h5 15.♗e5 ♗g5 Black seeks to exchange pieces to ease his defence. 16.h3

16…♗xf3 16…♗e6 was probably better, after which White only retains a slight pull: 17.♘xg5 hxg5 18.♕d1 ♘f4 19.♗xf4 gxf4 20.♘f3 and White is better due to his superior pawn structure and better bishop. Still, I feel this is better for Black than the game continuation. 17.♘xf3 ♗f4 18.♗h7+ ♔h8 19.♗xf4 ♘xf4

517

20.♗f5 Simple-looking but very strong play by Anand. First he has forced the black king on a worse square, now the bishop is ready to be traded on d7, paving the way for the knight to jump to e5. 20…♕c7 20…♘f6 is answered by 21.♘e5, followed by ♖e3, ♖ae1 etc. Black is in deep trouble already. 21.♗xd7 ♕xd7 22.♘e5 ♕e6 The only move to keep f7 protected and prevent ♕f5 at the same time. 23.♖e3 ♔g8 Black needs to get out of the knight checks. The alternative 23…♘g6 leads to a truly depressing endgame: 24.♘xg6+ ♕xg6 25.♕xg6 fxg6 26.♖ae1 ♖xe3 27.♖xe3. Not all rook endings are drawn. 24.♖ae1

518

Sample games

The threat is 25.♘d3, which is very tough to meet. 24…♕c8? The engines recommend 24…♖f8 as a better defence. At least it keeps material equality. White still is much better after 25.♕b3, but needs to avoid 25.♘d3? ♕g6! and Black is ne due to 26.♖g3 ♘xd3!. 25.♘xf7 Hou Yifan certainly saw that one coming, but probably thought it was better to give up a pawn and get rid of the mounting pressure. However we will see the coming endgame is probably hopeless. 25…♖xe3 26.♖xe3 ♔xf7 27.♖f3 ♔g8 28.♖xf4

519

It’s not just an extra pawn, but also the more active pieces that give White a close to winning advantage. 28…♕e6 29.♕f5 Anand has correctly evaluated that the rook endgame will be excellent for White. 29…♖e8 30.♕xe6+ ♖xe6 31.♔f1 White’s plan is to simply push his extra pawn majority on the kingside while Black has no counterplay. There are no open les for the black rook to seek activity. 31…a5 32.♖f3 ♖g6 33.a4 ♖d6 34.♔e2 ♖d8 35.♔d3 ♖b8 36.♖f5 b5 37.axb5 ♖xb5 38.♔c2

The king keeps the queenside protected. There’s no need to rush. 38…♖b7 39.f4 ♖b8 40.g4 a4 41.♖e5 ♔f7 42.f5 ♖a8 43.♖e6 ♖c8 44.♖e5 ♖a8 45.h4 a3 46.bxa3 ♖xa3 47.♔b2 ♖a8 48.g5 hxg5 49.hxg5 The plan executed in the game is impossible to prevent. 49…♖b8+ 50.♔c2 ♖h8 51.♖e6 ♖h5 52.g6+ ♔f8 53.f6 1-0 Anand won this game with very simple, yet very e ective means. The tempo advantage in these symmetrical 520

Sample games

positions often translates into a better endgame that you can press for a win and ultimately convert to a full point. Make sure to study the move order details. I think that Anand actually did not play the most precise move order. Small details matter a lot here. Game 20 FR 1.4 (C01) Pavel Ponkratov 2605 Davit Magalashvili 2506 Amol rapid 2018 (3)

1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3 ♗d6 5.c4 The only try for an advantage. It is important to understand that 5.♗d3 in this position is toothless due to 5…♘e7! and Black easily gets the bishop traded on f5. First 5…♘c6 and then …♘ge7 leads to the same scenario. 5…dxc4 6.♗xc4 ♘f6 7.♕e2+ This is the move I recommend as it poses the most problems for Black in the resulting endgame. If you like to keep the queens on, you may also play 7.♘c3, which is part of our repertoire anyway via 5…♘f6 6.♘c3 dxc4 7.♗xc4. 7…♕e7 8.♕xe7+ ♗xe7 8…♔xe7 usually signals Black’s intention to go for a …♗e6 based setup, accepting a pawn on e6. This is probably the approach I would go for, but it still remains a somewhat unattractive task to defend. Please see the theoretical section for details. 9.0-0 0-0 10.♖e1 ♗d6 521

White has gained some time and has active pieces. In particular the c4bishop promises to be a long-time annoyance for Black. 11.♗g5 I recommended 11.♘c3 in the theoretical section. Both moves are of about equal strength, keeping the slight pull that we have. 11…♗f5?! Black does not mind the doubled pawns. I think this is a questionable decision. There are two problems associated with doubled pawns: the pawns themselves might become weak and oftentimes there are weak squares around the pawns. This is the more serious issue here as the f5-square is an excellent post for White’s pieces – the knights in particular. In order to avoid the doubled pawns Black should choose 11…♘bd7. Let’s examine some more moves: 12.♘c3 h6 13.♗h4 c6 14.♗b3.

522

Sample games

analysis diagram

A useful prophylactic move, anticipating …♘b6. Black’s task is not easy here: 14…b6? (this move gets punished quite severly. She should have gone for 14…♖e8 instead, which is only slightly worse for Black) 15.♘e5 ♗b7 16.♘xd7 ♘xd7 17.♗e7 ♗xe7 18.♖xe7 ♖ad8 19.♘e4 ♗a6 20.♘d6 ♘f6 21.♘xf7 ♖xd4 22.♘d8+ ♔h8 23.♘e6 and Black resigned in GuninaGara, Gibraltar 2017. 12.♗xf6 gxf6 13.♘c3 ♘c6 The pawn on f6 is vulnerable, thus 13…♘d7 makes some sense.

It does not solve the problems though. After 14.♘e4 I don’t see a fully satisfactory move for Black. It is one of the most annoying ideas that Black faces: White takes on d6, followed by ♘h4 and ♖e7. 14…♗b4 avoids the capture, but 15.♖e3 keeps the initiative. Even ♖b3 is now on the agenda in many lines. 14.♘e4 ♔g7 The principled alternative was 14… ♗xe4 15.♖xe4, hoping to draw with opposite-coloured bishops. White’s pieces are still more active and the doubled pawn/f5-square remains a problem. One sequence to check is 15… ♖fe8 16.♖xe8+ ♖xe8

analysis diagram

analysis diagram

523

17.♗b5! and White gets rid of the opposite-coloured bishops scenario and further weakens Black’s pawns. 17…♖b8 (a clever move, intending … ♘e7) 18.♗xc6 bxc6 19.♖b1. The black rook now makes sense on b8. This position is very one-sided though, as only White can try to win the endgame against Black’s wrecked pawns. 15.♘xd6 cxd6 16.d5 ♘e5 17.♘d4 524

Sample games

This was White’s intention when taking on d6. He needs the knight to stay on the board in order to exploit the weak f5-square. 17…♗d7

18.♗e2 I was somewhat surprised by this move. I expected the more direct 18.♗b5, immediately trading the defender of f5. White is better here, for sure, but the game move does not spoil anything. After 18.♗b5 it continues 18…♗xb5 19.♘xb5 ♖fd8 20.♘d4 and Black su ers in a passive defence. 18…♖ac8 19.f4 ♘g6 19…♘c4 looks a bit tougher: 20.b3 ♘b6 21.♗f3. The b6-knight at least attacks d5 from this post, while on g6 it does nothing at all. The structural de cits don’t go away though, White’s edge is long-term. 20.g3

20…♖fe8 Here 20…a6 at least prevents 21.♗b5. One interesting scenario to check is 21.♖ac1 ♖xc1 22.♖xc1 ♖c8 23.♖xc8 ♗xc8. White has excellent winning chances in this minor-piece endgame: 24.♘c2 ♘e7 25.♘e3. This square is a good post for the knight. It controls all the key squares, like f5, and is ready to switch to the queenside via c4 if appropriate. 21.♗b5 ♖xe1+ 22.♖xe1 ♗h3 Instead, 22…♗xb5 23.♘xb5 wins a pawn immediately for White. 23.♔f2 ♖c5

24.b4! Tempting Black to grab on d5. 525

526

Sample games

24…♖c3 We need to check 24…♖xd5 25.♔e3 ♘e5 (a nice trick, but it does not help; 25…♘e7 26.♗e2 ♘c6 leads to the same position) 26.♗e2 ♘c6 27.♘c2 (amazingly, Black’s rook is almost trapped in the middle of the board) 27…♘e7 28.g4. Now …♘f5 is prevented and ♗c4 is the idea to pick up the rook, with excellent winning chances. 25.♖e8 ♖c8 26.♖e3 White wants to keep the rook on the board, hoping to exert more pressure. However, after 26.♖xc8 ♗xc8 27.♗d3 a6 28.♘c2 the endgame is clearly favourable for him as well, very much comparable to the note to move 20. 26…♔f8 27.♗e2 ♗d7 28.♖a3 a6 29.b5 This initiates the coming forced sequence. 29…♗xb5 30.♗xb5 axb5 31.♘xb5 ♘e7 32.♘xd6 ♖c2+ 33.♔g1 ♘xd5 34.♘xb7 ♘c3 35.♘d6

35…♘xa2? This loses the game. Black had to try 35…♘e2+. He is still in trouble after

527

36.♔f1 ♘d4 37.♖d3 ♘e6 38.a4, but it is not over quickly. 36.♘f5 The mate threat now forces Black into passivity, leading to material losses soon. 36…♔e8 37.♖a8+ ♔d7 38.♖a7+ ♔e8 39.♘d6+ ♔f8 40.♖xf7+ ♔g8 41.♖xf6

There is this infamous sentence, ‘The rest is a matter of technique’. Well, this sentence is applicable here, but that does not mean it is easy to play. The king on the rst rank is a problem. 41…♘c1 42.♘f5 ♘d3 43.♖e6 ♘f2 44.♔f1 ♘g4 45.♘h6+ ♘xh6 46.♖xh6 ♔g7 47.♖h5 ♔g6 48.g4 h6 49.h3 ♖a2 50.♔g1 ♖a4 51.♖f5 ♖a2 52.♖b5 ♖a4 53.♖b6+ ♔g7 54.f5 h5 54…♖a2 keeps the king passive, but White already wins without it: 55.h4 ♖c2 56.♖g6+ ♔h7 57.g5 hxg5 58.hxg5 and the pawns will promote soon. 55.♔f2 ♖a3 56.♔g2 ♔h7 57.♖g6 ♖b3 58.♖g5 hxg4 59.hxg4 1-0 The endgames after the early queen exchange are not as harmless as you might think. This game is a good example of how small advantages are 528

Sample games

piling up. The pawn structure damage and the resulting outposts for the white pieces proved to be very tough to handle. Game 21 FR 1.4 (C01) Magnus Carlsen 2838 Varuzhan Akobian 2645 Pro Chess League rapid 2017 (5)

1.e4 e6 2.d4 d5 3.exd5 exd5 4.♘f3 ♘f6 5.♗d3 ♗d6 6.♕e2+ ♕e7 7.♕xe7+ ♔xe7 7…♗xe7 8.♗f4 also gives White a small edge, as played in Carlsen-Meier in the same season of the Pro Chess League. 8.0-0 ♖e8 9.♖e1+ ♔f8 10.♖xe8+ ♔xe8

secures the outpost on e5, as we will see. 11…c6 12.♗d2 h6 Black’s position is somewhat uncomfortable. It is obvious that White intends to post his knight on e5, where it is very di cult to get rid of. Controlling g5 is useful for Black, but it is probably not the most urgent move to make. Engines suggest 12…♘a6, but after 13.♖e1+ ♔f8 14.♗xa6 bxa6 15.♘e5

analysis diagram

11.♘c3 This is an important nesse! The knight threatens to jump to b5 and provokes …c7-c6 by Black. ♗d2 and ♖e1 will follow, keeping a pull. The knight on c3 is often misplaced if there is a pawn on c6, but here speed of development is more important as it

529

Black has some trouble with the spoiled queenside structure: 15…♖b8 16.♘d1 c5 (instead, 16…♗e6 17.♗f4 ♘e8 18.c3 is also slightly better for White) 17.dxc5 ♗xc5 18.♗f4 ♖b6 19.♘d3 ♗d6 20.♗xd6+ ♖xd6 21.f3. This is a sample line to show how the play might continue. After …c6-c5 Black has weak pawns, and not much to show for them. 13.♖e1+ ♔f8 14.♘e5 ♘bd7 15.f4

530

Sample games

This is the set-up that White had in mind when going for 11.♘c3. The knight on e5 is extremely annoying and cannot be expelled easily. White has weakened the light squares though – this might be a point that Black can play for. It is notable that engines only give White a small plus, but I think for a human player Black’s position is awkward to handle. 15…♘b6 16.h3 Playing g2-g4 soon and gaining even more space is certainly useful. This move prepares the kingside expansion. 16…♗d7 It’s useful to check 16…h5.

This stops the immediate g2-g4, but it feels weird to play the h-pawn again. I mainly want to show White’s best reply as it is instructive: 17.f5!, still keeping g2-g4 in mind, but at the same time preparing ♗f4 to activate the bishop with a tempo (♘g6+ is on). White has the easier game for sure. This was a better option compared to the game, but again: a human does not play …h7h6 and then …h6-h5 shortly afterwards. 17.g4 ♗e8 Akobian tries for some counterplay with …c6-c5, but he needs to play this rst. The immediate 17…c5 does not convince due to 18.♘xd7+ ♘bxd7 19.g5 and White opens the game for his bishop pair. 18.♔f2 c5 19.dxc5 ♗xc5+ 20.♔f3 20.♔g3 was slightly more precise. The main idea is to play 20…♘fd7 21.♘f3, avoiding trades and keeping the pressure on. 20…♘fd7 21.♘xd7+ ♗xd7 22.b3 ♗c6 Black has improved his situation in the last moves. It is still easier to play White, but it is not much. 23.♔g3 ♖d8 24.♘e2

analysis diagram

531

532

Sample games

24…d4 This advance was not strictly necessary, but in general it is desirable to advance the IQP. Black gets a nice c6h1 diagonal and the d5-square, but White has some additional options as well. 25.b4 ♗e7 26.b5 ♗e8 27.♔f3 ♘a4

In this endgame only White can be better, but a rook endgame with equal pawn count almost always has some drawing margin. 34…♔e7 35.♔e4 ♔f6

28.♘g3 Surprisingly (at least to me), Carlsen allows 28…♘b2 and further trades. I would have expected 28.♖b1, aiming for ♗b4, for example after 28…♘c5 29.♗b4. This still looks slightly better for White as the d-pawn is somewhat exposed and the king is much more active. 28…♘b2 Now the bishop on d3 can no longer avoid the trade. 29.♘f5 29.♗f1? d3 is bad for White. 29…♗c5 30.♖b1 ♘xd3 31.cxd3 30.♖b1 covered b5 and now serves as preparation of ♗b4. 31…♗d7 32.♗b4 ♗xb4 33.♖xb4 ♗xf5 34.gxf5

36.♖a4 It would be a horrible blunder to play 36.♖xd4?? ♖xd4+ 37.♔xd4 ♔xf5 and White is lost! If you are unsure about evaluations like that, I suggest you analyse the position a bit on your own, not using an engine. Next you play the position against the engine in a rapid or blitz time control, and nally you analyse your game being helped by the engine and if possible with a training partner or teacher. 36…♖a8 The engine suggestion 36…♖e8+ 37.♔xd4 a6 38.bxa6 bxa6 39.♖xa6+ ♔xf5 should be a better chance to draw, but Akobian’s move is not terrible either. 37.♖xd4 ♖e8+ 38.♔f3 ♖e7 Otherwise ♖d7 will be too much to bear. 39.♖d5 h5 40.a4

533

534

Introduction

PART V

1.e4: other moves Introduction

40…b6? Black should have gone for the more active 40…♖e1! 41.♖d7 ♖a1 42.♖xb7 ♖xa4 43.h4. Here White is better, of course, but it’s not an easy win at all – rather, a tough technical task. 41.h4! Black is in zugzwang. Moving the rook will allow White to make decisive progress – and there is nothing else available. 41…♖c7 42.♔e4 ♖c1 42…♔e7 43.♖e5+ ♔f6 44.♖e8 is an easy win for White as well. 43.♖d7 ♖e1+ 44.♔d5 ♖f1 45.♖xa7 ♖xf4 46.♔c6 ♖xh4 47.♔xb6 White is just much quicker on the queenside. It is funny that often in those pawn race scenarios exactly one tempo decides. Here it is not that close, and quickly assessed as winning for White. 47…♖b4 48.a5 h4 49.♖a8 ♔xf5 50.a6 g5 51.♖h8 ♔g4 52.a7 ♖a4 53.♔b7 f5 54.a8♕ ♖xa8 55.♔xa8 f4 56.b6 f3 57.b7 f2 58.♖f8 And Black resigned.

535

This part deals with the second- tier defences to 1.e4. Some of these are played quite regularly on the club level, and also used by masters, but are rare guests in the classical time control top tournaments. Top players usually prefer the more respected openings discussed in parts I-IV, as they are considered to be slightly more reliable. We indeed have more chances for an opening advantage in this part, compared to more classical openings. Let’s get an overview of the next chapters. Chapter 25: 1…d6 2.d4 ♘f6 3.♘c3 g6

536

Introduction

against the Pirc. Quite often, a quick … ♘f6 by Black will indeed transpose to the Pirc. In general, the whole complex with …d7-d6/…g7-g6 is full of transpositions. Don’t be too intimidated by this fact – White has a strong centre and will have at least a decent position with common sense moves. The Modern Defence is covered in Chapter 26. This is the Pirc Defence. Black allows White to take space in the centre and gives us a wide choice of set-ups. 4.♗e3 is my choice against the Pirc. It is a exible approach that does not yet commit to a central pawn formation. Chapter 25 provides you with the details.

Chapter 27: 1…d6 Black may also play this move, but avoid the bishop placement on g7. The main part of this chapter deals with 2.d4 ♘f6 3.♘c3 e5.

Chapter 26: 1…g6

This is the ‘Modern Defence’. I always felt this is one of the weirdest opening names. Will it still be called that in 50 years time? Anyway, after 2.d4 ♗g7 3.♘c3 Black has tried many o beat moves that we’ll check brie y, but most of the time 3…d6 is played. Again we’ll go 4.♗e3, using the same set-up as 537

This is the modern way to reach the closed Philidor’s Defence. There are move order issues with the traditional 1.e4 e5 2.♘f3 d6, as examined in Chapter 2. Against the Philidor, I suggest playing the main line. It is easy to remember, and quite comfortable for White, though Black is certainly solid. Chapter 27 deals with the Closed Philidor and other rarities after 1…d6. Chapter 28: 1…d5

538

Introduction

approach against this line is analysed in Chapter 29. Chapter 30: 1…d5 2.exd5 ♕xd5 3.♘c3 ♕a5

The Scandinavian Defence. Black immediately de nes the central pawn structure by forcing the trade of his dpawn against White’s e-pawn. The coming pawn structure also has similarities with other openings, in particular the Caro-Kann. Chapter 28 shows how to handle the Scandinavian sidelines. Chapter 29: 1…d5 2.exd5 ♕xd5 3.♘c3 ♕d8

This is the old main line. White needs to play energetically against it to get an advantage – just making ‘natural’ moves does not promise much. For details how to attack against 3…♕a5 please see Chapter 30. Chapter 31: 1…♘f6

Currently the most respectable move. Black is content with a slightly worse position and ‘play chess’. We aren’t much against this approach, as we want to do the same thing with white and be better at the same time! Our 539

One of the most original openings – Alekhine’s Defence. It is actually not easy at all to get an edge against this provocative move that aims to lure the 540

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

white central pawns forward in order to attack them later. My choice against it is to play 2.e5 ♘d5 3.d4 d6 4.c4 ♘b6 5.exd6. White has good chances for an edge with this line while still being reasonable in terms of study time. Chapter 31 shows how to play against the quirky knight move.

CHAPTER 25

Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6 1.e4 d6 2.d4 ♘f6 3.♘c3 g6

Chapter 32: Rare black replies on the first move

Finally there is Chapter 32, containing the rarest of the rare lines. Most of these moves can be easily met on general grounds, but some are so dubious that some study may net quick points. Please enjoy this chapter, as it features some attractive refutations of bad play.

541

4.♗e3 Against the Pirc and Modern, I recommend a dangerous sideline that will often catch black players unawares. I think this line, involving ♗e3 and h2-h3, is a serious test for Black’s cause anyway, so an unprepared black player might struggle with his defensive task quite quickly. White still keeps his options open regarding pawn structures, but most of the time White will play with f2-f4, after having seen Black’s set-up. Note that quite often we will play f2-f4 right after ♗e3 and h2h3. It should be mentioned that starting with f2-f4, usually on move 4, is possible as well. So why don’t we start with that? There are two good reasons: the rst one is that ♗e3, rst, makes it more di cult for Black to strike at the centre with …c7-c5, a 542

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

strategy that is very common after the early f2-f4. The second and more practical reason is that black players will have likely looked at the early f2-f4 more, as it is perceived as the main line. A) 4…♘g4 This is rare and not very good, but useful to check. 5.♗g5 ♗g7 6.h3 Instead 6.♗e2 is also ne, obviously, but 6.h3 ts well with our repertoire. 6…h6 7.♗c1 ♘f6 8.♗e3

And we reach our main line with Black having the extra move …h7-h6. This is useful for White, as now ♕d2 will come with tempo or prevent Black from castling altogether.

B1) 5…♗g7 6.f4 This transposes to 4…♗g7 (Line D of this chapter) B2) 5…♘bd7 6.♘f3! We don’t play like this often, but here it is strong! Instead, 6.f4?! is met with 6… b5, which is quite annoying. The immediate 6.♘f3 is strong, because it threatens e4-e5 immediately. 6…e5 7.♗c4 ♗g7 8.dxe5 dxe5 9.0-0 0-0 10.a4 b6 11.♕d3 And White is slightly better in this position due to his more active pieces. B3) 5…b5?! This principled move is quite dubious as it ignores one point of White’s play. 6.e5!

B) 4…a6 Black delays the natural …♗g7 in order to accelerate his queenside play. 5.h3 With the knight on f6 this move makes a lot of sense, in order to prepare f2-f4.

