Sabotage the Griinfeld

Larry Kaufman

Sabotage the Griinfeld A Cutting-Edge Repertoire for White Based on 3 .f3

New In Chess 2014

20 1 4 New In Chess Published by New In Chess, Alkmaar, The Netherlands www.newinchess.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.

All photos: New In Chess Archives.

Cover design: Volken Beck Supervisor: Peter Boel Proofreading: Rene Olthof Production: Anton Schermer

Have you found any errors in this book? Please send your remarks to editors@newinchess.com. 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: 978-90-569 1 -440-0

Contents

Explanation of Symbols . 6

Introduction. . . . . 7

The Role of Komodo 1 1

Chapter 1 History of the f3 Anti-Grunf eld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3

Chapter 2 Third Move Offshoots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Chapter 3 The N eo-Grunf eld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. l 1

Chapter 4 The Samisch with ... cl -c5 .................................. 11 l

Chapter S The Samisch without ... cl -c5 ................................ 141

Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 3

Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 9

Index of Variations . 1 7 7

Index of Names. 1 8 3

Game List. . . 1 8 5

Bibliography . 1 8 7

5

Explanation of Symbols

;!; White stands slightly better The chess board :j: Black stands slightly better

with its coordinates: White stands better + Black stands better

8 +- White has a decisive advantage 7 -+ Black has a decisive advantage 6

= balanced position 5 00 unclear position 4 compensation for the material 3 > strong (sufficient) 2 < weak (insufficient)

better is a b c d e f g h :::; weaker is

good move !! excellent move

bad move

D White to move ?? blunder

Black to move !? interesting move King ?! dubious move Queen D only move II Rook /::,. with the idea Bishop - attack ttJ Knight

t initiative tt lead in development counterplay # mate

corr. correspondence

6

Introduction

This book is about the position arising after the opening moves l .d4 tt::lf6 2 .c4 g6 3 .f3 . It is sometimes called 'Alekhine's Anti-Grunfeld' due to the critical role played in its early days by then World Champion Alexander Alekhine, and to the motivation of White's 3 .f3 move to avoid a proper Grunfeld Indian Defense, which would arise after the natural 3 . tt::lc3 dS . Although a full White repertoire is part of this book, it also includes coverage of major non-repertoire lines so that a Black repertoire can also be derived from it. Explanations are aimed at club level players, but the actual analysis is intended to be suitable even for grandmasters.

The idea of the weird-looking 3 .f3 (weird because the g I -knight usually goes there) is to provoke 3 . . . dS by threatening to play 4.e4, at a time when White's knight is not yet on c3 . Then the Exchange Variation of the Grunfeld is more effective since Black cannot swap off his attacked knight on dS . The counter-argument is that White has paid a price for this, in the sense that f2-f3 is not as useful a move in general as tt::lf3 would have been. However it is still quite useful, as it goes well with queenside castling and a kingside attack similar to the Yugoslav Attack against the Dragon Sicilian or the Samisch King's Indian, into which play often transposes.

The move 3 .f3 does have some other drawbacks. Instead of playing Grunfeld-style (3 . . . dS) , Black can target the dark squares weakened by f2-f3 with moves like 3 . . . e S , 3 . . . cS , and 3 . . . tt::lc6 , the move chosen for the Black repertoire in The Kaufman Repertoire For Black and White (KRBW) , my last book. Black can also just choose the King's Indian, content with the fact that White is virtually forced to choose the Samisch Variation against it, which may not be his best option.

This book might seem an odd choice for an author who just enthusiastically recommended the Grunfeld for Black in KRBW But I am always keen to try to prove White's advantage in chess, and when I realized the strength of the move 3 .f3 I felt that it deserved a book of its own. I expected this to be the only book in the current century devoted to this position, but a similar book by grandmaster Svetushkin came out first. While we agree more often than not, I point out the many places where I could not agree with him after doing further analysis. Although we are both grandmasters, Svetushkin is considerably higher rated than I am, but on the other hand I probably have much better computer hardware and software than he is likely to have used. These days this is very important.

7

S a b o t a g e t h e Griinfeld

So what is my verdict on 3 .f3 from a theoretical standpoint? In my view, it is extremely difficult to demonstrate any meaningful advantage against the Griinfeld, but the 3 .f3 d5 neo-Griinfeld does seem to give White his normal opening edge. I think this is also true of the alternative third moves other than 3 . . . g7 (or 3 . . . d6 first) , the King's Indian. Against the King's Indian, I don't think that the Samisch is White's best option, but I do think it suffices for a normal opening edge. So White is giving up something against the King's Indian to get something against the Griinfeld. Since the Griinfeld has a much better theoretical reputation than the King's Indian, White has more latitude as to what to play in the second case. To summarize, 3 .f3 seems to give White his normal slight plus whatever Black plays, whereas the usual 3 .tlJc3 fails to do so against 3 . . . d5 ! . If all this is true, then 3 .f3 may simply be ' the best move' ! So the move 3 .f3 is suitable to use at every opportunity, but is especially appropriate against opponents known to prefer the Griinfeld over the King's Indian.

The 3 .f3 Anti-Griinfeld is by no means new, though it is newly popular. It was introduced in 1 929 in games by Flohr and Nimzowitsch, but was quickly taken up by World Champion Alexander Alekhine in his title defense against Bogoljubow, and he played it many times over several years thereafter. It was named 'Alekhine's Anti-Gruenfeld Attack' in one book with the same name written by Alan 1. Watson in 1 996 . It was also played by many other famous players of that era, including Fine, Yudovich, and Bondarevsky. However the majority of these early games transposed to the Samisch, and in the majority of games featuring 3 . . . d5 White refrained from queenside castling, which is now considered essential in this variation. Still , some of the early games do feature modern lines, and we trace their development in the historical section of this book.

This book attempts to cover all the reasonably important variations arising after 3 .f3 , but with the Samisch King's Indian being given restricted coverage to avoid duplication with Schandorff s excellent coverage of it in his recent White repertoire book. However, he only covers lines with 6 .e3 so I fully cover lines without it, as well as some improvements I have found on his lines. This is not strictly a repertoire book, but I do make clear which lines I recommend for a repertoire for White, as well as which ones are playable for Black (i.e. limit White to no more than a 'par' opening plus) .

Virtually everything in this book has been checked by the two strongest engines at the time of writing, Houdini 3 and (at a later stage) 4, and Komodo, for at least 1 5 minutes per position, usually more. Both of these engines are much stronger than the corresponding versions that were used for my last book, so the quality of analysis is that much higher. This analysis is done using the ' IDeA' feature of 'Aquarium' so that hundreds of positions can be scheduled for analysis overnight (one core per position) . With an eight core and a twelve core computer devoted to this task, you can see how it was possible to analyze deeply the many thousands of positions in this book in a reasonable time. Most writers just use the engines to check as they

8

In tr o d u c t i o n

write, so many positions will have less than a minute's scrutiny. My method insures that the quality of the analysis will be top-notch, at least to the extent that the computers can approach the truth. I use my own judgment as a grandmaster to decide which engine to believe when they disagree, as well as to identify the occasional instances where they both get things wrong, primarily in endgames and in severely blocked positions. Because I am a co-author of Komada (together with the late Don Dailey) , I am usually able to explain in words why it evaluates a position as it does. I tend to favor Komodo's analysis over Houdini both because I better understand where the scores are coming from, and because I believe that Komodo's evaluations are on average a bit more realistic in human terms. At the time of writing, Houdini is the stronger engine at blitz levels, while tests at levels averaging a couple minutes per move generally favor Komada. Based on this trend, I believe that at levels like 1 5 minutes per move as used for this book, Komada is likely stronger than Houdini, but no one tests at such long levels so I can't be certain of this. But regardless of which engine is objectively stronger, Komada seems to 'like' the white side of most of the recommended lines in this book more than Houdini , correctly so in my opinion as the lines in question do score well for White in human practice.

As with my previous books, my choice of recommended lines is primarily objective, and hence suitable even for the elite grandmasters, but since the text is aimed at ordinary club players, I hope that this book will appeal to a wide range of players , including even grandmasters.

Another feature of the previous book which I retained for this one is that all references to material values are based on my own scale, which is pawn = 1, knight = 3 . 5 , single bishop = 3 . 5 , two bishops together = 7 . 5 (i.e. half a pawn bonus for the pair) , rook = 5 . 2 5 , and queen = 1 0 . This is far more accurate and reliable than the traditional 1 -3-3-5-9 count.