543

544

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

B31) 6…b4 This just blunders a pawn at the end. 7.exf6 bxc3 8.fxe7 ♗xe7 9.bxc3 And White is a pawn up and has the better position. B32) 6…dxe5 7.dxe5 ♘fd7 7…♕xd1+ is even worse than the immediate retreat of the knight: 8.♖xd1 ♘fd7 9.♘d5 ♔d8 10.e6 fxe6 11.♘f4 and White has a very good, indeed almost winning position. 8.♕f3 c6 9.♕g3 And White has excellent prospects. f2f4, ♘f3 are easy moves to make, and Black is in trouble. B33) 6…♘fd7 Relatively best, but still rather uninspiring for Black. 7.♕f3 A very direct and strong move. Black is already struggling to keep the game going. 7…♘b6 8.a4 b4 9.♘e4 Not much fun for Black. Let’s have a look why 9…♗g7 fails: 9…♗g7 10.♘g5 Threatening both f7 and a4-a5. 10…♗e6 The only move, but Black’s position is already beyond repair. 11.♘xe6 fxe6 12.♕g4 ♕d7 13.♘f3 And White is close to winning. C) 4…c6 Black delays …♗g7, because he expects White to go for ♕d2 and ♗h6. In this case he would save a tempo. This is one of the upsides of a exible move like 4.♗e3. We still have many options, and 545

Black needs to take these into account and sometimes will play against ideas that are not that high on our agenda in the rst place. 4…c6 is a good idea against the mentioned ♕d2 idea, but less valuable versus a set-up with f2-f4. 5.h3

C1) 5…b5?! Going …b7-b5 early on with …♘f6 already played is often quite dubious because of e4-e5. This is no exception. 6.e5 b4 The compliant 6…dxe5 7.dxe5 ♕xd1+ 8.♖xd1 ♘fd7 9.f4 leads to an excellent position for White. Black’s queenside is weakened, and e4 is a great square for a knight or a bishop. 7.exf6 This initiates a forced sequence, leading to a nice advantage. 7…bxc3 8.fxe7 ♕xe7 9.bxc3 ♗h6 10.♕e2! And White is better. The point is: 10…♗a6 11.♕d2 This strange queen manoeuvre actually helps White to castle quickly. 11…0-0 12.♗xa6 ♘xa6 13.♘e2 ♗xe3 14.♕xe3 ♕xe3 15.fxe3 d5 16.♔d2 546

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

And it’s an extra pawn that is, admittedly, not easy to convert – but it’s an excellent opening result. C2) 5…♗g7 6.f4 And we have transposed to 4…♗g7 (Line D of this chapter). C3) 5…♕a5

6.a3! This is usually a good reply to the queen sortie. It is not that clear what the queen is actually doing on b6, because you can never take on b2 now due to ♘a4. White will continue with f2-f4 etc.

Just threatening the e4-pawn and trying to disturb White. However, the moves Black tries to provoke are not hurting us in any way… moreover, at the end of the day it’s not clear if Black’s queen makes sense on a5. 6.♕d2 b5 7.♗d3 We need to cover e4 as …b5-b4 was a threat. 7…♘bd7 8.a3! The immediate 8.f4?! is not that great due to 8…b4 9 ♘ce2 c5. 8…♗g7 9.f4 0-0 10.♘f3 And White is better, with e4-e5 on the agenda. C4) 5…♕b6

547

C5) 5…♘bd7 6.f4 I like this active set-up after Black has committed to a slow build-up with … c7-c6. That being said, there is nothing wrong with playing ♘f3 and a2-a4 for a slower game, if you like some variety. 6…e5 Quite frequently Black goes for 6…b5 in these lines. We should go for 7.a3. Note that 7.e5? b4! is not advisable. It is completely ne here to spend a move on a2-a3, securing our broad pawn centre: 7…♘b6 (this is quite a clever idea. Black intends to play …d6-d5 and, after White pushes e4-e5, to hop into e4. We need to address this point) 8.♗d3! (covering e4, so that the knight cannot go there) 8…♗g7 9.♘f3.

548

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

And White has the usual comfortable game. We will castle kingside and probably go for a kingside attack with ♕d1-e1-h4, for instance. This set-up is tough for Black to handle, as witnessed in our sample game 22, Testor-Bauer, Austria Bundesliga 2015/16. 7.♕f3 There is nothing wrong with 7.♘f3 either, but quickly going for 0-0-0 is tempting. 7…b5 The routine move 7…♗g7 is already too slow against White’s aggressive setup: 8.0-0-0 0-0 9.g4 and White’s position is much better and easier to play. 8.dxe5 dxe5 9.0-0-0 And again White is better. Black has a couple of moves now, but it is already a rather rare and speci c move order, so let’s check the most direct one: 9…b4 10.♘a4 ♕a5 11.fxe5 ♕xe5 12.♗d4 ♕e6 13.♔b1

549

With a very nice initiative for White. D) 4…♗g7 5.h3

The de ning move of our set-up against the Pirc/Modern. We prepare f2-f4 and prevent …♘g4 as a reply to e4-e5 as well. D1) 5…c6 6.f4 b5? This is wrong and gets punished quickly. Better is 6…0-0 7.♘f3, which leads to a position examined via 5…0-0 (line D41 in this chapter). 7.e5 ♘fd7

550

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

you to 6.g4!?, which is quite dangerous for Black! 6…e5 7.♕d2 White will castle queenside and go for ♗h6. I prefer White here, but Black is quite solid in this Philidor-type position.

8.h4! First h2-h3 and then h3-h4 is odd, but it does not matter. White has a fantastic position already. 8…h5 9.e6! fxe6 10.♗d3 ♘f8 11.♘f3 And Black’s defensive task will be tough, to say the least. White will just play ♘e4, c2-c3, and castle kingside to put pressure on e6 ultimately. Black has huge coordination issues. D2) 5…♘bd7

6.♘f3 A rare case where I like to avoid f2-f4. After 6.f4 Black has 6…c5. Not terrible for White, but this is quite OK for Black. If you want to play more aggressively than 6.♘f3, I would point 551

D3) 5…a6 Going for a quick …b7-b5. If Black is aiming for this type of set-up I think he should rather play the Modern and delay …♘f6. White has no problems getting a nice game. 6.f4 0-0 Instead, 6…b5 is already quite dubious, which does make 5…a6 look somewhat silly: 7.e5 ♘fd7 8.♕f3 ♘b6 9.0-0-0 and with g2-g4 coming White is in excellent shape. 7.♕f3 There is nothing wrong with 7.♘f3 either, but this prevents …b7-b5 altogether, and 0-0-0 and g2-g4 are obvious next moves. 7…e5 8.0-0-0 exd4 9.♗xd4 ♘c6 10.♗f2

And White is much better with his extra space and attacking ideas on the kingside. 552

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

D4) 5…0-0 6.f4 Just sticking to the usual set-up. Practice has seen a lot of moves here that I actually think are also ne for White. Even the simple 6.♘f3 or 6.♕d2 are good and won’t equalize for Black.

Now we have the nal split for this chapter. D41) 6…e5 Quite a radical try at solving the opening issues. It does not seem to work perfectly, though. 7.dxe5 dxe5 8.♕xd8 ♖xd8

9.fxe5 The only way to address this approach. Black has some issues now with 553

recovering the pawn. 9…♘fd7 10.0-0-0 ♗xe5 11.♘d5 White is better, with ♘f3/♗b5 to come. Black is not well coordinated. D42) 6…c5 This is an important move to check. Against f2-f4 based set-ups, this strike is often the right recipe, for instance after 1.e4 d6 2.d4 ♘f6 3.♘c3 g6 4.f4 ♗g7 5.♘f3, 5…c5! has the best theoretical reputation. One point of our set-up is that ♗e3 makes …c7-c5 more di cult to play. Here the c-pawn push is playable for Black, but it does not equalize, ultimately. 7.dxc5 Initiating a forced sequence. Note that 7.d5? is terrible due to 7…b5!, a typical tactic, the point being 8.♗xb5 ♘xe4!. 7…♕a5 8.♗d3 We need to cover e4 urgently. 8…dxc5

This is important to note. Black does not get to take back with the queen on c5 due to our bishop on e3, a key di erence to the Austrian Attack with 4.f4 against the Pirc. 9.♗d2! 554

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

This is the important move to remember. We absolutely want to play e4-e5 to shut out the bishop on g7, but the immediate 9.e5 fails to gain an advantage against 9…♘d5. 9…e5 The only critical move, stopping the epawn. It loses quite some time, though, and we will be better here as well. If Black allows the e-pawn to advance, things are even worse: 1) 9…♘c6 10.e5 ♘d7 11.♘f3 is just nice for White, with the space and the awkward black bishops; 2) 9…♖d8? 10.e5 ♘fd7 11.♘d5 and White wins immediately. 10.fxe5 ♘fd7 11.♘f3 ♘xe5 12.♘xe5 ♗xe5 13.♕f3

And White is better. ♘d5 and 0-0-0 are coming.

7.♘f3 The most natural, but 7.♕f3 is a nice alternative if you like to go for a quick attack with 0-0-0 next. 7…b5 Or 7…♘bd7 8.e5 ♘e8 9.♗d3 and again, White is better, with a space advantage and good attacking chances. Queenside castling is a serious option here as h4-h5 will be very dangerous. 8.e5 The usual push against passive play. 8…♘d5 After 8…♘fd7 9.♗d3 leads to a standard, favourable position for White. You may continue with 0-0 and ♕e1-h4, or play for 0-0-0, if Black stays passive. 9.♘xd5 cxd5 10.♕d2

D43) 6…c6

555

556

Chapter 26: Modern Defence: 1…g6

CHAPTER 26

Modern Defence: 1…g6 1.e4 g6

The b5-pawn could not be taken due to 10…♕a5+. Now the pawn is threatened and White enjoys a nice edge with the black g7-bishop biting on the e5-pawn. Conclusion The Pirc allows us to take the central control with pawns, an invitation we should happily accept. However our move order with ♗e3 and h2-h3 rst is a bit more cunning than just playing a quick f-pawn push. One issue of these ‘non-contactopenings’ like the Pirc and Modern are the exible move orders. Don’t be too concerned. We are better in the centre and have some leeway. If you forget something, it should not be fatal at all. The most move-order-sensitive part of this chapter is line D42, where you should know the right sequence to secure an advantage.

The Modern Defence. The main di erence between this opening and the Pirc Defence is that Black does not play his knight to f6 early on. This makes a di erence in many ways. One important point is that White is not able to operate with an early ♗e3/ ♕d2/♗h6 manoeuvre. Black also retains the option of developing the knight to e7, or even to h6 in some cases. The downside of Black’s game is clearly his lack of development and neglect of central control. Against the Modern, we will opt for a similar set-up as against the Pirc. Note that whenever Black plays …♘f6 in the next few moves we will transpose to the Pirc. 2.d4 ♗g7 1) 2…♘f6 is the North Sea Defence! It is dubious, but White needs to know it and must avoid one single line. Let’s

557

558

Chapter 26: Modern Defence: 1…g6

have a look: 3.e5 ♘h5 4.♘f3 (the only important point to know about this line is that 4 ♗e2 d6! 5 ♗xh5 gxh5 6 ♕xh5 dxe5 leads, amazingly, to some compensation for Black. White should avoid this and should just develop normally) 4…d6 5.♗c4 d5 6.♗e2. White is better here with c2-c4/♘c3 to come. What is the knight doing on h5?; 2) 2…d6 3.♘c3 c6 (instead, 3… ♗g7 leads to the 2…♗g7 line, while 3…♘f6 is a Pirc. With 3…c6 Black further delays any development) 4.♗e3. I always recommend this set-up against …d7-d6/…g7-g6. It is always playable and easy to learn. Now it is hard for Black not to transpose, for example 4… ♗g7 or 4….♘f6 would transpose to Modern/Pirc standard move orders. An independent try is 4…b5 5.a3. We also use this approach in the Pirc move order to stop …b5-b4. We will continue with f2-f4 and go h2-h3, if Black plays …♘f6, preventing any …♘g4 irritations. From here play will likely transpose elsewhere, as it feels silly for Black to delay normal moves like … ♗g7 and …♘f6 any further. 3.♘c3 Now 3…d6 is the most popular move, but Black has tried about everything here.

An interesting move order. Black might go for a Caro-Kann/Modern mix, but he also still has the option to play …d7-d6. 4.h3 I understand that this move looks a bit mysterious, but it has a point. We like to play a ♗e3 based set-up, mostly with h2-h3 inserted early to defend against …♘f6-g4 ideas. If we go 4.♗e3, Black may respond with 4…d5, which presents us with a move order problem, as we don’t play an early ♗e3 against …c7-c6/…d7-d5 set-ups (see the CaroKann chapter for this). So we start with h2-h3 instead of ♗e3. 4…d5 Instead 4…d6 5.♗e3 transposes to 3… d6 below. 5.♘f3 This position I have examined via the Caro-Kann move order. Please check that chapter for details.

A) 3…c6 B) 3…c5

559

560

Chapter 26: Modern Defence: 1…g6

This move intends to transpose to either a Benoni after 4.d5 or a Sicilian after 4.♘f3. 4.dxc5 This avoids the transpositions, and is strong. Black is in some trouble now. 4…♕a5 Trying to recoup the pawn. Black has tried 4…♗xc3+. Yes, our structure won’t be pretty – but we are excellent in other regards. The dark squares will be a problem for Black and as long as there is a pawn on c5 there are development issues with the c8bishop: 5.bxc3 ♕a5 6.♕d4 (this move, followed up by the next, secures White’s good position) 6…♘f6 7.♕b4 ♕c7 8.♘f3 and we are a lot better, as discussed on move 4: bishops count and the ♗c8 is not pretty. 5.♗d2 ♕xc5

6.♘d5 This is the sequence to play after 4… ♕a5. Black is in trouble, as we will see. 6…♘a6 Black needs some safety on c7. 7.♘f3 e6 8.b4 Starting concrete play. 8…♕f8 That was forced. What a weird position! 9.♘c3 White is in great shape here. The pawn on b4 is poisoned. 1) 9…♕xb4 10.e5 gives White a great attack; 2) 9…♘xb4 10.♘b5 ♘a6 11.e5 and Black should not survive.

561

562

C) 3…d5

Chapter 26: Modern Defence: 1…g6

A bit odd, but not terrible for Black. 4.exd5 Taking with the knight instead does not win a pawn due to 4…c6, and d4 will fall as well. 4…♘f6 5.♗c4 We keep the pawn for the time being, but are ready to return it later for a good position. 5…♘bd7 Black plays for just recapturing the d5pawn. 6.♘f3 We should simply develop and be happy with the sightly more comfortable position. 6…♘b6 7.♗b3 a5 8.a4 0-0 9.0-0 ♘bxd5 10.♘xd5 ♘xd5 11.h3 And we are slightly better with the more active pieces. This structure is very similar to g6/c6/d5 set-ups. D) 3…a6

This looks quite exotic, but without any direct contact between both armies Black is quite exible with move orders. …a7-a6 is part of Black’s agenda quite often, and here starting with this ank action early on is possible. In fact, this quick queenside play prevents the set-up we are going for against 3…d6 4.♗e3 a6 – let’s have a look. 4.♗e3 b5 5.♗d3 Here 5.f4 ♗b7 prevents 6.♗e2, which is our choice against 3….d6 4.♗e3 a6 5.f4 b5 6.♗e2. Black is one move faster with the pressure on e4. Instead of 5.f4 I suggest to develop sensibly with 5.♗d3. Instead, there are some other possibilities, for instance even 5.h4 looks interesting. 5…♗b7 6.a4 Clarifying matters on the queenside. 6…b4 7.♘ce2 ♘f6 8.♘g3 And with ♘f3 and 0-0 coming, White completes his development. With the better central control we should be better. E) 3…d6 Finally we have arrived at the main line.

563

564

Chapter 26: Modern Defence: 1…g6

4.♗e3

We play like this vs the Pirc Defence as well, so it makes a lot of sense as it keeps the repertoire consistent. In fact, any …♘f6 in the next moves will transpose into the Pirc. E1) 4…♘d7 5.f4 And we will transpose after Black goes …a7-a6 or …c7-c6. 4…♘d7 has little independent value. One set-up that may occur via various move orders is the Hippopotamus (or just ‘Hippo’), which involves …e7-e6/…d7-d6, knights to d7 and e7 and a Queen’s bishop anchetto. Let’s have a look at this via the 4….♘d7 move order, though Black might also start with … e7-e6, …a7-a6, …b7-b6 etc. – it is more of a set-up than a concrete line. 5…e6 6.♘f3 ♘e7 7.♗d3 This is the set-up that White should go for against the Hippo. We build a strong centre and support it with the bishops. You may also go for queenside castling as it happened in our sample game 23, Kotronias-Hebert, Montreal 2002. 7…b6 8.0-0 ♗b7 565

9.f5! In this concrete position, f4-f5 is already quite strong. Note that Black may go for di erent move orders, so you are not guaranteed to get precisely this on the board. But against the Hippo this is a standard idea. Another plan might be ♕d1-e1-h4, especially if Black has started with an earlier kingside build-up and castled already. 9…exf5 10.exf5 0-0 11.♕d2 And White has good attacking chances on the kingside. E2) 4…a6

Black prepares …b7-b5 immediately. One key idea for Black is to play …d7d6, …♘d7 later, and then …c7-c5, to 566

Chapter 26: Modern Defence: 1…g6

strike at White’s centre. After a subsequent …c5xd4 Black often reaches a very decent Sicilian structure. It is our goal to prevent that. 5.f4 b5 6.♗e2 I like this idea. White is preparing ♗e2f3, and later maybe e4-e5 to exchange the bishops on the long diagonal. 6…♗b7 Black has also tried 6…♘d7, in order to meet 7.♗f3 with 7…e5!. We should go for 7.e5 to prevent that. After 7…♘h6 8.♗f3 ♖b8 this line is assessed as ne for Black by Tiger Hillarp Persson in his book The Modern Tiger. I agree that it is a very double-edged position, and Black has chances to counter-attack the centre, but still I’d rather be White after normal play like 9.♘ge2. After 6…b4, 7.♘a4 is the best square. The knight looks a bit odd on the rim, but it is good to prevent …c7-c5: 7… ♗b7 8.♗f3 ♘f6 9.e5 ♘d5. Now Hillarp Persson only mentions 10.♗d2, but I like 10.♗f2 more. I think White is better here with a later ♘e2 and c2-c4. 7.♗f3 ♘d7 8.e5 ♕c8 9.♕e2

White is more comfortable here due to the extra space and the limited scope of the bishop on g7. E3) 4…c6 Preparing either …b7-b5 or …♕b6. This move allows us to gain lots of central space. 5.f4

The move h2-h3 is only useful when a knight, on f6, is threatening to jump to g4, harassing our bishop. So we might as well push f2-f4 immediately after Black has declared his intentions to play somewhat passively, with …c7-c6. E31) 5…♘f6 6.h3 This transposes to our line against the Pirc Defence. Viktor Moskalenko only considers 6.♘f3 in his book The Perfect PircModern, which gives Black more options based on …♘g4 jumps. E32) 5…b5 6.a3 A tempo worth spending. Black’s slow set-up gives White enough time to play this. Our set-up involves a future ♘f3,

567

568

Chapter 26: Modern Defence: 1…g6

♗d3, and probably e4-e5, pushing in the centre. 6…♘d7 7.♘f3 If Black plays …♘g8-f6 now, we would transpose to the Pirc Defence. We have easy moves anyway, involving ♗d3, 00 etc. You should look at the Pirc Defence lines as well, they are very similar in nature. E33) 5…♕b6

And after a later e4-e5 the black knight is stranded on g4. White is better here with a huge centre Black has little counterplay against. Conclusion The Modern and the Pirc are similar in nature, and a lot of the Chapter 25 conclusion applies to this chapter as well. There are, however, some quite speci c lines. In particular you should look at the lines with 3…a6 and 3…d6 4.♗e3 a6. The early …a7-a6 lines are attractive to many players as they gain space on the queenside and help to pressurize the centre by threatening … b5-b4 and placing the bishop on b7. Make sure to study these lines, they will occur frequently after 1…g6.

6.a3 A good reply to the queen sortie. White covers the pawn tactically and does not misplace any piece, as would be the case after 6.♖b1. Note that 6.♕d2 ♕xb2 is not completely convincing, and should be avoided, especially as there is this good alternative. 6…♘f6 6…♕xb2?? 7.♘a4 is the point of 6.a3 – the square is now taken away from the black queen. 7.♘f3 Here, 7.h3 ♘h5!? is somewhat messy and not so clear. The direct 7.♘f3 is ne, however. 7…♘g4 8.♗g1

569

570

Chapter 27: Philidor and rare lines after 1…d6

CHAPTER 27

Philidor and rare lines after 1… d6 1.e4 d6 2.d4 ♘f6 There are some transpositions possible instead: 1) 2…g6 3.♘c3 transposes to 1…g6 2.d4 d6 3.♘c3, examined in Chapter 26; 2) 2…e5 3.♘f3, transposing into the old Philidor move order, is best here. Instead, taking on e5 and trading queens is also somewhat better for White, but I prefer to transpose. This position we examine via 1.e4 e5 2.♘f3 d6 3.d4. 3.♘c3

Now the Closed Philidor with 3…e5 is the main part of this chapter, but we need to check di erent options rst. A) 3…c6

571

This is the Czech Defence, also called the Pribyl Defence after its main ‘developer’, IM Josef Pribyl. 4.f4! This is the most testing move here by far. It is quite logical to grab the full centre once Black has committed to … c7-c6, as Black is not able to strike with … c7-c5 in one go. In fact, Black must react here immediately, otherwise White’s centre will be too strong. 4…♕a5 The mentioned immediate reaction. Black attacks the e4-pawn by pinning. 4…g6 is a second-rate Pirc Austrian Attack. We usually only play f2-f4 after going h2-h3 and ♗e3 rst, but here Black has already committed to the passive …c7-c6, so White’s game is easy to play: 5.♘f3 ♗g7 6.♗d3 0-0 7.00 and White is comfortable. Pushing e4-e5 to gain space and restrict the g7bishop makes sense, as a future plan. ♕e1-h4 is another standard idea that looks promising. 5.♗d3 e5 Black needs some foothold in the centre. 6.♘f3 572

Chapter 27: Philidor and rare lines after 1…d6

White will retain a nice edge if he remembers not to take on e5, but instead to maintain the tension. A1) 6…♘bd7 This looks a lot like a second-rate Philidor. White has the f-pawn involved, which should be a nice asset. 7.0-0 ♗e7 8.♘e2 This is an idea that I played in a league game when my opponent surprised me with this opening and I had to nd moves over the board. White intends to support the centre with c2-c3, and the knight has a good square on g3, eyeing f5 later. A2) 6…exd4 7.♘xd4 g6 Black has given up the centre in order to attack the e-pawn later. 8.0-0 ♗g7 9.♔h1 This is mandatory after f2-f4. You can’t play with queen checks and pins in the air all of the time. 9…0-0

573

10.f5! This is the most popular move here, and it seems to be strong indeed! White aims for ♕d1-e1-h4 and a cavemanstyle attack. Well, those caveman clubs might be crude, but they hurt a lot! White’s approach is extremely dangerous. A3) 6…♗g4

7.♗e3 Never trade on e5, is the key thing to remember. It would give Black’s f8bishop undeserved space. 7…♘bd7 8.0-0 ♗e7 9.h3 ♗h5 10.♕e1 White is better with his space advantage. There is also a small trap

574

Chapter 27: Philidor and rare lines after 1…d6

here, typical of the …♕a5 vs ♕e1 setup: 10…0-0? 11.♘d5! ♕d8 12.♘xe7+ ♕xe7 13.♘h4 This is even worse for Black than it looks. He’s close to busted already.