Most chess opening books focus on master I grandmaster games, with analysis by engines of the moves. In this book, I work partly from databases which include large numbers of games played by computers. Consequently I focus more on the analysis and less on the actual games, since the analysis done at 1 5 minutes per move will be of much higher quality than almost all of the actual games, which are played at much faster time limits than this. I make an effort to cite relevant games in the variations, but I 'm not fanatical about it, I don't always try to track down which computer-recommended move is the actual novelty in a sideline, especially since many novelties are played in computer games anyway. Frankly, since most opening analysis used in games comes from computers , I don't think it's terribly interesting to know which grandmaster happened to get the first chance to use a computerinspired novelty. In at least one case, an elite game featured a novelty followed for many moves by a variation I had already written up for this book!

Working on this book has been somewhat of a novel experience for me. In my previous books, which covered the full range of openings, I had to cut off analysis at some reasonable depth in order to be able to write such books. This time, I was

9

S a b o t a g e t h e G riinfe ]d

able to analyze all the way into the endgame in many lines, since the range of coverage was so much narrower. I hope that I have come fairly close to the truth in at least most lines. The usual result of such deep analysis of a good White opening is that White should reach an endgame where he is the only one with winning chances , although Black should be able to hold. That is indeed the result shown here for many of the best Black defenses.

I also want to point out that I have been playing this 3 .f3 line in tournaments since late 20 1 2 myself (mostly as White, once as Black) , and so I have some firsthand experience with the lines. So far I believe I have gotten a clear (maybe winning) advantage out of the opening in every White game, though I make too many mistakes at the board at age 65 so my actual results aren't that wonderful.

For those who are not familiar with me, my tournament successes now span more than half a century, from second place in the Maryland Junior Championship in 1 96 1 to the World Open Senior Championship in 2 0 1 3 . I earned the grandmaster title by winning the World Senior Championship in 2008 . I won the American Open Championship in 1 966 , became an international master in 1 980 , played in four U.S. Championships and two Student Olympiads, and have been Maryland champion eight times, including as recently as 20 1 2 . My son Raymond is an international master. This is my third chess book. I have been involved with computer chess (off and on) since 1 96 7 , when I worked on 'MacHack' , the first computer to compete in human tournaments. More recently I worked on Rybka and now Komodo.

I would like to thank IM Eric Kislik for his help in analyzing many of the lines in this book, the late Don Dailey, my Komodo partner, for this engine without which the book just wouldn't be nearly as good, and of course all the players who played the games featured herein. The book includes relevant grandmaster games right up to February 2 0 1 4.

1 0

L arry K aufman P otomac MD, USA

F ebruary 20 14

The Role of Komodo

Because of my great faith in the Komada chess engine, I have decided to introduce a new feature in this book. Traditionally, most variations end with a symbolic assessment like ;!;. While retaining these symbols, in key positions in the analytical section of the book I also give Komodo's evaluation (in pawns) after a deep think (usually 1 S minutes or more) . That way I am free to express my own opinion either in symbols or words while letting the reader see the unbiased opinion of a very deep computer analysis by what I believe to be the best analytical engine available. That there is some objective basis for this opinion was demonstrated just before my deadline on this book when Komada won the 'TCEC' internet tournament, which is widely considered to be the real (if unofficial) world software championship, ahead of all major rivals including Houdini and Stockfish. Most regrettably, my partner in the Komada project died just 9 days before Komodo's TCEC triumph. It's hard to say what Komodo's rating would be in human terms, but there is no doubt that it starts with a ' 3 ' and that it would be hundreds of points above even World Champion Magnus Carlsen.

Normally computer evaluations are translated to symbols with the t evaluation covering the range 0 . 2 S to 0 . 7 S pawns. I think this is a bit too high, my guideline for that symbol is about 0 . 20 to 0 . 60 . But the symbolic evaluation shown often disagrees with the one implied by the last sentence, because it reflects my opinion. I try to consider how easy it is to play the position as well as its objective merit. So for example even with a score like +0 .80 or more I may give just;!; if the position is tricky and double-edged. On the other side I may give a +0 .07 position a t if it's not excessively drawish and if White is clearly (if only slightly) for choice. The computer's evaluation is probably the better guide for a correspondence player, but I would like to think that my own opinion has some added value for practical over-the-board players at least. Occasionally I totally disagree with Komada, but these instances are quite rare.

One more note about Komodo's evaluations. After deeply analyzing all serious opening lines in chess with Komada, my conclusion is that White's initial advantage is about 0 . 1 S pawns. This is not enough to win, but should be enough to retain some advantage into the endgame, so Black will at least have to be careful to secure the draw. Consequently positions evaluated as well above +0 . 1 S are deemed (by Komada at least) to be desirable for White, and those well below +0 . 1 S (which I call 'par') are acceptable for Black. With 3 .f3 White nearly always achieves 'par ' , the only exceptions being rare lines that just haven't been analyzed very deeply. All the major third move replies to 3 .f3 usually lead to this par score with best play, but in many cases Black must tread a very narrow path to reach this score.

I also use the known feature of the 'tip' in this book, indicated with the sign:

11

Chapter 1

History of the f3 Anti-Grunfeld This section is a chronological review of games played in the last century that introduced new opening ideas in this variation. By playing through these games, you will learn how to meet all sorts of unusual moves Black may try, as well as some unorthodox white tries, although often I have omitted white ideas that have been found wanting. You will also acquire the foundation to better understand and remember the modern ideas found in the theoretical section, as well as a better feel for how to play typical positions. All the analysis here has been checked by Komodo, though not always to the depth used in the theoretical section. Once we get past the point where a game has any theoretical value, I don't insist on really deep checking, since any mistakes won't affect your opening choices.

The first two games featuring l .d4 tllf6 2 .c4 g6 3 .f3 were played in 1 929 by two of the best players of the time on the white side, namely Salo Flohr and Aaron Nimzowitsch. I start off this chapter with the Nimzowitsch victory over Saviely Tartakower, also a top player back then. As happens more often than not, Black sim -ply allowed transposition to the Samisch King's Indian, but the game does illustrate a theme I emphasize in the book, namely that with a pawn on f3 it is usually good to play tllh3 if the knight cannot be profitably taken. See Game 1 . 1 . Studying these old games might be a waste of your time if you are a grandmaster, but for amateur players there is often more to learn from the old games than from the latest battle between titans, because the top players back then often made strategic mistakes similar to what many of today's amateurs will do. Instead of a wild battle with no real theme you are more likely to see how the stronger player exploits the mistakes of the weaker one.

One point really struck me when annotating these games of the 1 93 0s. It seemed to me that the superstar players played at a much lower level than today's stars, perhaps at the level of an ordinary grandmaster today, while most of the players of that time who are not famous today were likely not even of what we would now call master strength. This is partly due to unfamiliarity with what are now considered standard plans and ideas, but also to missing more tactics . Probably this is simply a reflection of the fact that today's stars play thousands and thousands of games on the internet to hone their tactical skills, while the top players back then rarely had much chance to play other players of their own level except in the occasional tournament. Perhaps this opinion will be somewhat controversial, but it agrees with my own experience as someone who played in many tournaments from the early 1 960s to today. Today's players are just much stronger, for many reasons.

1 3

S a b o t a g e t h e Gr ii n feld

Anyway, the next three games in my survey are all between World Champion Alexander Alekhine as White and his challenger Efim Bogoljubow. The first two are from their title match in 1 92 9 , the third is from a 1 9 3 1 tournament which Alekhine won by a huge margin. The first (Game 1 . 2) transposed to an unusual Samisch line where Black chose . . . c7 -c6 and . . . d7 -dS rather than . . . d7 -d6. This is a known idea in the King's Indian, a favorite of the late grandmaster Alex Wojtkiewicz, but it usually aims to meet eS by . . . tl'ie4, which is obviously impossible in the Samisch. The other two games featured the neo-Grunfeld move 3 . . . dS . For the most part I skipped over games that transposed to the Samisch in this section, with a couple exceptions. In the first of these two games (Game 1 . 3 ) we see the first use of the defense in which Black develops his queen's knight before castling, which is not recommended though it was tried in one recent elite game. Alekhine won a very nice game, although he missed multiple opportunities to win the game along the way. In the last of these games against Boga (Game 1 .4) , Alekhine played 9 .dS ? in the position where everyone castles queenside nowadays, and followed this error by four more in the next few moves ( ! ) , arriving at a dead lost position which the engines rate at about the equivalent of a rook up for Boga. Despite this, Alekhine miraculously drew the game ! It's hard to imagine this set of events happening in an elite game today, although I 'm sure someone will be able to point out a similar case in modern play. This game should be a lesson that this system is not without risk for White; a couple mistakes and White can be dead lost.