C) 3…e5

B) 3…♘bd7

Black intends to go for the Philidor, but likes to avoid the possible endgame after 3….e5. This endgame is not that scary for Black, actually (we avoid it), but maybe undesirable if Black is ambitious. 4.♘f3 This will just transpose to a regular Philidor. This is my recommendation to keep things simple when facing a rare line. If you actually suspect, before the game, that your opponent is going to play 3….♘bd7, it is worth checking out the aggressive 4.g4!?, which is quite dangerous for Black. It is quite complicated, though, so I feel it is easier to transpose to calmer lines, if you are surprised by this system. 4…e5 5.♗c4 And we have transposed to a regular Philidor, as examined below.

This is the modern move order to reach a Philidor Defence. Black intends to hold the centre (the e5-pawn does not take on d4) for the next few moves and only later decide how to proceed. In the traditional Philidor move order, 1.e4 e5 2.♘f3 d6 3.d4, both 3…♘d7 and 3… ♘f6 don’t manage to reach this set-up (please check these lines as they are bad for Black but still frequently played at club level). 4.♘f3 Trading on e5 and exchanging queens is not very dangerous for Black. The king in the centre is no huge issue with queens traded. Instead, we will simply put pressure on e5 by developing naturally. A general, important point: taking on e5, later, is only good if there is a very concrete reason for it, like material gains or a very favourable piece trade etc. d4 is a better pawn than d6 – it is not an equal trade. Also, note that the f8-bishop will bene t enormously from trading as well. Another important point is that …c7c6, by Black, should always be answered with a2-a4, stopping …b7-b5

575

576

Chapter 27: Philidor and rare lines after 1…d6

advances. Just allowing Black to expand on the queenside is usually not a good idea. 4…♘bd7 Again there are possible ways to transpose to a di erent chapter: 1) 4…exd4 5.♘xd4 we have looked at via the traditional Philidor move order 1.e4 e5 2.♘f3 d6 3.d4 exd4 4.♘xd4 ♘f6 5.♘c3 (Chapter 2); 2) 4…♘c6 5. ♗b5 transposes to Chapter 5. 5.♗c4

The bishop is best placed on this diagonal, eyeing the f7-pawn. 5…♗e7 The most popular move by a mile. Black prepares castling, which is very sensible play. Sometimes Black plays 5…h6 to prevent 6.♘g5, but this was not a serious threat anyway. It is mostly an unnecessary commitment, but given the slow/closed character of the position, we cannot refute this move straight away. One active, but risky, option for Black could be to keep the king in the centre and push …g7-g5 for an attack, an idea we will check now: 577

6.0-0 c6 7.a4 (stopping …b7-b5, an automatic reaction) 7…♕c7.

Black delays castling in order to potentially go for …g7-g5, to attack our king. This would be possible after playing h2-h3, so we should keep in mind not to rush with this move, which is very standard in these positions under normal circumstances (Black going …0-0 is ‘normal’). 8.♘h4! points out the drawbacks of Black’s set-up. Black has no satisfactory move here to address the knight jump to f5: 8…g6 9.a5! (we intend to play f2f4 without being annoyed by …♕b6 or …♘b6 at some point) 9…♗g7 10.f4 (with a terrible position for Black; one issue is that he cannot castle) 10…0-0? 11.dxe5 dxe5 12.♘xg6! and over. Another rarely seen option for Black is 5…exd4 6.♘xd4 (instead, taking with the queen is also ne, and transposes to a position that we have looked at via the 1.e4 e5 2.♘f3 d6 move order) 6… ♘b6 7.♗e2! and we transpose to 5… ♗e7 6 0-0 ♘b6, examined below. 6.0-0

578

Chapter 27: Philidor and rare lines after 1…d6

7…exd4 8.♘xd4 Now there is no …c7-c5, harassing the bishop on b3. 8…0-0 9.a4 a5 10.f4 Grabbing more space in the centre and preparing to cover e4 with ♗f3 without blocking the f-pawn. 10…♖e8 11.♗f3 ♗f8 12.b3

Now there are three entirely di erent options for Black: concrete play in the centre with C1, going for a risky kingside operation with C2, or playing C3, the solid main line. C1) 6…♘b6 This move is a product of the computer age – it was not played before 2005, according to my Megabase 2018. I guess it was never examined before as it does not seem to logically t with Black’s idea to hold the centre, initiated by …♘bd7. However, as usual concrete lines are more important than general considerations. Let’s have a look. 7.♗e2! This is a very important move to remember. The bishop is not that great on e2, and usually we would avoid this post, but here the issue is that the natural alternative, 7.♗b3, is actually worse due to concrete reasons. Black plays 7…exd4 against it, when 8.♘xd4 is bad due to 8…c5!, and 8.♕xd4 gets hit by a set-up involving …a7-a5, … ♘fd7 and …♗f6, which is surprisingly e ective. Black has no problems at all after 7.♗b3. 579

White has a slight advantage due to his space in the centre. After centralizing and consolidating the centre pawns White may plan kingside play with g2g4 next. C2) 6…c6

Black may delay or omit castling by playing this move. Play will be similar 580

Chapter 27: Philidor and rare lines after 1…d6

to lines we have examined via 5…h6. 7.a4 Always prevent …b7-b5. 7…♕c7 8.♖e1 h6 9.♗a2 This move is very common in the Philidor structures. It is a useful improvement for White, stepping out of possible tempo moves like ….♘b6. It also allows ♕d3(e2)-c4 in some cases. 9…♘f8 Of course, Black could castle to return to more conventional play (line C3 below). He also has 9…g5.

And White is better, with annoying pressure on Black’s position. C3) 6…0-0 7.♖e1

Universally useful to support the centre.

The most aggressive way to play this opening, but it weakens too much: 10.♕d3! (this prepares both ♕c4 and ♘c3-e2-g3, aiming at the f5-square) 10…♘f8 11.♘e2 ♘g6 12.♘g3 and White is a lot better, with many weaknesses to exploit. 10.♘h4! An important resource for White, aiming for the f5-square. You need to check if any …♘xe4 tactics work, but here this is clearly not the case as 11.♖xe4 covers the knight on h4. 10…♘e6 11.♘f5

581

C31) 7…a6 This move gained some popularity in recent years. Black prepares …b7-b5 with the a-pawn, instead of the more common …c7-c6 based set-up. 8.a4 Mandatory! We need to stop the bpawn. 8…b6 Black’s plan, for the next few moves, is to go …♗b7, …c7-c6 and then ultimately …b6-b5. In some cases, even the immediate …b7-b5 is possible, as we will see. 9.♗a2 The usual improvement move that is commonly seen in the Philidor. 9…♗b7

582

Chapter 27: Philidor and rare lines after 1…d6

10.♕e2! Not a common move in these positions, but it is actually quite strong here because Black was threatening …b6-b5. For instance, after something like 10.h3, 10…b5! is possible. The queen looks at the b5-square, preventing this idea. At the same time we get ready for ♖d1, which is more powerful than usual as c7 is not available for Black’s queen. 10…c6 Freeing c7 and preparing the …b6-b5 push. 11.♖d1 ♕c7 12.♘h4! The f5-square is too inviting to resist this move. The position is somewhat better for White. 12…♖fe8 This is not that great, but I like to show a typical idea.

583

13.♕c4! An additional point behind 9.♗a2 and 10.♕e2. 13…d5 Not much else is available. 14.exd5 exd4 15.♘f5! With a very nice position for White. C32) 7…c6 8.a4

No huge bonus points for guessing that one. …b7-b5 has to be prevented. Now Black has tried many moves, and studying these lines is actually a bit tricky because the move orders are exible for Black. Moves like …♕c7, … ♖e8, …h7-h6, …a7-a5, …b7-b6 can be played in many permutations. In fact, 584

Chapter 27: Philidor and rare lines after 1…d6

we need to distinguish between two di erent black approaches here. The rst one is to continue to hold the centre with moves like …h7-h6/… ♖e8/…♕c7, and still try to play for … b7-b5 ultimately, with …b7-b6, …a7a6 as preparatory moves. The second approach is to take on d4 and initiate concrete play. This intention is usually signalled by the move 8…a5, which secures c5 for the d7-knight after a subsequent capture. If Black goes for the slow waiting game of Plan 1 we should keep the central tension and improve our central support with moves like ♗a2 and h2h3. Moving the knight to f5 is an important idea as well. Plan 2 requires more concrete play, as we will see. C321) 8…b6 Black aims for the slow set-up, having …b6-b5 in mind at the end. 9.♗a2 We have seen this before. The bishop retreats to a good square and prepares additional ideas like a queen manoeuvre to c4, in some cases. 9…a6 10.h3 Again, this is a useful improvement move. 10…♗b7

11.♘h4! Of course, f5 is a great square for the knight, but we will always need to check the possible tactics involving the e4-pawn. Here it works perfectly. 11…♘xe4? This fails. After ‘normal’ moves White is just better, with the knight hopping to f5 – and …g7-g6 allows the bishop to h6. 12.♘xe4 ♗xh4 13.dxe5! Certainly not 13.♘xd6? ♗xf2+! 14.♔xf2 ♕f6+. 13…dxe5 Black has some other moves, but it’s too late to get a good position anymore. 14.♘d6 ♕c7? After this, instead of being clearly worse, Black loses immediately. 15.♘xf7! And it’s over. The knight transfer to f5 is a key idea in many lines. Please check sample game 24, Parligras-Solberg, Rethymno 2010, for an excellent illustration of this idea. C322) 8…♕c7

585

586

Chapter 27: Philidor and rare lines after 1…d6

16.♘h6+ is just one of the threats. Black’s game is beyond repair.

A di erent way to start slow queenside play. 9.♗a2 b6 10.h3 a6 11.♘h4! No surprises here! Whenever possible this should be played. 11…♖e8 12.♘f5 ♗f8 13.♗g5 Black is under quite some serious pressure, already. 13…♗b7 Best is probably 13…h6, but it is still far from great for Black. White must be better with his far more active pieces. 14.dxe5! Only trade on e5 if you have a very good reason! Here we have one! 14…dxe5

C323) 8…h6 Black goes for the slow set-up again. 8…h6 takes control of g5, which is useful – but there is a downside, as Black weakens the light squares around the king, in particular g6. This is serious as the bishop is on the critical a2-g8 diagonal. 9.♗a2 ♖e8 10.h3 ♗a2 and h2-h3 are useful here as well, just like they are against other ‘waiting game’ set-ups with …b7-b6/…♕c7 etc. 10…♕c7 11.♗e3 ♘f8 12.a5 ♘g6 13.♕d3

And White is certainly better, but as usual Black is quite solid. You need to build up against the Philidor patiently, slowly improving and pushing Black back step by step. C324) 8…a5

15.♕f3 587

588

Chapter 27: Philidor and rare lines after 1…d6

This is usually the start of the …exd4 plan. With 8…a5 Black gives up on playing …b7-b5 and instead secures the c5-square for the knight, after capturing on d4. 9.♗a2 exd4 It’s also worth examining 9…h6. Combining …a7-a5 with waiting is not so common. Our play is quite standard for waiting game set-ups: 10.h3 ♖e8 11.♗e3 ♕c7 12.♕d2 (a good move, but instead, 12.♕e2 or 12.♘d2, improving central support, is also ne. Note that 12.♘h4?! is not that promising here as f5 is covered after the reply 12…♘f8) 12…♘f8 13.♖ad1 and White must be somewhat better here. 13…♗e6 is not that great, but it is instructive to see why: 14.d5 ♗d7 15.♘xe5! dxe5 16.d6 and White is much better as the dark squares in Black’s camp will be fatally weakened. 10.♘xd4 I also looked at 10.♕xd4. It is not bad either, but 10…♘c5, with …♗e6 to come, seems ne for Black. 10…♘c5 Black tries to get counterplay against e4. 589

11.♗f4 A natural developing move, looking at the d6-pawn as well. 11…♕b6 Now 11…♗e6 is less e ective compared to the position with the queen having recaptured on d4. White will gain the pair of bishops, with little compensation for Black: 12.e5 dxe5 13.♘xe6 ♘xe6 14.♗xe5 with a comfortable game for White. 12.♕d2 ♖e8 Vladimir Barsky recommends 12… ♗e6 in his book The Modern Philidor Defence. Surprisingly, now 13.♘xe6 fxe6 14.e5 dxe5 15.♗xe5 ♔h8 is indeed ne for Black, as stated in the book. However, with 13.♗e3 White is still somewhat better. We are threatening to take on e6 now and will just consolidate further with f2-f3 if necessary. 13.h3

White is slightly better here, with his extra space and more active pieces. Conclusion The Closed Philidor is a quite playable line for Black. I have played it on 590

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

occasion as a surprise weapon as well. However if White knows how to handle the opening, Black will not equalize completely. It is important to keep the tension in the centre and try to play ♘h4 if appropriate. Instead of holding the centre, Black might also take on d4 at various points. Make sure to study these lines precisely, as they are more concrete, and details matter.

CHAPTER 28

Scandinavian: 1…d5 2.exd5, rare lines 1.e4 d5

The Scandinavian is quite popular at club level, but is not seen frequently in top events. What is the reason? Well, the top players usually just want to force a draw with black or, in some concrete cases, play for a win if the tournament situation demands it. The Scandinavian is unsuitable for either scenario: Black will often simply get a position that is just very slightly worse and that cannot be immediately drawn by some forced line. On the other hand it is not particularly suited for drumming up some crazy complications, either. An additional issue is that Black will give up the bishop pair frequently, which is an important factor for strong players. These considerations are not as relevant at club level, where Black will

591

592

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

usually get a playable, if slightly worse position, without much learning e ort. With white we will get a comfortable position against the Scandinavian most of the time, but don’t expect too large of an advantage – Black’s structure is quite solid. 2.exd5 A) 2…♘f6 Black intends to capture on d5 with the knight (most of the time). Now 3.c4 keeps the pawn but Black has 3…c6 and even 3…e6, both interesting gambit-style moves. The main move is, of course, 3.d4, but I think my suggestion 3.♗b5+ is just a bit better for White and avoids the murky lines with 3.d4 ♗g4!?. Let’s have a look. 3.♗b5+ A disturbing check.

always slightly better with this pawn centre. A1) 3…♘bd7 4.d4 Now 4.c4 is possible again, but 4…a6 5 ♗a4 b5!? is quite interesting for Black and should be avoided, in my opinion. 4…♘xd5 5.♘f3 c6 6.♗e2 Quite amazingly, this position often arises from the Alekhine Defence! The move order there is 1.e4 ♘f6 2.e5 ♘d5 3.d4 d6 4.♘f3 dxe5 5.♘xe5 c6 6.♗e2 ♘d7 6.♘f3. White is slightly better due to his space advantage and easier development. We have some easy moves to play next. 6…g6 Black may also develop without a anchetto, starting with 6…♘7f6 7.c4 ♘b6 8.h3 (a useful move to remember. Black is not allowed to go …♗g4, with a potential future capture on f3 to simplify matters and maybe weaken the d4-pawn in the process) 8…♗f5 9.0-0 e6 10.♘c3 ♗e7 11.♗e3 with a standard position that is somewhat better for White due to his space advantage. 7.0-0 ♗g7 8.c4 ♘c7 9.♘c3 0-0 10.♗f4

In this line all you need to know is that White will retreat the bishop to e2 soon and then go for a set-up with c2-c4, d2d4, ♘c3, ♘f3, 0-0 and often h2-h3 to prevent any …♗g4 ideas. White is

593

594

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

And again, White is better without any particular e ort. A2) 3…c6? This is an incorrect gambit, Black does not get enough play for the pawn. 4.dxc6 bxc6 After 4…♘xc6 5.♘f3 White is just a pawn up for nothing. 5.♗e2 e5 6.d3 ♗c5 7.♘f3 Again Black does not have enough for the pawn. Comparing this with the main line of the Two Knights Defence/Fried Liver is useful: 1.e4 e5 2.♘f3 ♘c6 3.♗c4 ♘f6 4.♘g5 etc. leads to a similar structure, but with White losing far more time with ♘f3-g5 and some bishop moves that are not played here. A3) 3…♗d7 4.♗e2 White is happy to have misplaced the black bishop on d7 and retreats to e2, at the same time preventing any …♗g4 ideas. 4…♘xd5 5.d4 ♗f5 Or 5…g6 6.c4 ♘f6 7.♘c3 ♗g7 8.♘f3 00 9.0-0 and again, simple moves have given White a comfortable game. 6.♘f3 e6 7.0-0 ♗e7

595

8.a3 This is the only spot, after 2…♘f6, where you need to remember a speci c move. Instead, 7.c4 ♘b4 is not terribly bad for us, but we can just avoid it with the useful move 7.a3. 8…0-0 9.c4 ♘b6 10.♘c3 ♗f6 11.♗e3 ♘c6 12.h3 And White keeps a small but stable edge. B) 2…♕xd5 I think this is probably better than 2… ♘f6. Against the knight move White gets a slightly better position without any e ort at all, as seen in part A) of this chapter. 3.♘c3 The most natural and probably best move. It has one drawback: White places the knight in front of the cpawn, so White does not manage to get the d4/c4 pawn duo early on (compare with 2…♘f6). I always had some interest in 3.♘f3, trying to delay ♘b1-c3 until White has played c2-c4 rst. However, nowadays this approach seems to be defanged by 3…♗g4 4.♗e2 ♘c6 etc. as also demonstrated in John Bartholomew’s Scandinavian Repertoire on Chessable. After 3.♘c3 Black has a choice of queen moves. We will play with d2-d4 and ♘f3 against all of them, in most cases followed by ♗c4. A common theme is that Black develops with …♗g4 or … ♗f5, against which we will often try to play with kingside expansion. 3…♕d6

596

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

This move enjoyed some sparks of popularity some years ago, but White seems to have worked out what to do by now. The most popular queen moves to a5 and d8 are covered in the following two chapters. There actually is a fourth queen move: 3…♕e5 check! This is actually not as stupid as it rst looks. Black will go … c7-c6 and …♕c7. This cannot be brilliant for Black, but it is still fairly solid. Note that after …♕a5 Black sometimes plays …c7-c6 and …♕c7 as well, which takes the same amount of moves. I suggest to play 4.♗e2 c6 5.♘f3 ♕c7 6.d4. Now Black has tried some moves, but I don’t think we need to check all of them. A key idea to remember is ♘e5 followed by ♗f4, for example: 6…♗f5 7.♘e5 ♘d7 8.♗f4 ♘xe5 9.♗xe5 with a nice initiative for White. 4.d4 ♘f6 Black sometimes starts with 4…c6 or 4…a6, but …♘f6 will be played soon, so these moves don’t have any independent value. 5.♘f3

After 5.♘f3 Black has a wide choice. Unfortunately, there is no simple, identical recipe that we can use against all of them. Let’s have a look at the details for each black try.

597

598

B1) 5…♗g4 The usual recipe against …♗g4 is to immediately question the bishop with h2-h3 and g2-g4: 6.h3 ♗h5 (after 6… ♗xf3 7.♕xf3 ♘c6 8.♗e3 e6 9.0-0-0 White is a lot better with his space and bishop pair) 7.g4 ♗g6 8.♘e5 and Black is under some pressure already. White threatens h3-h4-h5 to trap the bishop, usually preceded by ♗g2, to control e4 rst. 8…♘bd7 is the only move not leading to a serious disadvantage: 9.♘xg6 hxg6 10.♗g2 c6 11.d5 and White is slightly better due to the bishops in a wide open position. B2) 5…c6

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

With a lasting edge due to the bishops, a well-known issue in the Scandinavian, after the trade of the light-squared bishop for a knight. B3) 5…a6

It is useful to compare this to similar positions with the queen on a5 or d8. In those cases, we have the natural move 6.♗c4, allowing us to answer 6… ♗g4 favourably by taking on f7 (with the ♕ on d8) or h2-h3/g2-g4 (♕ on a5). Here, if 6.♗c4, 6…♗g4! is possible and we don’t get to play h2-h3 and g2-g4. The critical move now is not very natural, but it is strong: 6.h3! Just stopping …♗g4 and preparing to answer 6…♗f5 with 7♘h4!. 6…♗f5 After 6…g6 7.♗d3 ♗g7 8.0-0 0-0 9.♖e1 White should be a bit better, with the usual space advantage and easy moves to come, like ♗g5, ♕e2, ♖ad1 etc. 7.♘h4 Instead, 7.♘e5 is also good, but a bit more complicated to learn. Our move is simple and good. 7…♗g6 Black keeps the bishop with 7…♗e6, but 8.♗e3 g6 9.♗d3 is also nice and easy to play for White. 8.♘xg6 hxg6 9.g3

599

6.♗d3! Not very intuitive to me, but experience and the engines suggest it is the strongest move. A closer inspection shows why it is good. 6…♗g4 1) 6…b5? is one idea of 5…a6, but it fails miserably: 7.♘e4! ♘xe4 8.♗xe4 c6 9.a4 and this looks a lot like a win in under 20 moves in the making; 2) 6…♘c6 7.♘e4 will transpose to 6… ♗g4, most likely. 7.h3 ♗h5 8.0-0 ♘c6 9.♘e4 The key idea, trading the unimpressive knight on c3 for the vital colleague on f6. 9…♘xe4 10.♗xe4

600

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

CHAPTER 29

Scandinavian: 2…♕xd5 3.♘c3 ♕d8 1.e4 d5 2.exd5 ♕xd5 3.♘c3 ♕d8

White now has a great piece on e4 and is able to support the centre with c2-c3, if necessary. This position was reached in sample game 25, Anand-Demidov, Pro Chess League rapid 2018. White is already for choice, but studying this game will help to understand the arising positions much better. Conclusion One of the easier chapters to summarize. White is better in all lines, given that he knows some details. Most important is the 3…♕d6 line, when after 4.d4 ♘f6 5.♘f3 Black has a choice of moves. There is no move that equalizes, but you need to answer each one with a di erent set-up. If you remember this correctly, 3…♕d6 is a line to be happy to play against.