In the next game we see the first use of the system with 8 .f4 , which is still topical as it was played in a very recent elite game (see theoretical section) . Black responded poorly and soon fell victim to a standard h-file attack. These early games are very good at showing how to punish routine or inaccurate play by Black. This game (Game 1 . 5) was rather one-sided.

In Game 1 . 6 we have a 1 9 3 3 battle between two of the titans of American chess at the time, Fine and Dake, both of whom I met in the 1 9 7 Os. It featured the first use of 8 . . . eS (as in Anand-Gelfand from their recent title match) , but White played the natural but weak 1 O . a4? instead of Anand's 1 O .h4. Black missed the proper way to punish this move, giving White the edge, with the advantage tilting back and forth after multiple errors on both sides, with Fine finally winning brilliantly.

Next we have the first use of the move 3 . . . e6, which Maxime Vachier-Lagrave has brought into the limelight this year. I was truly shocked to see that this move dated back to 1 934! Naturally both sides made dubious choices in the next few moves, with White emerging on top and winning a nice miniature. See Game 1 . 7 .

The next battle, again to my amazement, featured Black switching to the Benko Gambit with 3 . . . cS 4.dS bS , long before Benko ever played his Gambit! ! However it's not so good here because in the real Benko Gambit Black would meet an early f3 by . . . e7-e6 rather than by . . . g 7-g6. In the game White declined the pawn and Black should have equalized, but he blundered twice on moves 20 and 2 1 . Black (Opocensky) was famous for his opening innovations , but perhaps his practical playing strength was not on a par with his originality. See Game 1 . 8 .

1 4

Chapter 1: H i s t o ry of the f3 Ant i - G riinfeld

Next we see the first use of 8 .bS against the premature knight development of Game 1 . 3 . On moves 1 5 and 1 6 White missed the chance for advantage with the thematic d4-d5 ! , something which I doubt a modern master would miss. The way he played allowed Black easy equality and a quick draw. Black was one of the titans of the day, Grigory Levenfish. See Game 1 . 9 .

In our next game Black essays a very rare gambit with 4 . . . c 6 , which shouldn't give full compensation for the pawn but does offer some practical chances. White declined the gambit and still obtained some advantage. Later Black equalized but White again built up a winning advantage, only to throw the game away with missed wins on moves 3 3 and 3 4. See Game 1 . 1 0 .

Next we come to another example of the line from Game 1 . 9 , with the new move 9 . . . eS . This could have led to a fascinating piece sacrifice (analyzed in theory section) but instead Black just conceded a modest plus to White, which White went on to exploit for a win. The game featured two heavyweights of 1 948 , Bondarevsky vs. Aronin. Curious that Lev Aronin was a top player then, while Levon Aronian is a top player now! See Game 1 . 1 1 .

Now we come to two games with the Enevoldsen brothers behind the white pieces, introducing a long variation that remained popular into the current century. Black plays 9 . . . eS 1 O.dS tlJd4, getting a nice outpost, which White proceeds to undermine. In Game 1 . 1 2 Jens missed several chances for a modest plus, then one chance for a repetition draw, and quickly went down to defeat. In Game 1 . 1 3 Harald had the more dangerous attack until a very understandable error on move 2 3 turned things around dramatically and Black went on to win, although missing a nice mating combination on move 26 . This subvariation should favor White, but it is very tricky and the price of a mistake by either player is high.

Our next game takes place after my own playing career had begun, in 1 964, between two of the best American players of that time, Larry Evans and Donald Byrne. Black played the original but dubious 9 . . . aS ? ! , and after the unnecessary 1 O . . . e6? he was already losing. Later Black missed a chance to complicate, but then went down to a one-sided defeat. The game is a good illustration of the power of the h-file attack if Black fails to find sufficient counterplay. See Game 1 . 1 4.

Next comes our first example of the Benko Gambit (with f2-f3 and . . . g7-g6) Accepted. I think that White did not adopt the best plan in this variation and Black missed some chances to equalize, but eventually he blundered and lost. See Game 1 . 1 5 for both the actual game and the recommended line for White.