This is currently Black’s most respectable choice. Black just retreats the queen on the safest square and argues that the lost time is compensated by White’s slightly misplaced knight. In many cases Black gets very close to equality, but White, more often than not, keeps a slight edge. This is mostly due to Black giving up the bishop pair by trading on f3. 4.d4 ♘f6 1) Here, 4…g6 is a rare move, but it was played in a nice game that saw Fischer dismantling this set-up in a very forceful way: 5.♗f4 (any normal developing set-up with ♗c4/♘f3/♗g5 etc. is comfortable for White, but Fischer’s approach is strong) 5…♗g7

601

602

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

6.♕d2! ♘f6 (6…♕xd4? 7.♕xd4 ♗xd4 8.♘d5 ♗b6 9.♗xc7 is excellent for White) 7.0-0-0 c6 8.♗h6 and Fischer already had a promising attacking position against Robatsch at the Varna Olympiad 1962; 2) 4…c6 has a subtle point. If we go 5.♘f3 Black may now play 5…♗g4, and after 6.h3 ♗xf3 7.♕xf3 grab on d4! This is risky for Black, but I think just going 5.♗c4 instead is trickier for Black to handle anyway. After 5.♗c4 it is best for Black to just play 5…♘f6!, after which 6.♘f3 transposes to a line examined below via 4…♘f6 5.♘f3 c6 6.♗c4. We need to check however why 5…♗f5?! is dubious for Black.

603

This is the natural-looking move that is, in fact, rather questionable: 6.g4 (here, 6.♗xf7+!? ♔xf7 7.♕f3 leads to a very strong attack, but it is tough to analyse. It’s not a very practical line to learn for a rather rare move order, so I suggest something aggressive that is less dependent on move-by-move knowledge) 6…♗g6 7.♘ge2 (this is played to answer …h7-h5 with the annoying knight jump to f4) 7…♘f6 8.♘f4 e5 (otherwise h2-h4 will be very annoying to meet) 9.♘xg6 hxg6 10.♗e3 and White is comfortably better – the bishop on c4 is particularly strong. 5.♘f3 Black has a choice now between two lines that lead to entirely di erent play. Line A will get sharp quickly, but White is quite clearly for choice. Line B is the main line, solid for Black and better than line A – still not equal though. A) 5…c6 We need to mention that 5…♗f5 has little independent value. After 6.♘e5! c6 7.♗c4 it quickly transposes to positions analysed here.

604

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

6.♗c4! A perfectly normal move, but it is also clearly the strongest here. Black now has problems playing the bishop from c8. 6…♗f5 After 6…♗g4? 7.♗xf7+ ♔xf7 8.♘e5+ White wins. 7.♘e5 e6 8.g4 ♗g6 9.h4

A usual scenario, pushing on the kingside and threatening to trap the bishop. The main idea is actually f2-f3, rst, then h4-h5 ! 9…♗b4 9…♘bd7 10.f3! (Black is helpless against h4-h5 now) 10…♘xe5 11.dxe5 ♘d7 12.h5 and White wins material – the problem bishop is trapped. 10.f3! ♘d5 10…h6 11.♘xg6 fxg6 12.♕d3 leads to a disaster! 11.♗d2 >Instead, 11.♗xd5 is also better for White, but this seems even more e ective. 11…♗xc3 12.bxc3 Black has no good way to stop h4-h5 anymore. 605

12…b5 13.♗xd5 ♕xd5 14.h5 f6 The only way to not lose the piece straight away. 15.hxg6 fxe5 16.♖xh7 ♖g8 17.♖h5 With a decisive advantage for White. B) 5…♗g4 This leads to much more quiet positions, compared to 5…c6. 6.h3

6…♗xf3 This is forced. White now gets the usual type of edge against the Scandinavian with the bishop pair and a space advantage. Black is very solid though as we will see. We need to prove why 6…♗h5? is indeed unplayable. This is a very bad version of an already risky idea. Engines already show a winning position for White. It’s not that obvious to a human, but White’s clear advantage is indisputable: 7.g4 ♗g6 8.♘e5 e6 After 8…♘bd7, 9.♕e2! is the only somewhat difficult move White needs to know/find: 9…♘xe5 10.dxe5 ♘d7 11.♗g5. Many moves are winning here, but this may be the most convincing. We castle queenside and 606

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

simply storm Black’s position. 9.♗g2 c6 10.h4 ♗b4 There are no reasonable alternatives for Black anymore.

10…♗b4 allows …♗e4 due to the pin, so our answer renews the threat: 11.0-0 We have seen this idea frequently. Concrete play is strong here, overruling the impression that castling looks risky after g2-g4 and h3-h4. 11…♘bd7 12.♕e2 and h4-h5 is coming next, with decisive effect. 7.♕xf3 c6

Black has to cover b7 and now attacks d4 for real. 8.♗e3 I think White should castle queenside here to be able to push his kingside 607

pawns without exposing the king. 8.♗e3 prepares 0-0-0 and covers d4. 8…e6 9.♗d3 I prefer this slightly over 9.0-0-0, as 9… ♗b4 forces 10.♘e4 ♘xe4 11.♕xe4 ♕d5 and Black trades a lot of material. I still think White is slightly better, but my main recommendation of 9.♗d3 exerts more pressure and keeps pieces on. 9…♘bd7 After 9…♗b4, queenside castling is not advisable, but White is better anyway: 10.0-0 ♘bd7 11.♘e4 and White has the usual space/bishop pair advantage. Here the black bishop is misplaced on b4 and will have to move again soon. 10.0-0-0 ♗b4 By far the most popular move and best.

Instead 10…♕a5 is answered by 11.♔b1!. ‘Long castling’ is long because it takes two moves – you need to play ♔b1 at some point. Jokes aside, the king move is indeed useful as the queen is eyeing a2. Black has no clear play here, while White improves easily on the kingside. Please check sample game 26, Moskalenko-Petrov, Orsha 608

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

2018, to see how the play develops from here. 11.♘e2 Avoiding trades after ♘e4. The knight might prove useful on g3 or f4 later. 11…♘d5 12.♕g3 This attacks g7 and prepares ♗g5 if appropiate. 12…g6

After 13…♗f8 14.♗xf8 ♖xf8 15.c4, White is better in the centre and Black’s king has still not yet castled. 14.h4 ♗f8 Not the only move, but it makes sense to try to trade this annoying bishop. 15.♗g5 One point of 14 h4. 15…♗e7 16.♔b1

This is the best way to deal with the threat against the g7-pawn: 1) 12…0-0? fails quite drastically: 13.♗h6 ♕f6? (13…g6 was losing the exchange, but it was the lesser evil) 14.♗g5 is extremely embarrassing for Black; 2) 12…♘xe3 13.fxe3 13.fxe3 is quite comfortable for White. We might play h3-h4 and e3-e4 soon, to seize the initiative. 13.♗h6 We stop Black going …0-0 and preserve the bishop from a possible trade against the knight on d5. 13…b5 Claiming some space and securing the knight on d5.

I think White is slightly better here. Black is solid, but White has the classical advantage of extra space.

609

Conclusion With this Scandinavian line Black is content to get this type of position, being solid but very slightly worse while not needing to study much theory – a sensible trade-o if you want a ‘Keep It Simple’ approach for Black. From our White perspective it is nothing to complain about either – we have preserved a small edge in an easyto-play position.

610

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

queen around and then opening up the position in the centre feels wrong. 5.♘f3 ♗g4

CHAPTER 30

Scandinavian: 2…♕xd5 3.♘c3 ♕a5 1.e4 d5 2.exd5 ♕xd5 3.♘c3 ♕a5

This is the traditional main line, but it is not as popular anymore, as black players began to appreciate that the queen is a bit safer on d8 and that there is little concrete advantage in it being on a5. 4.d4 There are some interesting ideas here in delaying or omitting this move altogether, but I like to recommend something more classical for our repertoire. Going d2-d4, and ♘f3 later, certainly makes a lot of sense. Now Black has many moves, but the main line is 4…♘f6 (line D). The deviations at best transpose to line D, but are mostly bad. A) 4…e5? Black ignores some opening principles in the rst four moves – moving the 611

This is the most popular move here, and worth checking, as the refutation is not well known and rarely played. The alternatives for Black don’t even demand precision from White: 1) 5…♗b4 6.♗d2 exd4 (White was threatening to take on e5 and to play a2-a3, so this is forced) 7.♘xd4 ♕e5+ (relatively best, as after the queen trade Black at least won’t get mated quickly) 8.♕e2 ♕xe2+ 9.♗xe2. The substantial lead in development is still very promising for White, even after the queens are o the board. Castling queenside and ♖he1, to centralize, will pose Black serious problems; 2) After 5…exd4, 6.♕xd4 is even better than 6.♘xd4 ♗b4 7.♗d2, transposing to 5….♗b4. White has excellent chances to score a quick win from here. 6.♗c4! This quite logical developing move has only been played ve times out of a total of 128 games with this position. 612

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

Stock sh already evaluates this as winning for White, which is probably true for engine standards. In human play it is still tricky, but Black certainly has the tougher task ahead of him. The simple threat is now to take on f7, followed by 8.♘g5+. 6…♘f6 Or 6…e4 7.♕e2 ♘f6 8.♗d2 ♕f5.

The only move to pose some problems at all. 9.♗xf7+!. I’ll only give the main line after this, not all possible lines that Black may play. It is a very obscure line that may never occur in any of your games, so learning lots of these lines by heart is not sensible. However, it makes sense to use this position as a starting point for practicing tactics and attacking play – just analyse it a bit on your own and check with the engine later. 9…♔xf7 10.♘g5+ ♔g6 11.f3 exf3 12.gxf3 ♗h5 13.♘ce4 with a winning attack for White. 7.♗d2 This was the second serious threat, besides 7.♗xf7+. The knight jump to d5 will be deadly now.

613

7…exd4 8.♘d5 ♕c5 9.♕e2+ ♔d8 10.♗b4 ♗xf3 11.gxf3 ♕c6 12.♗xf8 ♘xd5 13.♗b5 ♕e6 14.♗xg7 A fairly long line, but all forced and easy to calculate. White will win d4 and have a winning position. B) 4…♗f5?!

Not very common, and for good reason, but in fact the strongest reply is not played all that often. 5.♕f3! ♘c6 Even worse is the natural-looking 5… c6 because of 6.b4!. This leads to a very long line, but it is forced and easy to reproduce once you remember the b2b4! idea: 6…♕xb4 7.♖b1 ♕xd4 8.♕xf5 ♕xc3+ 9.♗d2 ♕d4 10.♕c8+ ♕d8 11.♕xb7 ♘d7 12.♕xc6 with a clear advantage for White. 6.♗b5 ♗d7 7.♘ge2 And White is better here with the superior development and the vulnerable queen on a5. C) 4…c6

614

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

D) 4…♘f6 5.♘f3

Black starts with 4…c6 and delays the normal-looking move …♘f6. Quite often it will transpose anyway, but let’s check the exceptions quickly. 5.♘f3 1) 5…♘f6 6.♗c4 transposes to the main line below (Line D); 2) 5…♗g4 6.h3 ♗h5 (6…♗xf3 7.♕xf3 ♘f6 8.♗d2 transposes to 4… ♘f6 and is just better for White) 7.g4 ♗g6 8.♘e5 – now 8…♘f6 will transpose to 4…♘f6 and there is no real alternative as 8…e6 is basically the same thing. White is clearly better; 3) 5…♗f5 6.♘e5!. Immediately preparing to attack the bishop with g2g4. The game will now almost always transpose into our repertoire’s main line (D below). Black can try to avoid the transposition with 6…♘d7, but the cure is worse than the disease: 7.♘c4 ♕d8 (7…♕c7 8.♕f3 e6 9.♗f4 ♕d8 10.0-0-0 is excellent for White as well) 8.d5! (this is surprisingly tough to meet) 8…♘gf6 9.dxc6 bxc6 10.♕f3 (and White wins c6! The pawn on c2 is poisoned) 10…♗xc2?? 11.♕e2 ♗g6 12.♘d6# Oops!

615

D1) 5…♗g4 This early bishop sortie seems to be somewhat imprecise. It is better for Black to wait until White has played his bishop to c4, and only then to pin. Let’s see why 5…♗g4 here is too early. 6.h3 ♗h5 Instead, 6…♗xf3 is always a concession, leading to a comfortable game for White: 7.♕xf3 c6 8.♗d2 and White enjoys the bishop pair without any downsides. 7.g4 As usual, White needs to act quickly to exploit Black’s play. 7…♗g6 8.♘e5 This is already very critical for Black. One misstep might lead to a quick defeat now, as the bishop is under threat due to h3-h4-h5. 8…e6 The position after 8…c6 9.h4 ♗e4 10.♘c4 ♕d8 11.♘xe4 ♘xe4 12.♕d3 ♘f6 13.g5 ♘d5 14.♗d2 looks like Black has skipped some moves – White is clearly better. 9.♗g2 616

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

This gains a tempo and controls e4. 9…c6 10.h4 Black has no fully satisfactory move now. 10…♗e4 11.♗xe4 ♘xe4 12.♕f3! ♘d6 13.♗f4

With a great position for White. D2) 5…♘c6

This has a somewhat dubious reputation. As far as I can see, it is even worse than that. With best play White should get a comfortable advantage, at least. 6.♗d2! This is a key move to remember. After …♘c6 the black queen has few retreat 617

squares compared to 5…c6. White prepares ♘b5, followed by c2-c4-c5, to exploit that. 6…♗g4 This leads to a huge advantage for White, even though the refutation is not played all that often. Black’s best move to limit the damage is probably 6…a6, but after 7.♗c4 White is threatening 8.♘d5 to trap the queen: 7…♕h5 8.♘e5 (this leads to a very comfortable endgame. White may try for more with 8.♗e2, but 8.♘e5 is simple and very good) 8…♕xd1+ 9.♖xd1 e6 10.♘xc6 bxc6 11.♘a4 and Black will su er due to his permanently damaged pawn structure. 7.d5! ♘xd5 Or 7…♘e5 8.♘b5 ♕b6 9.♗e3 ♕a5+ 10.c3. Black is helpless against the threat 11.b4. The queen can be saved with 10…c5, but this is already quite sad. 8.♘b5 ♗xf3 8…♕b6 9.c4 amounts to the same thing as after 8…♗xf3. Black will lose a piece for insu cient compensation. 9.gxf3 ♕b6 10.c4 0-0-0 After 10…♘f6?? 11.c5 it’s over immediately. 11.cxd5 ♖xd5 12.♘c3 ♖d8 13.♘a4 ♕d4 14.♖c1 Black does not have enough for the piece in this position. D3) 5…♗f5

618

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

This move order is somewhat tricky as 6.♗c4 e6 is quite decent for Black. We need to tweak our move order as well. 6.♘e5! c6 The black queen needs some squares! 7.♗c4 e6 8.g4 And we have transposed to the main line (D4) below. D4) 5…c6

The most active square for the bishop. Note that against early …♗f5 moves we go ♘e5, intending ♗c4 and g2-g4. If Black delays the bishop move to f5 we should start with ♗c4, as the immediate 6.♘e5 may be answered with …♗e6, as played by Anand vs Kasparov in the only ever World Championship game that started with the Scandinavian. This line seems to be less challenging for Black, in my opinion. Now we need to check Black’s two bishop moves available. D41) 6…♗g4 This pin is not particularly problematic, but as usual we need to act quickly and need to be aware of one subtle point on the next move. 7.h3 ♗h5

This is the most common move, giving the queen retreat squares. Often, this position is reached via the 4…c6 5.♘f3 ♘f6 move order as well. By playing … c7-c6 Black is not yet committing where the c8-bishop will be developed. 6.♗c4

8.♕e2! It is important to understand and remember that 8.g4 ♗g6 9.♘e5 looks incredibly similar to the main line with …♗f5, but here the ♘e5 does not threaten f7 as the black bishop is already on g6. This means that Black can do without …e7-e6 and can

619

620

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

challenge the e5-knight immediately with 9…♘bd7!. This is quite annoying as we don’t even get h3-h4 in. Fortunately, we have a very good alternative, namely the queen move to e2. We simply delay the g2-g4 push until a more appropriate moment arises. 8…♘bd7 8…e6 9.g4 will very likely transpose to 8…♘bd7 9.g4 as Black needs to play both …♘bd7 and …e7-e6 quickly. 9.g4 ♗g6 10.♗d2 e6 The c2-pawn is covered. After 10… ♗xc2??, 11.♘b5 ♕b6 12.♘d6+ is the point. With the queen on e2 this is always in the air. 11.0-0-0 ♗b4 12.a3 ♗xc3 13.♗xc3 ♕c7 14.♘e5

And White is at least slightly better with the bishop pair and more space – in the Scandinavian this is a very typical scenario. D42) 6…♗f5

621

7.♘e5 e6 8.g4 As usual we need to play forcefully to put pressure on Black. 8…♗g6 8…♗e4 is no solution for Black’s problems, but needs to be answered precisely: 9.0-0 (threatening 10.g5 to win a piece) 9…♗d5 (the only move to avoid an immediate loss) 10.♗d3 (the direct 10.f4 is also ne, but this avoids trading the somewhat unstable piece on d5, and is more logical, in my opinion) 10…♘bd7 11.f4 with an excellent position for White. 9.h4 ♘bd7 Black has tried 9…♗e4 here as well: 10.♖h3 (castling one move before, when the pawn was still on h2, was just ne. Here it would be OK as well, but using the fact that the h-pawn has moved is tempting) 10…♗d5 11.♗d2 (this forces the action. Black is behind in development and needs to move the queen again) 11…♗xc4 12.♘xc4 ♕d8 13.g5! (ignoring the threat to the dpawn. This position is just excellent for White) 13…♕xd4.

622

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

This loses by force. It’s fun to explore, even though you will probably never get it in a real game: 14.♕e2 ♕g4 15.gxf6! ♕xh3 16.0-0-0. White’s main threat is 17.♘e4, combined with ♕f3/ ♖h1 ideas, if the black queen moves. I’ll stop here, and suggest that you look at this position independent of opening study and treat it as an exercise in solving/analysing an attack. 10.♘xd7 And here is the nal split of lines:

puts the king on the kingside, where we have just pushed our pawns. In fact, White’s king is not in danger as Black has no attacking pieces. After, instead, 12.♖h3 Black does not have an easy life either, but it seems less convincing. 12…♗d5 Against 12…♖d8 or a similar move White will take on e4 with a nice initiative: 13.♘xe4 ♘xe4 14.♗d3 ♘f6 15.g5 ♘e8 16.♕f3. It is complicated, but White is better. 13.♘xd5 cxd5 14.♗d3

D421) 10…♔xd7 11.h5 ♗e4 And White is much better. We have some options how we might continue according to what Black does. After something like …♖d8/…♔c8, for instance, a strong plan would involve preparing b2-b3/c2-c4 to play on the queenside.

12.0-0! In order to punish Black, we need to undertake some extreme measures. This move unpins immediately, but 623

D422) 10…♘xd7 11.h5 ♗e4 12.0-0 12.♖h3 was played in the famous game Anand-Lautier, Biel 1997. If that does not ring a bell, please turn some pages of this book to nd sample game 27. It is such a great game that I needed to feature it, even though I’d rather recommend a di erent move at this 624

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

point. Anand’s choice leads to very sharp play where ‘Keep It Simple’ does not really apply, but maybe once you looked at the game you want to give it a try, anyway?

♗c4/g2-g4 play we put lots of pressure on Black. It can easily end up disastrous for Black – one mistake can prove fatal. Best play still leads to an advantage for White, but it’s still a ght.

12…♗d5 13.♘xd5 cxd5 14.♗d3 ♗d6 15.♗d2 ♕c7 16.♕f3 We need to cover f4, preventing …♗f4 by Black. I prefer White here, but it is quite clearly a complicated position. Don’t be afraid of the seemingly open king position. Black does not have any attacking pieces to threaten us directly, so it is not particularly risky – that being said, it’s not like being completely safe as in a typical castling scenario. Conclusion Finding a good approach against the 3…♕a5 Scandinavian was surprisingly di cult. With the queen being active on a5, all lines are more tactically tense, compared to 3…♕d8 lines. It is very important to realize that ‘normal’ moves are not enough to get anything against 3…♕a5, you need to play very energetically. With the suggested ♘e5/ 625

626

Chapter 31: Alekhine’s Defence: 1…♘f6

CHAPTER 31

Alekhine’s Defence: 1…♘f6 1.e4 ♘f6

2.e5 That must be best, but I need to mention that I checked 2.♘c3 as well. The main reason to consider this alternative is that we play the position after 2…e5 3.♘f3 anyway, arising from the Petro /Russian Defence move order. However, the issue is 2…d5, after which I don’t see anything too promising. You can investigate 3.e5, but after 3…♘fd7 I didn’t see anything better than to allow a transposition into the French Defence – and this is a line we don’t have in our repertoire. Long story short – just push the epawn! 2…♘d5 1) The retro move 2…♘g8 should just be answered by simple development. It looks odd, but Black has no weaknesses… well, if you don’t move pawns this is what you get in terms of 627

structure – not a strategy you can keep up the whole game! 3.d4 d6 4.♘f3. Just continue to develop and don’t worry about ‘not nding’ a refutation of the silly 2…♘g8 move. There is no refutation, but we are very comfortable; 2) 2…♘e4? just makes no sense. Simple play gives White a great edge in the centre: 3.d3 ♘c5 4.d4 ♘e6 5.f4 with a clear advantage for White. 3.d4 d6

4.c4 Theoreticians usually view 4.♘f3 as the critical move, and I agree. The practical problem behind 4.♘f3 is that Black chooses the defensive set-up, so you need to be prepared for a wide range of lines. In the Playing 1.e4 book by John Shaw the author devotes a huge chunk of the book to ghting the Alekhine’s with this line, but I don’t think it is very practical to learn all the details when you will only get the Alekhine’s once in a blue moon. Our choice still carries some punch, while being easier to learn. 628

Chapter 31: Alekhine’s Defence: 1…♘f6

structure with the black knight being slightly o side, on b6. 6.♘c3

4…♘b6 5.exd6

The Exchange Variation of the Alekhine is our repertoire choice. It still puts some considerable pressure on Black without being overly theoretical. Now Black has a very fundamental choice between recapturing with either c-pawn or e-pawn. There is a general consensus that taking with the e-pawn is solid for Black, but White usually should have a slight pull. On the other hand, taking with the c-pawn is sharper, but has been under a cloud for quite a while due to a certain set-up that we will investigate soon. I think this is actually quite playable for Black, but White can still pose some serious problems, of course. We have the obvious split now, with Black having the choice how to recapture.

Our set-up includes ♘c3/♗d3/♘ge2. The main point of this set-up is to make the black c8-bishop look a bit silly. Black usually tries to pin the f3knight by …♗g4, so we avoid ♘f3 completely. The ♘ge2 based set-up has some more concrete ideas that we will investigate now. A1) 6…g6 This does not t well with the e-pawn recapture. 7.♘f3 ♗g7 8.♗g5 This is awkward for Black to meet. He needs to weaken with 8…f6 or misplace pieces.

A) 5…exd6 The solid, conservative move. The symmetrical structure should be a little bit better for White, as we have more space and the more active pieces. A way to view this position type is to call it a Petro /Russian Defence

A2) 6…♗f5 This is quite uncommon, but it is useful to know how White should react. 7.♗d3 Simple, but good! This way White gets quick development and, therefore, good support for the centre. 7…♗xd3 8.♕xd3 ♗e7 9.♘f3 0-0 10.0-0

629

630

Chapter 31: Alekhine’s Defence: 1…♘f6

Next we go ♖e1/♗f4, and maybe double on the e- le. White is very comfortable here. A3) 6…♘c6 Black is putting some pressure on d4, so 7.♗d3 is not possible now. This move still has little independent value, if we play similarly to the main line 6… ♗e7 (A4 below). 7.♗e3 Instead, 7.d5 is a try to punish Black, but it is not much better than just this natural move that will transpose to 6… ♗e7, most likely. 7…♗e7 At this point, 7…♗f5 8.♗d3 is very similar to line A2 above. 8.♗d3 Now usually we will transpose to line A4 below or very similar positions with a sequence like 8…0-0 9.b3 ♗f6 10.♘ge2, but we need to check one deviation that involves a common idea in this line: 8…♘b4

631

This leads to nothing due to White’s strong reply. 9.♗b1! And Black cannot take on c4 now due to 10.♕a4+. This is a very common tactical trick in this line against the Alekhine. 9…0-0 10.a3 ♘c6 11.♘ge2 And again, we set a little trap. 11…♘xc4?? A blunder, of course. Against normal moves White has a slightly better version of the main line because Black has lost time with 8…♘b4. 12.♕d3 And the quirky Alekhine knight drops due to the double attack. A4) 6…♗e7 The most popular move, preparing to castle. 7.♗d3 ♘c6 8.♘ge2 Instead, 8.♗e3 would transpose to the lines we looked at via 6…♘c6 7.♗e3. Keeping the bishop exible is even more promising for White here. 8…♗g4 We don’t have to worry about 8…♘b4, as again 9.♗b1! is the key move: 9…0-0 632

Chapter 31: Alekhine’s Defence: 1…♘f6

10.a3 ♘c6 11.0-0 (the c4-pawn is protected by tactics) 11…♘xc4?? 12.♕d3 and we win the knight. 9.♗e3 Protecting the d4-pawn, which was just attacked, due to the pin on the e2knight. Now 9…♘b4 10.♗b1 again leads nowhere.

up with are d4-d5 or answering Black’s …d6-d5 with c4-c5. In both cases we expand nicely. A42) 9…0-0

A41) 9…♗f6

Black immediately puts more pressure on the centre. 10.f3 This is a key point of avoiding ♘f3 setups. 10…♗h5 The bishop sortie 10…♗h4+? leads to huge trouble for Black: 11.g3! ♗xf3 12.0-0! (both bishops are attacked!) 12…♗xe2 13.♕xe2 ♗e7 14.c5 with a great attack for White. Engines already see a win for us, but it is not very economical to study any further. 11.0-0 ♗g6 12.♗f2 A good prophylactic move, securing the bishop from future attacks on the ele. White has a nice space advantage now. The most common ideas to follow 633

10.b3! This is an important nesse. Alternatives don’t convince as much. 10.f3 is not so great because of 10… ♗h4+, disrupting White’s set-up. Instead, 10.0-0 allows 10…♘b4, as 11.♗b1 is not working now – Black just grabs on c4. 10…♗f6 Now after 10…♘b4?! we simply play 11.♗b1. Now c4 is protected, allowing this move. In general, White likes to keep pieces on the board due to his advantage in space. 11.0-0 ♖e8 12.h3 ♗h5 13.♕d2 Unpinning the knight and threatening to go ♘c3-e2-f4, getting the pair of bishops. 13…♗g6 14.d5 ♘e5 15.♗xg6 ♘xg6 16.♘g3

634

Chapter 31: Alekhine’s Defence: 1…♘f6

White enjoys a very easy game. We have more space and ideas to push the f-pawn. This whole line with 5.exd6 promises White a small, but very comfortable edge. The game GelashviliBerkes, Minsk 2017, is our sample game 29 and well worth studying to better understand how this variation is played properly.

White now has a very dangerous setup available, based on early queenside development with ♘c3/♗e3/♖c1 and b2-b3. The idea of this so-called Voronezh Variation is to clear the long diagonal and prepare to play d4-d5 at the right moment. With a black bishop on g7 this advance needs to be prepared properly. 6…g6 Black sometimes goes for 6…♘c6 immediately.

B) 5…cxd6

6.♘c3 The recapture with the c-pawn is far more ambitious compared to the symmetrical choice. Black intends to anchetto his bishop on g7, putting pressure on White’s central pawn formation. 635

This we should punish with 7.d5! ♘e5 8.f4 ♘ed7 (8…♘exc4?? 9.♗xc4 ♘xc4 10.♕a4+ – a common trick in the Alekhine) 9.♘f3 g6 10.♗e2 ♗g7 11.♘d4 (White has a dominant position in the centre) 11…0-0 12.0-0 ♘c5 13.f5 and Black is su ering in a passive position. 7.♗e3 ♗g7 8.♖c1 The main point is to prepare d4-d5, often in combination with b2-b3, which keeps the knight o side (control of c4). 8…0-0 Or 8…♘c6 9.d5 ♘e5 10.♗e2 (this is played to cover g4, preparing f2-f4. 636

Chapter 31: Alekhine’s Defence: 1…♘f6

Black cannot take on c4 again due to ♕a4+ ideas) 10…0-0 11.b3 with an excellent position for White that we will examine via 8…0-0 9.b3. 9.b3

At this point Black needs to choose his defensive concept.

12.♘h3! (just waiting to push f2-f4 for a moment until we have developed completely) 12…♘bd7 (maybe best, but I think Black is in dreadful shape anyway) 13.0-0 ♘f6 14.f4 ♘eg4 15.♗d4 and White is a lot better. The plan is 16.♘f2 next, with pressure mounting on the e- le. Black has many weaknesses, like e6 and e7. 12.f4 Consistent with the earlier play, but the simple 12.♘f3 is also better for White. White has static advantages (space, centre, backward e-pawn) and Black playing …e7-e6 won’t solve all of them. 12…♘ed7 13.♘f3 ♘f6 14.0-0 e6 15.dxe6 ♗xe6 16.♗d4

B1) 9…♘c6 10.d5 ♘e5 11.♗e2 Preparing to play f2-f4, kicking the knight again. 11…♖e8 Here, 11…f5 is the most popular move with more than 200 games in Megabase 2018. I think this is quite remarkable, as the strongest reply leads to a terrible position for Black: And White has a nice edge. Note that … d6-d5 is usually answered with c4-c5, keeping control over d4. Black su ers with the potentially weak d6-pawn as well. B2) 9…e5 Black strikes in the centre. This move is better than 9…♘c6, but does not equalize either. 10.♘f3

637

638

Chapter 31: Alekhine’s Defence: 1…♘f6

This is actually a rare move. White also scores excellently with 10.dxe5, but 10.♘f3 is not easy for Black, either. 10…♘c6 11.♗e2 ♗g4 Black’s last two moves also could have happened in reverse order. 12.dxe5! So we capture after all. This opening of the position is favourable for White, mostly because the b6-knight can be kicked back. 12…dxe5 13.c5 ♘d7 14.0-0

And White has the better game. His majority on the queenside is dangerous and we have good squares like c4 and d5 available. Note that d6 is also an inviting post for a knight. The play in this line is excellently illustrated in sample game 28, Nguyen-Moussard, Karlsruhe 2017. B3) 9…♗f5 This is Black’s best bet against the Voronezh for a ght, leading to complicated play. I actually played this with the black pieces in a must-win game in the penultimate round of the Lille Open 2017 against French player, Frederic Clement. My opponent did not 639

know this move, but found an excellent solution over the board.

10.g4! This is a strong move, but in case this feels too extreme for you: there is nothing wrong with a move like 10.♘ge2, intending 11.♘g3. 10…♗d7 11.h4 ♘c6 This was my choice, but it does not equalize. The whole line is still quite uncharted territory, so we cannot rely on game examples, but I don’t see a better move for Black even after looking at it with the engine.

12.d5! This is better than the move played in the game. After 12.h5?! d5! Black 640

Chapter 31: Alekhine’s Defence: 1…♘f6

would have been OK, but I failed to nd it and went 12…e5?!, still being worse. 12…♘e5 13.♗e2 The black knight is in trouble in the middle of the board. With f2-f4 the knight would be trapped. 13…♖c8 The engine suggestion, based on 14.f4? ♘exc4!, with good play for Black. 14.♗d4 This prepares f2-f4 with much greater e ect. 14…f5 15.g5 f4 16.♘f3

In the symmetrical line with 5…exd6 you just need to remember the set-up and make sure not to run into …♘b4 at an inappropriate time. 5…cxd6 is a lot sharper, but also slightly better for White at least. Make sure to study 9…♗f5 and 9…e5 properly. Alekhine players sometimes are quite booked up or have loads of experience, so they might have some idea even on move 10 or later.

A very tough position, but White must be somewhat better. Black has some weaknesses and still needs to activate the b6-knight. Conclusion Alekhine’s Defence is not that easy to handle from the white side. The positions are quite unusual, and while Black might have it on the board every other game, we will face it infrequently. Therefore it is sensible to play a line where you fare well with common-sense moves. Our selected approach ts the bill perfectly.

641

642

Chapter 32: 1.e4: rare lines

CHAPTER 32

1.e4: rare lines

but against 1.d4 and 1.e4 it is a slightly dubious choice as Black does not get active play against the strong centre that White will build. 2.d4 ♗b7 3.♗d3!

1.e4

We are back where we started, right on the very rst move. This chapter deals with the rarest of the rare lines that Black may choose. A) 1…b6

This is called Owen’s Defence. 1…b6 is actually quite playable against 1.c4 and 1.♘f3 (as seen, for example, in the game Eljanov-Carlsen, Douglas 2017), 643

Not that 3.♘c3 would be terrible, but exibility with the c-pawn is an important asset for White. In some cases, c2-c3 lends important protection to the centre, in other cases c2-c4 builds up an even broader centre. A1) 3…♘f6 4.♕e2 Again keeping c2-c3 as an option. 4…♘c6 A really weird way to play, but it is not losing by force or anything like that. 5.c3 e5 Black needs to commit a centre pawn at some point. 6.♘f3 d6 7.0-0 White is just better here as there is no serious pressure on our centre. The simplest way to play is to go d4-d5 soon, getting a structure that looks like an Old Indian gone wrong for Black. Alternatively, just playing ♖e1 and ♘d2-f1-g3 in Ruy Lopez style is also quite strong. 644

Chapter 32: 1.e4: rare lines

A2) 3…f5?? This just loses, but you will still meet it sometimes in blitz games. 4.exf5! ♗xg2 5.♕h5+ g6 6.fxg6 ♗g7 6…♘f6? leads to a cute mating nish after 7.gxh7+ ♘xh5 8.♗g6#. 7.gxh7+ ♔f8

8.♘f3! In fact, here there is more than one move that wins, but this one is quite simple to learn. 8…♘f6 Black can’t take the rook: 8…♗xh1 9.♘e5 ♕e8 10.hxg8♕+ ♖xg8 11.♕f5+ ♗f6 12.♗h6+ ♖g7 13.♗xg7+ ♔xg7 14.♕h7+ ♔f8 15.♘g6+ and we win the queen – at least! 9.♕g6 ♗xf3 Again, 9…♗xh1 10.♗h6 ♖xh7 11.♘g5 leads to a disaster. 10.♖g1 ♖xh7 11.♕g3! This is the key move that decides the game. 11…♗e4 12.♗xe4 ♘xe4 13.♕f3+ ♘f6 14.♕xa8 And White just wins material.

That’s best, and most frequently played. 4.♘f3 After our natural development move, we will look at some alternatives for Black now. They all lead to a comfortable game for us, at the very least.

A31) 4…c5 5.c3 ♘f6 6.e5 ♘d5 7.♗e4 Introducing the idea to play c2-c4 at the right moment. 7…cxd4 8.♘xd4 I think White is just more comfortable here. Stock sh now recommends that Black allows the deterioration of the central pawn structure. 8…♘c6 9.♗xd5 exd5 10.0-0 And White is extremely comfortable here. A32) 4…g6 Black might switch to a Hippopotamus set-up with this move or 4…d6. 5.0-0 ♗g7

A3) 3…e6

645

646

Chapter 32: 1.e4: rare lines

6.c4 This is an important point. Against the Hippo we should always try to put 3 pawns in the centre: the c-/d-/e-pawns, like here, or the d-/e-/f-pawns, against the Pirc Defence move orders. The main point of having three pawns in the centre is to support the advance of the d- or e-pawn. If you intend to play d4-d5, it is excellent to have a pawn on c4 to support this. If you want to play e4-e5, it is the mirror image situation with the f-pawn. It is tough to analyse the Hippo move-by-move, but I want to show a typical sample sequence. 6…d6 7.♘c3 ♘e7 8.♖e1 0-0 9.♗e3 h6 10.d5 e5 11.b4 With a very favourable King’s Indian position for White. A33) 4…d5 Black may play this here or after some preparatory moves. It is important to understand that e4-e5 is always going to be somewhat better for White as this is a quite favourable version of the French with a strange bishop on b7. In some cases, however, we have even better options than advancing the e647

pawn, mostly exploiting Black’s bad development. 5.exd5! As mentioned, e4-e5 is also possible, but this is an attempt to get Black into immediate complications while he is underdeveloped. 5…♕xd5 5…♗xd5 will very likely lead to the same position as after the queen capture: 6.0-0 ♘f6 7.c4 ♗b7 and we have transposed to 5…♕xd5. 6.c4 ♕d8 6…♕h5?! 7.♗e2!. The black queen is actually quite a liability on h5. Note that ♘c3-b5 will be a huge problem, as well. 7.0-0 ♘f6 8.♘c3 ♗e7 9.♖e1

We will continue with simple development, like ♕e2/♗f4/♖ad1, and enjoy our central space advantage. B) 1…a6

648

Chapter 32: 1.e4: rare lines

This is the St George Defence. It had its ve minutes of fame when Tony Miles used it to defeat World Champion Anatoly Karpov in 1980, a time when Karpov was basically unbeatable. Black intends to go …b7-b5 for the extended anchetto. This is slow and weakens the queenside somewhat, but it is not bad enough to lose outright. We are comfortable, but not much more. 2.d4 b5 3.♗d3 Our set-up against this opening involves ♗d3/c2-c3/0-0/♖e1, supporting the centre. 3…♗b7 4.♘f3 e6 5.0-0 c5 6.c3 ♘f6 7.♖e1 We have all the classical advantages: central control and healthy development – just play chess from here.

The Nimzowitsch Defence. It is not that bad if you use it just as a transpositional tool to reach 1.e4 e5 positions – the independent lines, however, are not that reliable for Black, or just downright bad. 2.♘f3 Of course, 2.d4 is ne, but 2.♘f3 is logical as well and easier to handle. C1) 2…d5 Scandi-style play – it is very questionable though. 3.exd5 ♕xd5 4.♘c3 ♕a5 Going the other way with 4…♕h5 5.♘b5 leads to quite an embarrassing moment for Black – now 5…♔d8 is the only move and it is not pretty.

C) 1…♘c6

649

650

Chapter 32: 1.e4: rare lines

5.♗b5! This is important to remember. The pin is annoying and the black queen is deprived of some squares on the 5th rank. 5…♗d7 Relatively best. 6.d4 White’s ideas are ♗d2/a2-a4 related. The queen gets into some real trouble on a5. 6…a6 7.♗xc6 Here, 7.♗a4 might be even better, but taking on c6 is a simple road to an advantage. Against rare lines, simple recipes are very sensible. If you go and learn some complex refutation, you will likely forget it when that rare line actually pops up after a long time. 7…♗xc6 8.0-0 0-0-0 9.♘e5 ♗e8 10.♕f3 White is clearly better. C2) 2…f5?

3.exf5 d5 4.♗b5 ♗xf5 5.♘e5 Black has tons of weaknesses that White is eager to exploit. 5…♕d6 6.d4 ♘f6 7.0-0 ♘d7 8.♗f4 All very natural. We develop quickly and occupy the weak spots. The only tricky part is about to come. 8…♘dxe5 9.♗xe5 ♕g6

10.♕f3! This is the move to remember! It is given by Stock sh as a winning advantage. It is not that simple to prove, but White is close to landing a quick knock-out. Amongst others, White is intending to play c2-c4 next – d5 and c7 are also hanging. 10…♗e4 11.♕h3 e6 12.♘d2 ♗xc2 13.♕c3 And Black is busted.

The Colorado Gambit. It’s completely incorrect, and therefore it’s hard to say why some titled players have tried it. If you know the line shown here you are already close to winning.

C3) 2…d6 The best of the independent lines. After it White will get a small advantage, but more ambitious tries like 2…d5 or 2… f5 back re quite badly, as we have explored already. The best move is quite clearly 2…e5, transposing to double king pawn territory. 3.d4

651

652

Chapter 32: 1.e4: rare lines

C31) 3…e5 4.♘c3 we have examined via 1…e5 (see Chapter 5). C32) 3…♘f6 4.♘c3 ♗g4 With 4…g6 Black is aiming for a mix of the Nimzowitsch and Pirc Defences. Our Anti-Pirc repertoire is based on a delayed f2-f4 in most cases, so we can’t just play as we do against the Pirc: 5.♗b5 (this works quite well here, just as after 4…♗g4) 5…a6 (after 5…♗d7, 6.e5 is very strong as Black does not have the d7-square anymore for the king’s knight. White is clearly better with simple moves) 6.♗xc6+ bxc6 7.00 ♗g7 8.h3 0-0 9.♖e1.

And I prefer White here. Black’s pawn structure is not great and the bishop pair is not e ective at the moment. White would like to trade the g7bishop, so ♗f4/♕d2 is a natural follow-up. 5.♗b5 This leads to an easy-to-handle position that is slightly better for White due to the better pawn structure. 5…a6 Not pretty, but d4-d5 was an e ective idea. 6.♗xc6+ bxc6 7.h3 ♗h5 8.♕d3 Just unpinning and preparing to castle queenside. 8…e6 9.♗g5 ♗e7 10.g4 ♗g6 11.♕e3

White had an attractive position in Kotronias-Vouldis, Psichiko (15 minutes) 2002. Please check sample game 30 for the full game, a very complicated ght with a mysterious twist at the end. C33) 3…♗g4

653

654

Chapter 32: 1.e4: rare lines

8.cxb7+. 7…♕a5+ This is the most demanding move to meet. After 7…a6 8.c7+ axb5 9.cxd8♕+ ♖xd8 10.♘xf7 ♔xf7 11.♔xd1 White is a pawn up with the better pawn structure on top. 8.♘c3

This pin is somewhat premature compared to 3…♘f6 4.♘c3 ♗g4. 4.d5! Just gaining time. The most natural reply happens to fail spectacularly, which is a nice added bonus. 4…♘e5 Better is to drop back to b8, but after that, normal development is enough for White to ‘remember’.

5.♘xe5!! The following complications are not that easy to learn, but White is close to winning and you get to bash out some cool moves! 5…♗xd1 6.♗b5+ c6 7.dxc6 Wow, quite a cool position. Note that White is threatening both 8.c7+ and 655

Now with the black queen on a5 our main threat is to take on b7. C331) 8…a6 9.b4! Extremely important to remember! 9…♕d8 After 9…♕xb5 10.♘xb5 axb5 11.cxb7 ♖b8 12.♘c6 ♖xb7 13.♔xd1 Black is busted again. 10.c7+ This is now very similar to 7…a6. 10…axb5 11.cxd8♕+ ♖xd8 12.♘xf7 ♔xf7 13.♔xd1 And White is a pawn up with a good position. C332) 8…0-0-0 This is the toughest nut to crack again. 9.♘c4! ♕c7 The other queen move, 9…♕b4, also fails due to 10.a3 ♕c5 11.♗e3 ♕h5 656

Chapter 32: 1.e4: rare lines

12.cxb7+ ♔xb7 13.♖xd1.

White has a winning attack here! We have two pieces and a pawn for the queen, but once you spot the idea ♘a5+, followed by ♖d1-d4-b4!, it becomes obvious that Black is out of defences. 10.♘d5 ♕b8 It’s simple after 10…♗xc2 11.♘xc7 ♔xc7 12.cxb7 ♗xe4 13.♗e3, when you don’t need to know or nd anything special. 11.c7 ♕a8

It makes a lot of sense to check these great lines a bit for yourself. It might take a while for you to actually get them in a real game – but it is great attacking training to analyse these lines, completely independent of the practical relevance of the speci c opening moves. Conclusion Unsurprisingly, the rarest of the rare lines are not that great for Black. Some of the lines are playable, some can be directly refuted. I actually looked at some of the refutations in detail, even though you might never actually get them on the board (rare lines, you know). It was fun to analyse, so enjoy the ride and hope to get some of those odd lines on the board.