Game 1 . 1 6 features a return to the 8 .f4 attack (which seems to have been the main line in the last century) , this time with the now usual response 8 . . . t2Jc6 , followed by 9 . . . tlJaS when the knight is attacked. The game went back and forth between equality and white advantage until Black made a fatal error on move 28 . The game and notes show the importance of exchanging off Black's dark-squared bishop when possible.

In Game 1 . 1 7 Black tries to improve on the previous game by retreating his knight back home when it is attacked, but I believe this was actually a worse move. However White missed the only way to punish the move ( 1 O . a4 ! ) and Black equal-

1 5

S a b o t a g e the G riinfeld

ized easily. The queens were traded into a virtually equal endgame, but somehow Black was outplayed and went down to defeat.

Next we see the first example of meeting the then-popular 7 . . . CiJc6 by the 'safe' 8 .dS rather than the complex 8 .bS . White wins a pawn in the game and goes on to victory after multiple errors on both sides, but the notes show that Black could have regained the pawn with only a moderately inferior game. This 7 . . . CiJc6 line is a bit dubious but not clearly refuted. See Game 1 . 1 8 .

In Game 1 . 1 9 White tries to improve over Game 1 . 1 6 on move 1 1 , but I think his new move was worse than the old one, and Black equalized. However Black almost immediately went astray on move 1 3 and found himself in a losing position, which White went on to win fairly easily.

In our next game Black simply plays to prevent d4-d5 by 7 . . . e6 . This allowed White to obtain a dominant center and space advantage, although at the price of the bishop pair. But White missed the way to consolidate his plus and Black equalized. Later White was fortunate that he could still draw the game. See Game l . 2 0 .

In Game l . 2 1 we have a return to the Benko Gambit Accepted, this time with White adopting a different plan than the one I recommended in Game 1 . 1 S . His choice also seems sufficient for at least a small plus, and although both players missed a nice equalizing trick on move 1 4, White went on to win a fairly nice and instructive game.

Next we have a return to the 8 .f4 fS variation of Game 1 . 5 , but this time with a much better reply to the check on move 9 . It seems that Black could have equalized by keeping queens on the board on move 1 S . I think Black's queen is more active than White's in this line and that White would have trouble playing actively with the black queen present. As the game went he was able to keep improving his position and won a nice and instructive endgame. See Game l. 2 2 .

Now we come to the first game featuring the initial 1 4 moves of the recent Anand-Gelfand game. However White went astray with 1 S .g4?, after which Black missed first a chance for advantage and then an exchange sac for equality. Instead Black chose a line giving him two pawns and the bishop pair for a knight, which was not enough, and he lost. See Game 1 . 2 3 .

Our final game in this section features the first use of 9 . . .fS , a still-topical variation. White played the sharp 1 O .h4, which I analyze to a draw in the theoretical section. In this game though, Black chose to suffer for a pawn, a poor decision in my opinion, and he lost rather badly. See Game 1 . 24.

This brings us to the present century. We have seen examples of all major variations (excluding King's Indian transpositions) except for the very recent 9 . . . 1/Wd6 and for

3 . . CiJc6 , for which see the theoretical section.

1 6

Chapter I: H i s t ory of t h e f3 Ant i - G riinfeld

(ES 7) Game 1 . 1 Nimzowitsch, Aaron Tartakower, Saviely Karlsbad 1 929 (2 1 )

1 .d4 tllf6 2.c4 g6 3.f3 This was one of the first two games played with this move according to the database.

3 ... g7 4.e4 d6 5.tllc3 0-0 6.e3 tllbd7

7.tllh3! With a pawn on f3 , f2 is usually the best home for the king's knight if it can reach that square. I think 6 . . . tllbd7 is strategically refuted by this reply.

7 ... e5 S.d5 as 9.tllf2 b6 1 0.'id2;!; tllc5 1 1 .gS d7 1 2.g4 cs 1 3.h4

1 3 ... @hS? This is ridiculous, moving toward the attack. Tartakower was one of the great players of the era; was the standard of

Aaron N imzowitsch

play really that low then? Apparently Tartakower judged that winning the bishop pair was worth having his kingside pawn structure ruined and his king attacked, but if so that is very bad judgment. Black should try l 3 . . . c6, though White's attack should prevail.

14.hS gxh5 15 .xf6 xf6 1 6.l::rxhS g7 1 7.tllh1

I 7 .O-O-O es 1 s .d3 f6 1 9 .l:!h2 @gs 20 . .ld.dh l g6 2 1 .c2+-; White has a strategically won game.