12.♗d2!! With the amazing threat of 13.♗a5!!. Black is completely lost. 657

658

Sample games

 

Sample games Game 22 PU 9.4 (B07) Sebastian Testor 2295 Christian Bauer 2627 Austria Bundesliga 2015/16 (2)

1.d4 d6 2.e4 ♘f6 3.♘c3 g6 4.♗e3 c6 5.h3 ♘bd7 6.f4 b5 7.a3 ♘b6

Black gives up on the idea of playing … e7-e5 soon. I believe Black badly needs to stake a claim in the centre and can’t a ord to ignore White’s broad set-up like he does in the game. 8.♘f3 I think that 8.♗d3 is somewhat more precise and thus I have recommended that move in the theoretical section. Black does not exploit the knight move in the game, so we transpose to normal positions where ♘f3/♗d3 was just played in a di erent order. 8…♕c7 659

This is very slow. The reason why I preferred to start with the bishop move is 8…d5 9.e5 ♘e4. White is still better after this, but I prefer to have complete control like in the game. Now 10.♘xe4 dxe4 11.♘d2 ♘d5 12.♕e2 f5 13.exf6 exf6 14.♘xe4 ♗g7 is almost forced, and it looks too murky for my taste, even though the engine likes White. 9.♗d3 ♘c4 This sends the bishop back home, but the knight is on a vulnerable square. If Black continues to develop with 9… ♗g7 10.0-0 0-0 11.♕e1 his position is quite passive. The simple attacking plan of f4-f5, ♕e1-h4, ♗e3-h6 etc. is very dangerous to face. 10.♗c1 a6 11.♕e2 ♗g7

12.0-0 Castling is a good move, but here 12.a4! was quite strong. After the natural 12…♗b7 13.0-0 0-0 14.e5 ♘d5 15.♘xd5 cxd5 White can strike with 16.e6, highlighting the weak light squares after the bishop has left c8. Stock sh already evaluates this as a clean win for White, so Black has to deviate earlier, but there is nothing 660

Sample games

great available. White is much better after 12 a4!. White absolutely had to avoid 12.♗xc4? bxc4 13.♕xc4 ♘xe4!, which gives Black good play after 14.♘xe4 d5 15.♕e2 dxe4 16.♕xe4 0-0 17.0-0 ♗f5. 12…0-0 13.e5 This gains space and makes the g7bishop look silly. 13…♘d5 14.♘xd5 cxd5 15.b3 e6?

This turns out to be a mistake. Black should have played the simpler 15…♘a5, even though White still has a useful initiative after a line like, for example, 16.♕f2 ♘c6 17.♗e3 f6 18.exd6 ♕xd6 19.c4. 16.♕e1 White goes for queenside play, now attacking the knight for real and covering the b4- and a5-squares. Less subtle people may choose 16.h4 here. It leads to a strong attack as well. 16…♘a5 16…♘b6 17.♗d2, intending ♗b4 or ♗a5, is no improvement for Black. 17.a4 bxa4 Not a happy move, but keeping the pawn on b5 wasn’t better. 661

After 17…♗d7 18.axb5 axb5 19.♗a3 ♘b7 20.♗b4 White is spoilt for choice what to attack next. 18.♖xa4 ♘b7 18…♘c6 19.♗a3 will win the exchange immediately. 19.♕e2 a5

20.♗d2 Here the engine suggestion 20.b4 ♗d7 21.b5 is very strong as well, but Testor’s move is increasing the pressure nicely. He is outrated by 300 points in this game, but his position is easy to play and already clearly superior. 20…♗d7 21.♖a2 ♕b6 22.♖fa1

22…f6 662

Sample games

Bauer does not want to wait passively anymore and decides to strike against the cramping e5-pawn. This is a very understandable and human decision, but it ultimately back res. This kind of development is very common if you strive for positions with a space advantage and no clear active plan for the opponent – they want to ‘break out’ and worsen their situation in the process. A patient move like 22…♖a7 was better. Here White needs to nd more complex solutions to increase his advantage. Stock sh nds the clever 23.c4! here. Now after 23…♕xb3 24.♗e3 dxc4 25.♗xc4 ♕b6 26.d5 ♘c5 27.dxe6 ♗xe6 28.♗xe6 fxe6 29.♘g5 White has a strong initiative and a clear advantage as well. 23.♔h1 A routine move, but probably unavoidable at some point. White already could have taken on f6, but Black can’t do much about White delaying it for one move. 23…♖a7 24.exd6 24.exf6 ♗xf6 25.h4 was even better, switching attention to the kingside. 25…♕d8 26.h5 gxh5 27.♘g5 ♗xg5 28.♕xh5 is a sample line. White wins in the attack. 24…♕xd6 25.c4 dxc4 26.bxc4 ♗c6

663

27.c5 Here White had a strong, but not very human alternative: 27.♕e3! is very strong. It threatens d4-d5 and after 27…♗xf3 28.c5 ♕c6 29.gxf3 the damaged structure is not important. White is just too quick, with moves like ♗e4 coming and the c-pawn on the way. 27…♕d7 28.♗c4 ♖e8 29.♗xa5? This has a tactical aw that went unnoticed in the game. 29.♗c3 instead would have kept all the advantage. 29…♗xf3? This should have lost by force. Instead, 29…♖xa5 30.♖xa5 ♘xa5 31.♖xa5 ♗h6 holds for Black. He’s got enough play now with the bishops against d4 and f4. 30.♕xf3 ♘xa5 31.♖xa5 ♖xa5 32.♖xa5 ♕xd4

664

Sample games

33.♖a8? 33.♕c6! was the win! 33…♖f8 34.♗xe6+ ♔h8 35.♕c7, and 36.♖a7 is the decisive threat. Note that the queen on c7 covers f4, so the white king is perfectly safe. 33…♖f8 34.♖xf8+ 34.♗xe6+ is a bit more obvious, but should be a draw at the end as well. 34…♔h8 35.♕a3 still poses a problem. Black only has one move to draw: 35… ♗h6!. Now c5-c6 is prevented and White needs to take care of his king with a move like 36.♖a4. Instead, 35… ♕xf4? loses after 36.c6 ♕f1+ 37.♔h2 ♕f4+ 38.♔g1 ♕d4+ 39.♔h1 ♕d1+ 40.♔h2. 34…♗xf8 35.♗xe6+ ♔g7 36.♕e2? This leads to serious trouble for White all of a sudden. 36.♕b7+ was the way to draw: 36… ♔h6 37.♗g8 ♕xf4 38.♕xh7+ ♔g5 39.♕d7 ♗xc5 40.♕d5+. 36…♕xf4 37.♕c2 ♕e3 38.♗g4 ♗xc5 39.♕d1 h5 40.♗f3 ♕f2 41.♗e2 ♕g3 42.♕f1

42…♕e3 At rst I thought that 42…♗d6 wins here, but now I think it is a draw as well: A) If White opts for 43.♔g1, his king needs to go for a walk: 43…♕h2+ 44.♔f2 ♗c5+ 45.♔f3 ♕e5 46.g4 ♗d6. Now Black wins a second pawn. There are still drawing chances for White though, as many endings without the queens will be a draw; B) 43.♕g1 looks very passive, but it may be a fortress: 43…♔h6 44.♗a6. White can only mark time with the bishop. Is there a win for Black? The engine does not nd one, and looking a bit myself I don’t see anything convincing either. One important line goes 44…g5 45.♗c4 f5 46.♗a6 g4 47.♗e2 gxh3 48.♗f1! (this is a clean draw!) 48…hxg2+ (48…h2 49.♕d4 will either lead to a perpetual or a queen trade, which is also an easy draw) 49.♕xg2 and again there is no way forward. The ‘wrong-coloured bishop’ draw is one of those odd things in chess. Isn’t it unfair that you can’t win with such a huge material edge?

665

666

Sample games

43.♕d1 ♔h6 44.♕f1 ♗d4 45.♕d1 f5 46.♗f3 ♗e5 47.♕e2 ♕f4 48.♔g1 ♕h2+ 49.♔f2 ♗d4+ 50.♔e1 ♕g1+ 51.♔d2 ♕a1 52.♕c4 ♗f6 53.♕c2 ♗g5+ 54.♔d3

The white king is exposed, but Black can’t control any light squares so there is no mate. The black pawns are also under control by the bishop on f3. Black tried for a bit, but it is a draw. 54…♗f6 55.♕c6 g5 56.♔c4 g4 57.hxg4 hxg4 58.♗d5 ♔g5 59.♔b5 ♕b2+ 60.♔a6 ♗d4 61.♕b5 ♕a3+ 62.♔b7 ♕e7+ 63.♔c8 ♗e5 64.♕c6 f4 65.♕d7 ♕c5+ 66.♕c6 ♕b4 67.♔d7 ♕f8 68.♕e6 ♗f6 69.♕g8+ ♕xg8 70.♗xg8 f3 71.gxf3 g3 72.f4+ ½-½ This game provides an excellent example of how easy White’s position is to play once Black has settled for the passive set-up without pushing …e7e5 in the centre. If White manages to play his pawn to e5 he will have a huge advantage in space, and Black lacks serious counterplay. Game 23 KF 14.1 (B06) Vasilios Kotronias

2416 Montreal 2002 (8)

1.e4 g6 2.d4 ♗g7 3.♘c3 d6 4.f4 Not our move order, but we will soon reach a position that could have arisen from 4.♗e3 as well. 4…e6 5.♘f3 ♘e7 6.♗e3 b6

Black is going for the so-called Hippopotamus or ‘Hippo’ set-up. Against this exible but slightly passive set-up I recommend to set up a broad pawn centre with either f2-f4 or c2-c4 played in order to support possible pawn advances in the centre. Most of the time we will play f2-f4, like seen in the game. With the pawns on d4/e4/f4 there are two pawn pushes that White can play for. He may opt for e4-e5, blunting the g7-bishop, or go for f4-f5, having an attack on the king in mind. 7.♕d2 Kotronias goes for queenside castling. 8.♗d3 and 0-0 next was ne as well. 7…♗b7 8.0-0-0 ♘d7

2578 Jean Hebert 667

668

Sample games

9.♕e1 The Hippo is more of a set-up than a concrete line. Both sides have many move orders and options. 9.♕e1 has the idea to go ♕h4 after Black castles kingside, which will happen most of the time. It works out beautifully for White in this game, but we will see it is not a completely one-sided a air. 9…a6 10.♗d3 This looks perfectly normal, but blocking the d- le has the drawback of allowing Black to play …c7-c5 more easily. If you keep the bishop on f1 for the time being, for example with a waiting move like 10.♔b1, it is di cult to play 10…c5. After 11.dxc5 ♘xc5 12.♗d4 White exchanges the important bishop on g7 and has play on the d- le. 10…c5 11.♖f1 11.dxc5 ♘xc5 12.♗d4 was still worth considering. Kotronias instead removes the rook from the long diagonal and protects the knight for a later f4-f5 push. 11…0-0 12.g4

12…♘f6?! It’s not easy to defend against an avalanche of pawns storming towards your king. Here Black had two good defensive options, both leading to complicated play with Black being ne: A) The rst move is 12…b5. Black now immediately threatens to push the cand b-pawns. White has little choice and needs to react in the centre: 13.dxc5 ♘xc5. Black is OK here. One important line now: 14.♗xc5 dxc5 15.e5 ♗xf3! (it’s better to trade this bishop for the knight instead of waiting for White to play ♗e4. The bishop on d3 is not very useful here so it is better for Black to not trade it) 16.♖xf3 ♕c7. The second point of taking on f3 is that now f4-f5 is impossible as it would drop e5. The position is about equal; B) The central strike 12…d5 also came into consideration. The coming complications are very di cult to assess though: 13.e5 (after 13.exd5 ♘xd5 14.♘xd5 ♗xd5 15.♘e5 ♕c7 Black is ne. The bishops are strong now; for example after 16.♘xd7 ♕xd7 17.dxc5 bxc5 18.c4? ♕a4! Black has a

669

670

Sample games

winning attack) 13…c4 14.♗e2 b5 (it looks scary for White, but h4-h5 is quick as well) 15.♘b1 ♕b6 16.h4 is entirely unclear. 13.♕h4 Here 13.dxc5 ♘xg4 14.cxd6 ♕xd6 15.♗d2 was better for White, and far less murky than the game. But White continues full steam ahead, getting the queen into the action.

13…cxd4? This gives White a huge and close to winning advantage. Black should have tried to throw a spanner in the works with 13…c4 14.♗xc4, and now there is more than one line that needs checking: A) 14…d5 is best, e.g. 15.exd5 ♘exd5 16.♘xd5 ♗xd5 17.♗xd5 ♕xd5 18.♔b1. White is a pawn up now, but Black’s pieces are active and he’s got some ideas on the light squares. White is for choice, but Black won’t get checkmated in under 20 moves; B) 14…b5 does not work due to 15.♘g5 bxc4 16.e5 with a great attack for White;

671

C) More interesting is 14…♖c8 15.♗b3 ♖xc3 16.bxc3 ♘xe4, but after 17.f5! White is better. All this is of course very tough to gure out without silicon help. 14.♘xd4 ♘d7 14…♘c6 was no improvement: 15.♘xc6 ♗xc6 16.f5 ♘d7 17.g5 and I don’t see anybody or even ‘anything’ surviving this attack. 15.f5 exf5 16.exf5 ♘c6 Here 16…♗xd4 prolongs the game, but 17.♗xd4 ♘c6 18.♕xd8 ♖fxd8 19.♗e4 is truly depressing.

17.♗g5 ♗f6 17…♕e8 can be refuted in more than one way, the simplest being 18.f6 ♘xd4 19.fxg7 ♔xg7 20.♗f6+ ♘xf6 21.♕xf6+ ♔g8 22.♕xd4. 18.♘e4 And Black resigned. A really powerful display by Kotronias, who never spent a tempo on a defensive or prophylactic move. However Black had some possibilities against this very direct approach, in particular on move 12. A more prophylactic approach, as examined on 672

Sample games

move 10, was very much worth considering. Against the Hippo, aggressive play is ne, but don’t underestimate the counterplay that Black may initiate with moves like … c7-c5, …b6-b5 or …d6-d5. Game 24 KP 4.15 (C41) Mircea Parligras 2574 Joachim Solberg 2229 Rethymno 2010 (2)

1.e4 d6 2.d4 ♘f6 3.♘c3 e5 4.♘f3 ♘bd7 5.♗c4 ♗e7 6.0-0 0-0 7.♖e1 c6 8.a4 b6 9.♗e3 I recommend to play 9.♗a2 rst, but these positions where Black holds the centre, the so-called Hanham set-up, are highly transpositional. In fact we will transpose to a position covered in the theoretical section very soon. 9…a6 10.♗a2 ♕c7 11.h3 ♖b8

tactically viable. The knight coming to f5 is just extremely annoying to defend against for Black. 12…g6 A very understandable decision, stopping White’s idea completely. The price to pay are weaknesses around the king. We should have a look at the alternatives that allow the knight to jump to f5: A) 12…b5 13.♘f5 ♘b6?! 14.a5 ♘a8 15.♕f3. Not the best knight ever on a8. White quickly won in FernandezLacasta Palacio, Barcelona 2017; B) 12…exd4 13.♗xd4 ♖e8 14.♘f5 ♗f8 15.♕f3.

analysis diagram

This could easily have been reached by 9.♗a2 a6 10.h3 ♕c7 11.♗e3 ♖b8. 12.♘h4! I believe that White should play this move against the closed Philidor if it is 673

And White is better. The threat of 16.♘h6+ is more than annoying for Black. 13.♘f3 Mission accomplished! The knight has done its job and returns to be relocated to g5 soon. The immediate threat is 14.♗h6 and 15.♘g5 and White crashes through on f7. 13…♘h5

674

Sample games

Black insists on holding the centre and prepares 14…♘g7 in reply to 14.♗h6. Taking on d4 was better, I think. 13… exd4 14.♘xd4 b5 15.axb5 axb5 16.b4 is a possible continuation. I prefer White, but Black is still quite alright in this position. 14.♗h6 ♘g7 14…♖e8 15.♘g5 ♗xg5 16.♗xg5 can’t be good in the long run.

15.♘g5 Stock sh loves to play 15.d5 here. I am surprised that the engine evaluates this as better than the game continuation. I am not sure about this at all – keeping the strong bishop on a2 open also has its merits. Let’s have a brief look at a possible line: 15…c5 16.♗c4 ♔h8 17.♕d3 ♖a8 18.♖ad1 f5 and Black has chances as well – at least it is quite complicated. 15…exd4 This helps White to get the queen into the attack quickly. 15…b5 was still playable for Black, with just a small advantage for White, e.g. 16.axb5 axb5 17.b4 ♗f6 18.♘e2 and it is still a game. 16.♕xd4 ♗f6 17.♕d2 b5 18.♕f4 675

If you like to compare chess to the game of football/soccer, the role of the queen is the central striker, scoring the goals. Here the queen clearly is on the way to get nearer to the goal (the black king!). 18…♕d8 Black needs the queen to defend. 19.♖ad1 The engine already evaluates the following sequence as tactically winning: 19.axb5 axb5 20.♗xf7+ ♖xf7 21.♘xf7 ♔xf7 22.♕xd6.

analysis diagram

Now ♖a7 is the decisive idea in this position.

676

Sample games

In the game, Parligras plays the most natural move: let everybody join in to the party. Black’s position can no longer be defended against all White’s pieces attacking, at least it is impossible to do so in a practical game. 19…♘e5 20.axb5 axb5 21.♕h4 ♗e6 The tougher defence was 21…b4 22.f4 ♘f3+ 23.gxf3 bxc3 24.bxc3 ♗e6 25.♗xe6 fxe6 26.c4. White is much better here as well, but it’s not going to be a checkmate attack to end the game shortly. 22.f4 ♘c4 23.♗xc4 bxc4 24.♔h1 If you play f2-f4, you are often going to play ♔h1 as well to stop disruptive checks. Black has nothing better than taking a pawn for his worries. 24…♖xb2 25.♗xg7 ♗xg5 26.fxg5 ♔xg7 27.e5 h5 I need to mention 27…d5 28.♘e4 ♖g8 29.♕h6+ ♔h8 30.♘f6 ♖g7 31.♖d4

analysis diagram

Here Black resigned in view of the coming ♘e4. The main message of this game: in the closed Philidor, operating with ♘f3h4-f5 ideas is very dangerous for Black. If you spot an opportunity to realize this plan, go for it. This applies in particular to scenarios after Black has played …♗b7, so the knight can never be captured. The knight is close to the black king, and expelling it with …g7g6 creates new weaknesses. A simple, yet very e ective plan. Game 25 SD 8.12 (B01) Viswanathan Anand 2767 Mikhail Demidov 2523 Pro Chess League rapid 2018 (6)

1.e4 d5 2.exd5 ♕xd5 3.♘c3 ♕d6 4.d4 ♘f6 5.♘f3 One main task for White against 3… ♕d6 is this position, where Black has a choice of lines. They are all quite comfortable for White, but require entirely di erent responses. Unsurprisingly, the 15th World Champion is up to the task. 5…a6 6.♗d3! This move and the connected idea to play ♘e4 is key to remember against 5…a6. 6…♗g4 7.h3 ♗h5 8.0-0 ♘c6 9.♘e4 ♘xe4 10.♗xe4

and the Arabian mate (my favourite mating pattern!) after ♖h4 and ♕xh7 is the winning idea. 28.exd6 ♖xc2 29.♕d4+ 677

678

Sample games

White has accomplished something: He has traded his badly-placed knight on c3 against the well-placed knight on f6. Why is c3 a bad square for the knight? Well, it is not so much that the knight would be ‘o side’, but it blocks the cpawn. This is important, as the c-pawn is a natural defender of the d4-pawn, and also might be used more actively with pushing c2-c4. An additional accomplishment is, of course, the strong bishop on e4, looking at both sides of the board. 10…e6 11.c3 ♗g6

Black quite often emphasizes queenside development in the Scandinavian, which makes queenside 679

castling quite a frequent guest in this opening. Here it is very dubious though, as White quickly launches an attack: 11…0-0-0 12.b4. This is both dangerous and simple to play – quite a terrible situation for Black. 12.♗xc6+ A very interesting decision. White is giving up the pair of bishops because he is gaining lots of time to increase his initiative. Something more standard like 12.♗xg6 hxg6 13.♖e1 was also quite good for White, but less forceful. Note that Black has some king position issues here as well. Going ‘long’ is questionable due to the hook on a6, while going ‘short’ will also not be completely safe in view of a knight possibly appearing on g5. 12…♕xc6 13.♘e5 ♕b6 This is more or less mandatory. Instead 13…♕e4 14.♖e1 ♕f5 15.♕a4+ b5 16.♕b3 ♗d6 17.a4 is excellent for White.

14.c4 This is the most direct and most aggressive move. 680

Sample games

White had a good alternative in 14.a4. Now Black should stop this pawn with 14…a5. Castling into it cannot be recommended, as 14…0-0-0 15.b4 is awful for Black; if Black plays something slow like 14…♗d6 the problem is 15.a5 ♕a7 16.♕a4+ and Black is in bad shape again. After 14….a5 White may just play 15.♖e1, correctly claiming that Black’s position is not harmonious. It’s hard to play … c7-c5 and obtain counterplay. 14…0-0-0 This is risky, but very principled. It immediately points at d4, the pawn weakness White has just created. 15.♗e3

This is a critical position early in the game. Black now had the fundamental choice between allowing c4-c5 and stopping it. 15…c5 Let’s check the alternatives. First we need to get 15…♕xb2? out of the way. This fails due to 16.♘xg6 hxg6 17.♖b1 ♕xa2 18.♕f3

681

analysis diagram

and Black is falling apart. However a playable alternative was 15…♗e7 16.c5 ♕xb2. This is also quite risky, obviously, but things are not entirely clear, and surely not easy to play: A) The tempting 17.c6 b6 18.a4 threatens to take on g6 and play ♕d3, invading on a6. Black has only one defence: 18…♖d6! (this, surprisingly, seems to hold) 19.♕f3 f6 20.♘xg6 hxg6 21.♖fc1 ♖h4! is a possible line that is OK for Black; B) 17.♘xg6 fxg6 (17…hxg6 18.♕f3 is worse as f7 is just hanging, in addition to 19.♖b1 being threatened) 18.♕g4 ♕b5 19.♕xe6+ ♕d7 20.♕b3 and White keeps the better chances, using the open b- le to attack. 16.d5 Here Anand had a stronger move, one that he would have played for sure if he had had more time to consider his options (it was a rapid game after all): 16.b4!.

682

Sample games

analysis diagram

The kind of move that makes you smile when you are allowed to play it. White gets an extremely strong attack now. 16…♕xb4 is relatively best (other lines like 16…cxb4 17.c5 ♕a5 18.♗f4 or 16…cxd4 17.c5 ♕xb4 18.♗d2 ♕b2 19.♖c1 crumble quickly): 17.♘xg6 hxg6 18.♖b1 ♕xc4 19.♕f3 ♕d5 20.♕xf7 ♕d7 21.♕f4 and White is much better. 16…♗d6? Now White is allowed to transform his dynamic advantage into a long-term, static one. Black should have taken the pawn, it seems: 16…♕xb2 17.♗f4 ♗d6 18.♖e1 and White has compensation for sure, but no big advantage. Of course this is not easy to gure out, in particular in a rapid game. 17.♘xg6 hxg6

683

18.♕g4! Strong! Now White retains the clearly better position long-term. The next moves are quite forced. 18…♖he8 19.b3 ♔b8 20.♖ad1 White is better due to his ongoing pressure and the awkward black bishop on d6. There is also a potential weak square on d5, ideal for White’s rook. 20…e5 Now 20…♕c7 21.dxe6 ♖xe6 22.♖d5 is a possible line to illustrate the power of White’s rook on d5. There is pressure in this middlegame, but the endgames might be depressing for Black as well, with the c5- and g7pawns being on dark squares. In the game Black decides to close everything down, but giving White a protected passed pawn is a concession. 21.f4 Anand’s move is very consistent, but 21.♖b1 was also very strong. White aims for b3-b4, as happens after 21… ♕c7 22.♖fc1 ♔a8 23.b4, with a great attack. 21…exf4 22.♗xf4 ♖e7 23.♕g3 ♗xf4 24.♕xf4+ ♕d6 25.♕xd6+ ♖xd6

684

Sample games

A very one-sided endgame is on the board now. White is enjoying a huge advantage. 26.♖fe1 ♖dd7 27.♖xe7 ♖xe7 28.♔f2! ♔c8 29.♖e1 ♖xe1? This move loses, but the alternative was not great either. After 29…♖c7 30.♖e8+ ♔d7 31.♖g8 White should win as well, it just takes longer. 30.♔xe1 ♔d7 31.♔e2 ♔d6 32.♔f3 ♔e5 33.♔e3 ♔f5 34.♔f3 Here there was a much more direct win: 34.b4 b6 (34…cxb4 35.♔d4 is quite easy) 35.bxc5 bxc5 36.♔d3 and the king walks to the a-pawn. Anand’s move does not spoil anything though. 34…♔e5 35.♔g4 b6

685

36.a4? This spoils the win. King and pawn endings are very tough to play with little time as they rely too heavily on move-by-move calculation. Intuitive play, typical in blitz and rapid, is not much help here. Well, this is not a book on endings, so I won’t provide any extensive analysis here, but will keep it short with just some key lines. White keeps the win with 36.h4 ♔d6 37.♔f4 ♔d7 38.♔e5 ♔e7 39.d6+ ♔d7 40.♔d5 a5 41.a4 f6 42.g3 f5 43.♔e5. 36…f6? The move to draw was 36…a5 37.h4 ♔d6 38.♔f4 ♔e7 39.♔e5 ♔d7 40.g4 ♔e7. There is no way further now, e.g. 41.h5 gxh5 42.gxh5 ♔d7 43.♔f5 ♔d6 44.♔g5 ♔d7 45.h6 gxh6+ 46.♔xh6 ♔d6 47.♔g5 ♔e7 48.♔f5 ♔e8 49.♔f6 ♔f8 50.d6 ♔e8 51.d7+ ♔d8 52.♔xf7 ♔xd7. One important line to check is 36… ♔d6? 37.a5!, which wins for White: 37…bxa5 38.♔f4 ♔d7 39.h4 ♔d6 40.g4 ♔d7 41.♔e5 f6+ 42.♔f4 ♔d6 43.h5 gxh5 44.gxh5 ♔e7 45.♔f5 ♔f7 46.d6. 37.h4 a5 38.h5 f5+ 39.♔g5 gxh5 40.♔xh5 ♔f6

686

Sample games

41.d6! ♔e6 42.♔g5 ♔xd6 43.♔xf5 ♔e7 44.♔g6 ♔f8

The di cult part is already done, but now Anand errs. Even with very little time it is easy for such a strong player to calculate this, so this is quite surprising. 45.g4? 45.g3! ♔g8 46.g4 ♔f8 47.♔h7 ♔f7 48.g5 was the way to win. 45…♔g8 46.♔g5 46.g5 ♔f8 47.♔h7?? ♔f7 even loses for White! 46…♔h7 47.♔f4 ♔g6 48.g5 ♔h5? 48…♔f7 was the drawing move. Again, it was a rapid game. 49.♔f5 ♔h4 687

50.♔g6! ♔g4 51.♔xg7 ♔xg5 52.♔f7 ♔f5 53.♔e7 ♔e5 54.♔d7 ♔d4 55.♔c6 ♔c3 56.♔xb6 ♔b4 57.♔c6 ♔xb3 58.♔b5 Black resigned. Against 5…a6, Anand’s move 6.♗d3 is an excellent antidote. There are more complicated lines available, but the idea to play ♘e4 is very e ective. In the opposite-side castling scenario White is quicker to attack, as we saw in the game. Anand always had an advantage, but at some point could have even played more powerfully with 16.b4!. The 3…♕d6 Scandinavian is a line we should be happy to face. Game 26 SD 8.7 (B01) Alexander Moskalenko 2522 Dmitri Petrov 2222 Orsha 2018 (2)

1.e4 d5 2.exd5 ♕xd5 3.♘c3 ♕d8 Currently the queen retreat to d8 has the relatively best theoretical reputation. With precise play White retains a small advantage, but Black’s position is very solid. 688

Sample games

4.d4 ♘f6 5.♘f3 ♗g4 6.h3 ♗xf3 7.♕xf3 White has gotten the pair of bishops without much of a struggle. The coming structure is quite common in our repertoire, for example the CaroKann Two Knights may lead to very similar play. 7…c6 8.♗e3 e6 9.♗d3 ♘bd7 10.0-0-0 ♗b4 11.♘e2

White has avoided further piece trades, as he should when enjoying an advantage in space. Black, on the other hand, would not mind trading, especially minor pieces. In a majorpiece endgame, for example with a queen and two rooks for each side, Black could even try to double against the d-pawn and go for …c6-c5 or …e6e5. This scenario is impossible with minor pieces on the board. 11…♕a5?! This is the start of a wrong plan, as we will see in the game. Black should play 11…♘d5, as examined in the theoretical section. White retains a small plus, but Black is not in any immediate trouble. 12.♔b1 0-0-0 13.g4

689

White certainly also considered 13.c4. Seizing control over d5 immediately also looks very promising. The black queen does not have options on the fth rank as all squares are controlled now. In the game White allowed 14… ♕d5, trying to play against the queen in the centre. This strategy is also perfectly ne. 13…h6 14.h4 ♕d5

15.♕g3 White does not trade, of course. Besides general considerations (more space – don’t trade), in this situation the black queen might become a tactical target. 15…♗d6 With 15…♘b6 16.♖hg1 ♘c4 17.♗c1 Black activates his knight, but it is not stable on c4 at all. Stock sh already suggests 17…b5, but this can’t be healthy in the long run. 16.♘f4 Provoking Black to take the knight and getting the two bishops against two knights. The alternative 16.f4 looks ne as well, intending to su ocate Black with pawns: 16…h5 17.c4 ♕a5 18.g5 ♘g4 690

Sample games

19.c5 ♗b8 20.♗d2. Black is in deep trouble here. 16…♗xf4 A huge concession, but Black had no satisfactory solution. 16…♕a5 is answered by 17.♕f3 with g4-g5 to come. 17.♕xf4 You would like to play the more natural 17.♗xf4?!, but after the cold-blooded 17…♕xd4 White has no great discovered attack. 17…g5 Black needs to get concrete play going. Playing in a slow way and allowing White to organize his bishops sets no practical problems. 18.♕g3 ♘b6

19.b3? It was much better to play the cautious 19.f3, taking away the e4-square. The next step is to expand with b2-b3 and c2-c4, kicking the queen back. Note that Black has absolutely no good outposts for his knights. 19…♘e4! The best chance to put up a ght. 20.♗xe4 ♕xe4 691

At least Black has managed to get rid of the bishop pair. Still, the e3-bishop is better than the knight, but most important in the next couple of moves is precise move-by-move calculation. 21.hxg5 Stock sh suggests 21.♖de1! ♕d5 (or 21…♕g6 22.h5 ♕h7 23.f4) 22.c4 ♕a5 23.♖c1, both with a substantial white advantage.

21…♘d5? Here 21…hxg5! 22.♖xh8 ♖xh8 23.♗xg5 ♘d5 24.♔b2 is an extra pawn for White, but Black’s pieces are all actively placed. White still has a di cult task ahead of him. 22.♗c1? Missing 22.gxh6! ♘c3+ 23.♔b2 ♘xd1+ 24.♖xd1 ♖dg8 25.♗f4 with fantastic compensation for the exchange. White is much better and even winning by engine standards. 22…hxg5 23.♖he1 23.♖xh8 ♖xh8 24.♗xg5 fails to 24… ♖g8 and Black is taking over the initiative. 23…♕g6 24.♖e5

692

Sample games

White has lost the huge advantage that he had out of the opening. I still think the position is easier for him to play. The text move is not that easy for Black to answer correctly. 24…♖dg8 24…♘b4 was correct. Now White has nothing better than 25.♖e2, after which 25…♘d5 holds the status quo. 25.♔b2 I don’t know why 25.♖xg5 was avoided. White seems to be better after 25…♕e4 26.♖xg8+ ♖xg8 27.g5. Black is quite active, but still has to prove adequate compensation. 25…♕f6 26.♖e4 ♖h4? 26…♕h6 is better. Now 27.c4 ♘f6 28.♖e2 ♕h3 shows the di erence to the game continuation. It is about equal. 27.c4 Now the black knight lacks a good square, like f6 in the previous line. 27…♘c7 Instead, 27…♘b4 28.a3 ♘a6 29.♖e5 is also quite sad for Black.

28.♖e5 Probably both players were having some time issues around here as this and some of the next moves are not the best. 28.f4! was a strong move for White; 28…gxf4 29.♖xf4 ♕e7 30.g5 gives him a huge initiative. 28…♕h6 29.♖a5 a6 30.♗e3?! 30.♖c5 was better, preventing 30…b6. 30…f6? This move fails against White’s powerful reply. Instead, 30…b6 was called for: 31.♖a3 (please note that 31.♖e5?? f6 32.♖e4 f5 is a way for White to lose quickly!) 31…♔b7. Here Black is obviously ne due to the o side rook on a3. Maybe Black missed 32…f5 in the line above and thus avoided 30…b6.

693

694

Sample games

resigned. Against the Scandinavian with 3…♕d8 you can’t expect to get a huge advantage, but it will look like this game quite often. It’s comfortable, with chances to put some pressure on Black. Here the set-up with 11…♕a5 was questionable and led to a nice initiative for White that – after some adventures – was converted into a full point. 31.d5! A successful pawn breakthrough on a well-protected square is always a very attractive solution. 31…♖d8? The engine’s defence is still tough to crack: 31…exd5 32.cxd5 ♖h3! (32… ♘xd5? 33.♖dxd5 cxd5 34.♕d6 leads to mate) 33.♕d6 ♖xe3 34.fxe3 ♖d8 35.♕g3 and White is much better, but Black is still ghting. 32.♗b6 Now it’s over in just a couple of moves. 32…♖d7 33.dxc6 ♖f7

Game 27 SD 16.5 (B01) Viswanathan Anand 2765 Joel Lautier 2660 Biel 1997 (1)

1.e4 d5 The Scandinavian is a rare guest on the very top level. Note that in 1997 this game was World No. 2 vs World No. 20 – nowadays these ratings would roughly mean No. 10 vs No. 80. The Scandinavian is not terrible at all, by the way, but White has well established ways to get a small edge. 2.exd5 ♕xd5 3.♘c3 ♕a5 4.d4 ♘f6 5.♘f3 c6 6.♗c4 ♗f5 7.♘e5 e6 8.g4 It is absolutely essential to learn a principled line against the Scandinavian. Slow play allows Black to equalize quite easily. Our repertoire choice keeps the pressure up. 8…♗g6 9.h4 ♘bd7 10.♘xd7 ♘xd7 11.h5 ♗e4

34.♖d8+! 1-0 An attractive nish! The b-pawn is unstoppable and therefore Black 695

696

Sample games

Simple’ approach made me avoid this position. 14.♗d3 ♘d5

12.♖h3 In the theoretical section I recommend to play 12.0-0. I prefer that slightly over Anand’s move, but the later World Champion’s choice also leads to interesting play. What follows is one of the most attractive attacking games by the Tiger from Chennai. 12…♗g2 Black tries to provoke the rook to go to an unstable square. 13.♖e3 Anand has spotted a creative idea, and is happy to oblige. 13…♘b6 Black is going for the most principled line. A very understandable decision, but it seems to lead to some trouble. The line I didn’t like so much for White and the reason why I recommend 12.00 is 13…♗d5!. With the king on g1 you could just take on d5 now with the knight. Now after 14.♗e2 0-0-0 the position is rather unclear to me. The engines evaluate it as roughly equal, but it is surely tough to play, as the exotic rook position on e3 poses problems. Sticking with the ‘Keep It

697

15.f3 This was Anand’s excellent idea. The bishop on g2 is now cut o from the black camp. Of course, Black may take the exchange on e3, but this trades o his only other active piece and possible support for the bishop. Unsurprisingly the game gets extremely sharp now, as slow moves by Black simply lose to ♔f2, to pick up the bishop with the king. 15…♗b4? It sounds a bit dramatic, but after this move Black is lost against perfect play. There was no equality anymore, but tougher defences were available: A) 15…♘xc3 16.bxc3 ♕xc3+ 17.♗d2 ♕xd4 18.♖b1 (the simpler 18.♔f2 is also better for White, but the rook move is even stronger) 18…♗d6 (instead, 18…0-0-0 gets refuted by 19.♕e2 ♗h3 20.♗a6!) 19.♔e2! 0-0-0 20.♕g1 ♗h3 21.♕h1 and White is clearly better after he picks up the stranded bishop; 698

Sample games

B) The most resilient is probably 15… ♘xe3 16.♗xe3 ♗a3 (trying to muddy the waters more than anything else, but there is nothing great available anyway) 17.bxa3 ♗xf3 18.♕d2 ♗xg4 19.h6. An important point, to weaken the dark squares. White is still clearly better here, but it is messy at least. 16.♔f2 ♗xc3 Lautier goes for the critical line. 16… ♘xe3 17.♗xe3 ♗h3 18.♕h1 would have been far too easy for White to play. 17.bxc3 ♕xc3 18.♖b1 ♕xd4 19.♖xb7

Modern engines quickly evaluate this as completely winning for White. But take a look at it with a human eye – it’s a complete mess! Not at all easy to play and evaluate over the board. Black now has a tough choice. 19…♖d8 This is a good defence, challenging Anand to nd the brilliant game continuation. The alternatives are very well worth checking though. Let’s look at 19…♗h3 rst. This leads to great lines: 20.♖xf7! (according to Anand’s annotation he had seen this 699

shot, but did not calculate further) 20…c5!.

analysis diagram

Covering the queen and threatening to take the rook now. 21.♖f5!! (according to Anand, Lautier saw this great move during the game! It’s a bit of a shame that the prosaic 21.c3 wins as well) 21…c4 (if Black just takes with 21… ♘xe3 the simple 22.♗xe3 ♕d7 23.♖xc5 is enough to win. The bishop on h3 is still hopeless and the white pieces dominate completely) 22.♖f4!! ♕b6 23.♖xc4 and the rook has nished its job. White has played ♖a1b1-b7-f7-f5-f4-c4 in the last seven moves! What a brilliant line! After something unconcrete like 19… g6 White wins with the simple 20.c4 ♘xe3 21.♗xe3 ♕e5 22.f4 and the bishop on g2 will fall with a lost position for Black. 20.h6!! gxh6 We’ll see the point of it all on the board! Knight moves did not help: A) 20…♘xe3 21.♗xe3 ♕e5 22.hxg7 ♖g8 23.♕c1 ♗h3 24.♕a3 ♕h2+

700

Sample games

25.♔e1 ♖d7 26.♖xd7 ♔xd7 27.♕f8 and White wins; B) 20…♘f4 21.hxg7 ♖g8 22.♗a3 ♘d5 23.♕e1 ♖xg7 24.♔xg2 ♘xe3+ 25.♕xe3 ♕xe3 26.♖e7+ ♔f8 27.♖xe6+ ♔g8 28.♖xe3. This position is even more winning than you might think at rst. Bishop to f6 is an absolutely killing idea.

And here Lautier resigned. A great attacking miniature by Vishy Anand! The 3…♕a5 Scandinavian is a very dynamic opening. We need to play aggressively to get something, so the game will be sharp and often quite tactical. On move 12 I suggest to play a bit more conservatively, compared to Anand’s choice. However, if you like to add something really exciting to your repertoire, go for the game continuation – you might win a similar attacking gem like Anand did. Game 28 AL 7.14 (B03) Thai Dai Van Nguyen 2441 Jules Moussard 2573 Karlsruhe 2017 (4)

21.♗g6!! How many lame wins would you trade in for a win like this? This move must have been such a pleasure to play! 21…♘e7 Taking the queen leads to mate: 21… ♕xd1 22.♖xe6+ ♘e7 23.♖bxe7+ ♔f8 24.♗xh6+ ♔g8 25.♗xf7#. If Black covers f7 for the moment with 21…♕f6 White wins with 22.♗xf7+ ♕xf7 23.♖xf7 ♘xe3 24.♕xd8+ ♔xd8 25.♗xe3 ♗h3 26.♖xa7 and the bishop is still stuck on h3. 22.♕xd4 ♖xd4 23.♖d3 There are multiple wins now. 23.♖xe6 ♖d7 24.♖xd7 ♔xd7 25.♗xf7 worked as well. 23…♖d8 24.♖xd8+ ♔xd8 25.♗d3

701

1.e4 ♘f6 2.e5 ♘d5 3.d4 d6 4.c4 ♘b6 5.exd6 cxd6 This is Black’s more combative and ambitious choice. Against our set-up, the so-called Voronezh Variation, Black is struggling to equalize, as you can see in the theoretical section. While Black may not be equal, the play always stays very complicated, as you will witness in the game examined here. 6.♘c3 g6 7.♗e3 ♗g7 8.♖c1 0-0 9.b3 The starting position of the Voronezh. There is no sensible black alternative before this point. 9…e5 This move and 9…♗f5 are the most reliable for Black. 10.dxe5

702

Sample games

10.♘f3 is my recommendation in the theoretical section, to keep the game a bit calmer. It will probably later lead to the same position type as in our game though, so it makes a lot of sense to examine it anyway. 10…dxe5 11.♕xd8 ♖xd8 12.c5

This is the position type I was referring to above. After c4-c5 White gains space on the queenside, where he has got a healthy majority. There are weak squares on d5 and d6 that are potential outposts for white pieces, in particular for knights. Black’s asset is the extra centre pawn and the pawn duo that he may set up with a later …f7-f5. 12…♘6d7 13.♗c4 ♘c6 14.♘f3 h6

After this prophylactic move, White has tried a range of options, but it is not apparent that any of them lead to a substantial plus. This position is the main reason why I opted for the calmer 10.♗e2 as I felt that you would need to explore even more from here. Of course you can do that, but for an opening that you may face once in a year it is not very economical to analyse beyond move 20. Black has an entirely di erent approach in 14…♘a5 15.♗e2 b6. After the engine suggestion of 16.0-0 bxc5 17.♘e4 ♘c6 18.♘xc5, White is slightly better, but Black holds his own. 15.♘b5 15.♘e4 is the main try. Now 15…♘a5 16.♗e2 ♘c6 17.♘d6 f5 18.♗c4+ ♔h7 19.h4 ♘f6

analysis diagram

leads to a complicated game where White is not better, as far as I can see. 15…e4 After 15…♘a5 16.♗e2 ♘c6 17.0-0 we see a di erence to 15.♘e4 as now the knight has the additional idea to go to c7. White is slightly better here. 703

704

Sample games

16.♘d2 ♘de5 Black went for a concrete solution based on his activated pieces. With very precise play he should equalize. 17.♘xe4 There is nothing better. 17…♘b4 18.0-0 ♘xa2 19.♖cd1

Black has regained the pawn, but he still is the one to be precise. 19…♗g4 This is imprecise, but why is impossible to gure out without machine help. The open-centre position with all pieces on the board is extremely di cult to calculate correctly. The engine nds a path to equality with 19…♗f5. It is very complicated, so I’ll just show one line that looks possible in a human game: 20.♘ed6 ♘xc4 21.bxc4 a5. The combination of the a-pawn and the bishop on g7 is very reminiscent of the Grünfeld. Black gives up the b7-pawn but acquires enough play to draw: 22.♘xb7 ♖xd1 23.♖xd1 ♘c3 24.♘xc3 ♗xc3 25.c6 ♖c8 26.♘d6 ♖xc6 27.♘xf5 gxf5 and it’s equal. 705

20.f3 ♘xc4 21.bxc4 ♗d7 Black has also tried 21…♗e6, but after 22.♘c7 White was very much better in Aranda Gonzalez-Mosquera, Barcelona 2012. 21…♗f5 22.♘ed6 a5 needs to be compared to 19…♗f5, of course. White’s extra move f2-f3 makes a huge di erence, as the following line shows: 23.♘xb7 ♖xd1 24.♖xd1 ♘c3

analysis diagram

25.♖d2!. This is now possible due to 25…a4 26.c6 a3 27.♘xa3! ♖xa3 28.g4 ♗e6 29.♘c5 and White is clearly better. 22.♘c7 ♖ab8 23.♘d5 Now White has the dual threats of 24.♘f6, to win the bishop, and 24.♖f2, to trap the o side knight. 23…♗c6 23…♗e6 24.♗f4 ♖a8 25.♗c7 is very similar to the game. 24.♗f4 ♖a8 25.♗c7 ♖f8 25…♗xd5 26.cxd5 ♖d7 27.d6 makes it easier for White. The idea is c5-c6, followed by ♘e4-c5. 26.♖f2

706

Sample games

This nets a pawn and keeps a large edge. Stock sh wants to play 26.♘ef6+ ♔h8 27.♗d6, which is even stronger. 26…f5 27.♖xa2 fxe4 28.fxe4 This position is forced after 26 ♖f2. 28…♖ae8 29.♖e2 ♗a4 29…♖f7 30.♗d6 ♗e5 was the toughest defence, but of course Black is still su ering. 30.♖d3 ♗e5 31.♗d6 ♖f7

32.h3 Here White had a clean way to win: 32.♗xe5 ♖xe5 33.♖f3 and the king will join with a march to d4. This would end the game quite quickly. 32…a6 33.♖a3 ♗c6 34.♖f3 The winning idea from above, 34.♗xe5 ♖xe5 35.♖f3, is in fact not entirely clear in this position: 35…♖xf3 36.gxf3 ♔f7 37.♔f2 ♖h5 and it’s still hard work to convert. 34…♗d4+ 35.♔f1 a5 36.e5 ♖xf3+ 37.gxf3 ♔g7

38.e6? This is a mistake, but the reason why is impossible to gure out. Instead 38.♘c7 was winning convincingly. 38…a4 39.e7 a3? The reason why 38 e6 is bad is 39… ♔f7. This miraculously holds a draw. The key line is 40.♘c7 ♖c8 41.e8♕+? (41.♘e6 or 41.♘b5 still draws) 41… ♗xe8 42.♖e7+ ♔g8 43.♘xe8

analysis diagram

43…a3 44.♖e2 ♗b2! and the game has turned around! 40.♘c7 Now White is back on track to win. 40…♖c8 707

708

Sample games

It’s move forty. However 40…♔f7 41.♘xe8 ♔xe8 42.♖a2 ♗b2 43.f4 ♗e4 44.c6 ♗xc6 45.♗xa3 wouldn’t have saved anymore. 41.♘e6+ 1-0 The Voronezh Variation is easier to play for White, but it remains very complicated. After 9…e5 you may play in true ‘Keep It Simple’ fashion with 10.♗e2 (theoretical section), or explore the complications presented in this game.

game is easy to remember and not very move-order critical. 6.♘c3 ♗e7 7.♗d3 0-0 In the theoretical section I didn’t check this move order, but we will soon transpose. Black’s moves are …♘c6/… ♗g4/…♗f6 and …0-0 anyway, there is nothing else available. 8.♘ge2 ♘c6 9.0-0 ♗f6

Game 29 AL 7.5 (B03) Tamaz Gelashvili 2563 Ferenc Berkes 2654 Minsk 2017 (7)

1.e4 ♘f6 Alekhine’s Defence is an opening you won’t face very often. When you do, it might happen against a specialist who knows it well or against someone who wants to surprise you. For both cases it makes sense to learn a set-up that is easy to handle. 2.e5 ♘d5 3.d4 d6 4.c4 ♘b6 5.exd6 See the note to move 1. The Exchange Variation leads to clearer positions compared to the messy Four Pawns Attack or the strategic complexities after 4.♘f3, where Black has a wide range of options. 5…exd6 This was the choice of Danish legend Bent Larsen. The symmetrical recapture is the solid option for Black, but it poses very few problems for us to solve. The white set-up used in the

10.b3! One of the few move order nesses to remember. White needs to deal with the …♘b4 idea properly. For more about this point, please check the theoretical section. 10…♖e8 11.♗e3 ♗g4 12.♕d2 In the theoretical section I went for 12.h3, which leads to similar positions and is of equal value. The main decision for Black is to decide between allowing White to play d4-d5 or pushing the d-pawn himself. 12…♗xe2 Black does not have any clear purpose for this bishop and just takes the knight. This is one positive point about the ♗d3/♘ge2 set-up – there is no pin on the f3-knight. 13.♘xe2

709

710

Sample games

sides of the board and will probably push g2-g4 next. 16.♖ae1 It is premature to play 16.a4?!