1 7 .. .f6 1 S.h2 h6 1 9.tllg3 @h7 20.e2 I:igS 21 .@f2

2 1 . 0-0-0 looks even better. 21 ... l::lhS 22.l::rh4 ?! es 23 . .l:rg1 fS 24.@g2?! tllb7 25.tllhS 96 26.f4 tlldS?

17

S a b o ta g e t h e G r ii n feld

27.f3? 2 7 .gs fxg5 28 .fxg S +- .

27 ... Ci:Jf7? 2a.Ci:Je2 28 .gS ! fxg5 2 9 .fS +-.

2a ... e7 29.\t!h1 @ga 30.Ci:Jeg3 @fa 31 .Ci:Jfs .:g:ga 32.d2 :t:Ica 33 . .l::!.h2 @ea 34.b3 @da 35.a3 :t:Iaa 36.c1

White is already clearly better here, but after the next move the game is decided.

36 ... fa?? 37.Ci:Jh4 h7 3a.Ci:Jxf6+- ha 39.Ci:Jxga xga 40.gS exf4 41 .gxh6 h7 42.xf4 xh6 43.f6+ @ca 44.Ci:JfS xf5 45.exfS @b7 46.g6 l::!.ha 47.xh7 .l:!.xh7 4a . .l:i.g6 @ca 49.f6 .l::!.ha so.g4+ @da 51 .e6 @ea 52.xf7 + @xf7 53.hxh6 1-0

If someone told me this was a recent game, I would guess the players to be rated around 2 000 . But they were among the top five at the time!

(E70) Game 1.2 Alekhine, Alexander Bogoljubow, Efim Germany/Netherlands Wch m 1 929 ( 1 5)

1 .d4 Ci:Jf6 2.c4 g6 3.f3 g7 4.e4 0-0 5.Ci:Jc3 c6 6.e3 d5

1 8

7.e5 7 .cxd5 ! cxd5 8 .e5 is a more accurate move order to reach the game without allowing the gambit mentioned on move 8 .

7 ... Ci:Jfd7 7 . . . Ci:Je8 8 .cxd5 cxd5 9 .h4 is slightly better for White.

a.cxd5 cxd5 After 8 . . . Ci:Jb6 ! 9 .dxc6 Ci:Jxc6;\; Black has some but not full compensation for the pawn.

9.Ci:Jxd5 as+ 1 O.Ci:Jc3 Ci:Jc6

1 1 .c4?! l l.a3 ! prepares b4 and prevents . . . Ci:Jb6 , leaving Black little compensation for the pawn.

1 1 ... Ci:Jb6 1 2.b3 e6 1 3.Ci:Je2?! 1 3 .xe6! fxe6 1 4.Ci:Je2 still gives White an edge.

1 3 ... Ci:Jc4 14.xc4 xc4 1 5.0-0 .l::!.fda 1 6.f4

Chapter 1 : H i s t o ry o f the f3 An t i - G riin f e ld

16 ... .ld.acS 1 6 . . . f6 , and with the bishop pair, a bad enemy bishop, and pressure on the backward d-pawn, Black has full compensation for his pawn.

1 7.a3 1 7 .b3 .te6 1 8 .a3;!; .

17 ... 'iY'b6 1 8.'fid2 tt:Ja5 19 . .ld.ae1 'iY'c6 20 . .if2 b6 21 .'iY'e3 f5?!

2 1 . . . .ie6 2 2 . .ld.d l tlJc4 23 .'ii'c l 'ii'b7=. 22 . .ih4 .ld.d7 23.b4

23 . .i::!d 1 .if8 24.b4;!;. 23 ... .txe2

23 . . . tt:Jb3 24 . .l::td l e6=. 24.tt:Jxe2 tt:Jc4 25.'ii'b3 b5 26 . .l:!.d1 e6 27 . .tf2

2 7 .'ii'f3 t. 27 ... .ifS 28 . .l::!.d3 as

29.bxaS With 2 9 . .ld.fd l ! axb4 3 0 .dS l:i'.xdS 3 1 ..ld.xdS exdS 3 2 .axb4 .l::!.a8 3 3 .tlJd4;!; White returns his extra pawn, retaining the better pawn structure.