13…d5 This is strategically very sound play. Black places his d-pawn on a light square and xes the pawn on d4, the target of Black’s f6-bishop. He should try to increase the pressure on this pawn, maybe with the help of a knight regrouping to f5. This is not easy to do, as we will see in the game. If Black does not play …d6-d5, our plan is to push our d-pawn after moving the a1-rook rst. 14.c5 Grabbing useful space. Black may have a target on d4, but we have space and the pair of bishops. 14…♘d7 Black intends to manoeuvre this knight to e6 via f8. The other knight jump, 14…♘c8, is a frequently seen idea in the Alekhine’s. The knight has …♘c8-e7-f5 in mind, but here this is di cult to realize: 15.b4 a6 16.a4 and now 16…♘8e7 17.b5 ♘b8 is just too passive. 15.b4 a6 This is an important resource for Black. Instead, 15…♘f8 16.b5 ♘e7 17.f4 is awful for Black. White is better on both

due to 16…a5! 17.b5 ♘b4 and now Black is able to trade material. The d3bishop can’t be saved with 18.♗b1? due to 18…♘xc5!. The a4?! a5! sequence is important to remember. 16…♘f8 17.♗b1 17.♗c2 was ne as well. One idea is to direct this bishop against d5. In the game White goes for a2-a3, followed by ♗b1 and a possible ♗a2 later. ♗c2 is a bit quicker, but playing the pawn to a3 is useful to cover b4 and free up the queen to move. The second plan to go for is to gain space on the kingside with pushing the f- and possibly g-pawn. 17…♕d7 17…♘e6 is also answered by 18.a3. This is very similar to 18…♘e6 below. 18.a3 b6?! This is not the best, it seems. The straightforward 18…♘e6 looks better. After, for example, 19.♗f5 g6 20.♗h3 ♗g7 21.g3

711

712

analysis diagram

Sample games

analysis diagram

White still is for choice, but Black is quite solidly placed. 19.h3 19.cxb6 cxb6 20.♖c1 is the computer’s suggestion, but the play in the actual game is ne as well. 19…bxc5 20.bxc5 20.dxc5 was the sharper choice, and also quite promising. Gelashvili goes for the symmetrical recapture, intending the awkward pin after ♗b1c2-a4. 20…♖ab8 21.♗c2 ♘e7

Stock sh wants to play more aggressively with 22.♘f4 g6 23.g4. This is not an attacking move, but prophylaxis against …♘f5. ♕d1 followed by ♗a4 is an idea to take note of – Black is under heavy pressure. 22…♕b5 Maybe White underestimated this move. The queen has to go back now. 23.♕d2 Grabbing with 23.♕xc7?! ♘e6 24.♕g3 ♕c4 leads to trouble for White. 23…♕d7 Black decides to retreat as well, o ering a repetition. White now has the promising ♘f4 option again, just like on move 22. If Black tries to stay on b5 he will face something like 23…g6 24.♘f4 ♗g7 25.♖b1 and White stays on top. 24.♘c3 We looked at 24.♘f4 above. Gelashvili’s move is a bit less precise, as we will see. 24…♘f5? Black blunders the most important pawn on the board. He should have played 24…g6. After 25.g4 we may compare this position to the one analysed for 22 ♘f4. The di erence is that Black now has 25…h5, which is impossible with the knight on f4. With 26.f3 White is still better, but it’s a game, of course.

Black can’t allow the pin with ♗a4, so there is no real alternative. 22.♕a5 713

714

Sample games

25.c6 ♕c8 Black is unable to cover both d5 and f5. 26.♘xd5 White has not just gained one pawn, but domination with the monster on d5. Black is lost, and Gelashvili converts easily. 26…♗d8 27.♗f4 ♘e7 28.♕d3 ♖b5 29.♘c3 ♖a5 30.a4 ♘fg6 31.d5 ♔f8 32.♗g3 ♕b8 33.h4 h5 34.♖e3 ♕b4 35.♖fe1 ♕g4 36.♖e4 ♕f5 37.♕d4 Black had seen enough and resigned. The symmetrical response to the Alekhine’s Exchange leads to a slight edge for White and an easy-to-play position. In the game White had good control when Black blundered the d5pawn and with it the game. Game 30 VO 5.9 (B00) Vasilios Kotronias

control games for the more obscure lines. 1…d6 2.d4 ♘c6 A really odd move order! Of course, White now could also play 3.d5, but the simple 3.♘f3 takes us back into our repertoire. 3.♘f3 ♗g4 4.♗b5 I recommend 4.d5, as it is just too tempting to invite 4…♘e5 5.♘xe5!! ♗xd1 6.♗b5+ with excellent play for White. Please see the theoretical section for all the juicy details. 4…a6 5.♗xc6+ bxc6 6.♘c3 ♘f6 This position is usually reached via 1… ♘c6 2.♘f3 d6 3.d4 ♘f6 4.♘c3 ♗g4 5.♗b5 a6 6.♗xc6+ bxc6. 7.h3 ♗h5 8.♕d3

1.e4 It needs to be mentioned that this was a rapid time control game. It is not easy to nd higher-level classical time

Black’s main issue in this type of position is the di culty to ght the white centre with pawn breaks. White has much more space so Black should attack the centre with pawn pushes like …e7(e6)-e5, …d6-d5 or …c6-c5. All are very hard to achieve without weakening the pawn structure even further, if possible at all without losing material. 8…e6

715

716

2580 Angelos Vouldis 2515 Psichiko rapid 2002 (5)

Sample games

There is no other way to get developed. 9.♗g5 Very natural play, keeping the tempo up. 9…♗e7 10.g4 ♗g6 11.♕e3 I also examined 11.0-0-0, ignoring the bishop on g6 being pointed at the queen: 11…d5 12.♗xf6 ♗xf6 13.♔b1. This is not a routine move. It covers a2 in case of trades on e4, followed by … ♕d5. White is slightly better now, but I think the game move order is even more promising. 11…♘d7 Black opts for a cautious move. It looked more challenging to play with …d6-d5, but this pawn move has its drawbacks as well: A) 11…0-0 12.0-0-0 d5

analysis diagram

13.♘e5 (this is a key idea to remember. The knight immediately attacks the weak pawn on c6) 13…♘xe4 14.♗xe7 ♕xe7 15.h4 (taking on g6 and e4 also was good enough for an advantage, but hunting the bishop is promising, too) 15…♘xc3 16.♕xc3 f6 17.♘xg6 hxg6

717

18.h5 g5 19.♕xc6 with a comfortable advantage; B) 11…d5 12.♘e5 ♘xe4 13.♗xe7 ♕xe7 14.0-0-0 is very similar to 11…00. 12.♗xe7 ♕xe7 13.h4 I prefer 13.0-0-0 in this position. White keeps his pawn structure more exible compared to 13.h4 h5! 14.g5. One sample line to illustrate this point: 13… d5 14.♖he1 ♖b8 15.a3 ♕d6 16.♘d2. Now f2-f4-f5 comes into consideration. White is much better.

13…f6 Surprisingly, Black avoids 13…h5!. White is still better after 14.g5, but the game looked much worse. 14.0-0-0 ♗f7 Black should try to play …e6-e5 in this position. Here it fails due to 14…e5 15.h5 ♗f7 16.♘h4. 15.♘d2?! This allows Black to play …e6-e5. White needs to get e4-e5 and h4-h5 in to kill the f7-bishop, so 15.e5! ♖b8

718

Sample games

analysis diagram

16.h5! and Black will have a very hard time to ever activate the bishop, for example: 16…fxe5 17.dxe5 d5 18.g5. This threatens 19.g6!. Black is in bad shape. 15…0-0 15…e5 would have nally taken a claim in the centre, but maybe Black had some bet going that he couldn’t move a central pawn to the fth rank up to a certain move number?! Anyway, White can no longer prevent …e6-e5 now, so he might as well do it one move later. 16.f4 a5 16…e5! – I know, the broken record… 17.g5 ♗h5 17…e5 18.gxf6 ♘xf6 19.dxe5 dxe5 20.f5 is a slightly better version of the line in the notes to move 19 for Black. White’s attack on the g- le will still be very hard to defend against. 18.♖dg1 e5 Finally, but it’s too late now.

719

19.dxe5 19.gxf6 ♘xf6 20.dxe5 dxe5 21.f5 was very strong as well. White will pile up on the g- le while Black has no play at all. 19…dxe5 20.f5 a4 21.a3 A good, human move. 21.♘xa4 is pointed out by the engine, and is indeed strong. Still, Kotronias’ choice is ne, keeping all the pros in his position and stopping counterplay. 21…♔h8 22.♘f1 Doubling on the g- le was strong as well. 22…♘b6 23.gxf6 ♖xf6 24.♕g5 ♘c4? This is incorrect, but very understandable, especially considering this was a rapid game. Instead 24…♗e8 25.♘e3 was very depressing for Black. ♖h3-g3 is on the cards. 25.♕xh5 ♖b8

720

Sample games

26.♕e2 The cold-blooded engine defence 26.♖h3 ♖xb2 27.♘e3 ♘xe3 28.♔xb2 ♘c4+ 29.♔c1 ♕xa3+ 30.♔d1 dodges all the bullets and wins. Tough to play, even with lots of time on the clock. Kotronias’ move is still winning though. 26…♘xb2 27.♘b1 ♕c5

(this is the only move to win and not lose! For instance: 29.h5?? ♘d3+ 30.♔d1 ♘f4 and White is toast); B) 28.♖g3 ♘c4 29.♘fd2 is the second winning line. 28…♖d6! Ouch! What a shot… White is completely lost now! It is fascinating that the game ends at exactly this moment in the database, but the result is a win for White! This is mysterious, but my guess is that the result is correct and only the notation stopped here as it is a rapid game that most likely was not transmitted live. Black might be winning by computer standards, but over the board anything can happen. So why is it a winning position for Black? 28…♖d6 29.♘bd2 ♘d3+ 30.♔d1

White to play and win! This is not that easy to play and in fact there are some mines you may easily step on – as we will see! 28.♕e3?? White was winning with… A) 28.♘fd2 (this is convincing, stopping 28…♘c4) 28…♖d6 29.♖g3!

30…♘f4! (this shot for example is easy to miss!) 31.♔e1 ♖xd2! (or this one!) 32.♔xd2 ♖d8+ 33.♔e1 ♕xc2 34.♘d2 ♖d3. Not a trivial line at all – maybe Black missed something along the way and indeed lost.

721

722

analysis diagram

Index of players

Damiano 13 Demidov 337

Index of players (numbers refer to pages) A Aagaard 241, 243 Adams 218, 220 Akobian 248, 262-263, 265 Anand 60, 69-70, 72, 74, 151-152, 167168, 170, 192, 256-258, 308, 311, 337, 339-341, 345-348 Andriasyan 66 Aranda Gonzalez 350 Aronian 11, 44, 50, 52-53 B Barsky 293 Bartholomew 296 Bauer 328-329 Berkes 170, 174, 351 Bogoljubow 35 Bok 155 Botvinnik 186 Burg 106 Bu Xiangzhi 62 Byrne 115 C Carlsen 24, 83, 104, 110, 148, 150-152, 155, 167-170, 192, 246-248, 262, 264, 320 Caruana 246 Clement 318 Colon Garcia 156 D Dambacher 29 723

E Eljanov 320 F Fernandez 335 Fischer 185-186, 299 Franco Ocampos 61 G Gao Rui 164 Gara 259 Gelashvili 351, 353-354 Gelfand 151-152, 154-155 Godena 176, 179 Goldbeck 164 Grigoryan 66 Grischuk 206, 246-247 Gunina 259 H Hebert 332 Hillarp Persson 282 Holtel 159 Houska 200 Hou Yifan 256, 258 I Ivanchuk 60, 119 K Karpov 323 Kasparov 241, 252, 308 Klein 74-75, 77 Kotronias 161, 332-334, 354, 356-357 Kramnik 11, 44, 53, 77

724

Index of players

L Lacasta Palacio 335 Larsen 20, 216, 351 Lautier 345-348 Lokander 55 M Magalashvili 258 Mamedov 210-213 Maze 158 Meier 248, 262 Miles 323 Moskalenko 283, 341 Mosquera 350 Motylev 218, 220 Moussard 348 N Najer 77, 79 Nguyen 348 Ni Hua 164 Ntirlis 241, 243

Riazantsev 213 Robatsch 299 Rozentalis 148-149, 151 Rublevsky 77, 79 S Schandor 202 Seirawan 57 Short 252 Sielecki 159 Solberg 291, 334 Spasov 179, 182 Spassky 115 Spielmann 35 Steinitz 29, 33 Sveshnikov 151 Szostko 179 T Tal 186 Ter Sahakyan 176 Testor 328-329 Tiviakov 105, 155, 167, 170-173, 175 Topalov 61, 110, 202-203

O Osmanodja 180

V P Paichadze 210-211 Panjwani 146-147 Parligras 291, 334, 336 Pelikian 60 Petrov 341 Plenca 158 Polgar 62-64, 66 Ponkratov 258 Pribyl 284 Pröhl 180

Vachier-Lagrave 192, 203, 213-214 Van Wessel 74 Vescovi 60 Vidit 51 Vocaturo 156 Vouldis 354

R

X 725

W Wang Hao 69, 71 Wei Yi 51 Williams 106

726

Explanation of symbols

Xu Yi 164

Explanation of symbols

Z Zaitsev 48

The chessboard with its coordinates:

White to move Black to move ♔

King

♕

Queen

♖

Rook

♗

Bishop

♘

Knight White stands slightly better Black stands slightly better White stands better Black stands better

727

+–

White has a decisive advantage

–+

Black has a decisive advantage

=

balanced position

!

good move

!!

excellent move

?

bad move

??

blunder

!?

interesting move

?!

dubious move 728

Bibliography

N

novelty

Bibliography Compiling a complete repertoire for White from move one is a tough task. You de nitely should compare your ideas and assessments with the books that have already been published before. However with such a wide range of Black options it is basically impossible to check all books that are on the market. I had some books available in my own library and of course I used them to improve this repertoire. However I did not call my friend who owns a chess mail order business to order every single book ever published for Black against 1.e4, trying to nd aws with my lines. The chosen lines are fundamentally sound and chosen to provide good practical weapons – it is called ‘Keep It Simple’ for a reason. Here is the list of resources used to write the book. In fact I looked at some more books in my own library but in some cases the lines that I suggest were not even around when the books were written. Examples are some ideas in the French Exchange or Kramnik’s 10.h3 in the Scotch Four Knights. In older books these lines are simply missing. Book resources Playing 1.e4 – Caro-Kann, 1…e5 and minor lines, John Shaw, Quality Chess, 2016

729

730

Author’s biography

e3 Poison, Axel Smith, Quality Chess, 2017 The Hyper Accelerated Dragon, Raja Panjwani, Thinkers Publishing, 2017 The Modern Philidor Defence, Vladimir Barsky, Chess Stars ltd., 2010 The Modern Tiger, Tiger Hillarp Persson, Quality Chess, 2014 The Perfect Pirc-Modern, Viktor Moskalenko, New In Chess, 2013 Opening Repertoire: the Caro-Kann, Jovanka Houska, Everyman 2015 The Caro-Kann, Lars Schandor , Quality Chess, 2010 Playing the French, Jacob Aagaard & Nikolaos Ntirlis, Quality Chess, 2013 Playing 1.e4 e5 – A Classical Repertoire, Nikolaos Ntirlis, Quality Chess, 2015 opening repertoire: the Open Games with Black, Martin Lokander, Everyman Chess 2013 New In Chess Yearbooks Electronic resources Chessbase 14 Mega Database 2018 with updates Stock sh 8 and 9, Stock sh based cloud engines John Bartholomew’s Scandinavian repertoire on Chessable.com

731

Author’s biography

IM Christof Sielecki is a professional chess teacher and content creator from Germany. As a player he has competed in the top leagues of Germany, the Netherlands and Belgium. The best result was winning the Dutch team championship in 2012 with the team of Voerendaal. In Germany he has led his home town club of Dinslaken on the rst board, promoting from the 7th divison to the 2nd division. The best results in tournaments were winning the Latschach Open 2013 and sharing rst place in San Sebastian 2018, ahead of many GMs. He is a very active content creator. This includes a black repertoire book on the Nimzo- and Bogo-Indian in 2015 and many online publications for various platforms like Chessable.com, Chess24.com and The Internet Chess Club. He is also providing live chess commentary online and on site, most recently at the Berlin Candidates tournament. 732

Author’s biography

Table of Contents Copyright Page

3

Contents

4

Preface

7

Part I: Open Games: 1.e4 e5

11

Chapter 17: 2.♘c3: sidelines

376

Chapter 18: Two Knights: 2.♘c3 d5 3.♘f3 dxe4

383

Chapter 19: Two Knights with 3… ♗g4

394

Chapter 20: Two Knights with 3… ♘f6

411

Sample games

423

Introduction

11

Chapter 1: 2.♘f3: rare lines

17

Chapter 2: Philidor’s Defence: 2.♘f3 d6

25

Chapter 3: Petro Defence: 2.♘f3 ♘f6

Chapter 21: Exchange Variation: 456 2.d4 d5 3.exd5 exd5 4.♘f3, rare lines

41

465

Chapter 4: Three Knights Game: 2.♘f3 ♘c6 3.♘c3

Chapter 22: Exchange Variation with 4…♗d6

49

484

Chapter 5: Scotch Four Knights: 2.♘f3 ♘c6 3.♘c3 ♘f6 4.d4

Chapter 23: Exchange Variation with 4…♘f6

59

502

Chapter 6: 4…exd4 5.♘xd4

76

Chapter 24: Exchange Variation with 4…♗g4/4…♘c6

Chapter 7: 4…exd4 5.♘xd4 ♗b4

85

Sample games

515

Chapter 8: 5…♗b4 6.♘xc6 bxc6 7.♗d3 d5 8.exd5 cxd5 9.0-0 0-0 10.h3

99

Sample games

113

Part II: Sicilian Defence: 1.e4 c5 157 Introduction

157

Chapter 9: Rossolimo Variation: 2.♘f3 ♘c6 3.♗b5, sidelines

165

Chapter 10: Rossolimo with 3…g6

179

Chapter 11: Rossolimo with 3…e6

194

Chapter 12: Rossolimo with 3…d6

206

Chapter 13: Moscow Variation: 2.♘f3 d6 3.♗b5+ ♘d7

214

Chapter 14: Moscow Variation with 222 3…♗d7 Chapter 15: 2.♘f3 e6 3.c3

244

Chapter 16: 2.♘f3: other lines

274

Sample games

295

Part III: Caro-Kann Defence: 1.e4 c6 Introduction

372 372

733

Part IV: French Defence: 1.e4 e6 449 Introduction

449

Part V: 1.e4: other moves

536

Introduction

536

Chapter 25: Pirc Defence: 1…d6 2.d4 ♘f6 3.♘c3 g6

542

Chapter 26: Modern Defence: 1…g6

558

Chapter 27: Philidor and rare lines after 1…d6

571

Chapter 28: Scandinavian: 1…d5 2.exd5, rare lines

592

Chapter 29: Scandinavian: 2…♕xd5 3.♘c3 ♕d8

602

Chapter 30: Scandinavian: 2…♕xd5 3.♘c3 ♕a5

611

Chapter 31: Alekhine’s Defence: 1… 627 ♘f6 Chapter 32: 1.e4: rare lines

643

Sample games

659

Index of players

723

Explanation of symbols

728

Bibliography

730

Author’s biography

732

734