29 ... tt:JxaS= 30.'ii'b1 ? 30 .'fia2=.

30 . . . tt:Jc4 31 Jk1 .l::!.a8 32.dS .l:!.xd5 33.ld.xdS 'fixd5 34.tt:Jc3 'iY'd2 35.tt:JxbS

35 ... tt:Jxa3 A) 35 . . . 'ii'xf4? 36 .g3 iVg4 3 7 .'iYb3+-; B) 35 . . . .ld.c8 3 6 .ld.c3 Wixf4 3 7 .'ii'c l

'ii'xc l + 3 8..l:!.xc1 ld.b8 would slightly favor Black.

36.tt:Jxa3 .l::!.xa3 37.g3 .l::!.a2 38 . .ld.f1 a'.b2 39.'ii'c1 'fib4 40.'fic6 'ii'b3 41 .'ii'aS 'ii'c4 42.'ii'dS 'ii'c6 43 . .l:!.d 1 'ii'f3 44 . .l::!.f1 'ii' c6 1/2-1/2

(D70) Game 1 .3 Alekhine, Alexander Bogoljubow, Efim Germany/Netherlands Wch m 1 929 ( 1 7)

1 .d4 tt:Jf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tt:Jb6 6 . .te3 .ig7 7.tt:Jc3 tt:Jc6?!

1 9

S a b o t a g e the G r iinfeld

Castling is far more common here, and probably better, but the knight move was played in at least one recent elite game so it's not 1 00% clear.

8.d5 8 .bS ! ? is the greedier move, but it entails more risk due to the following piece sacrifice: 8 . . . 0-0 9 .tllge2 and now:

A) 9 . . . tllas 1 0 .b3 a6 1 1 .d3 tllc6 1 2 .c2 tllb4 1 3 . 0 -0 ( 1 3 .b l i) 1 3 . . . c6 1 4.l:i'.c l ( 1 4.b l i) 1 4 . . . tllxc2 1 5 . .l:txc2 aS (Borwell-Miillner, corr semifinal Wch- 1 9 1 992/94) 1 6 .a4i;

B) 9 . . . eS 1 0 .dS tllaS ! ? ( 1 0 . . . tlle7 Bondarevsky-Aronin. Moscow ch-URS 1 948 ; 1 l .tlla4i) 1 1 .xb6 axb6 l 2 .b4 c6 1 3 .d3 bS 1 4.bxaS b4 and Black gets decent if not full compensation for the sacrificed knight, see the theory section for details.

8 ... tlle5

9.d4 9 .f4 ! tllg4 1 0 .bs+ d7 1 I .d4 xd4 l 2 .xd4 0-0 (SpeelmanLputian, Hastings 1 986/87 ) 1 3 .xd7 xd7 l 4.h3 tllf6 1 5 .tllge2 c6 1 6 .a4i.

9 .. .f6? 9 . . . 0-0 1 0 .f4 and now:

A) 1 O . . . tlled7 1 1 .xg7 ..t>xg7 1 2 .d4+. Alekhine claimed a big edge here based on the attack with h2-h4, but . . . 1 2 . . . tllf6 1 3 .tllf3 e6 1 4.dxe6 xe6 1 5 .e2i.

20

Analysis diagram

White has no attack, just a more advanced majority and slightly better placed knights;

B) 1 O . . . g4? ! 1 2 .tllgxe2 tllg4 1 4 .d4+ tllf6

1 1 .e2 xe2 1 3 .xg 7 ..t>xg7

1 5 . 0-0 -0 cs (Laurentius-Alexander, Warsaw Olympiad 1 93 5 ) 1 6 .d3 ;

C ) 1 0 . . . tllg4 1 1 .xg7 'it>xg7 1 2 .d4+ tllf6 transposes.

1 0.f4 1 O .a4 ! aS l 1 .b3 0-0 1 2 .tllbS (by threatening tllxc7 White provokes . . . 'it'h8 , which makes h2-h4 even more effective) 1 2 . . . 'it>h8 1 3 .h4.

1 o ... tllf7 1 1 .a4 Now it's less effective.

1 1 ... e5! 12 .dxeG xe6 1 3.a5 tlld7 14.aG b6 1 5.b5 e7 1 6.tllge2 c5

1 6 . . . 0-0 1 7 . O-Oi.

1 7.f2

Chap t er 1: H i s t o ry of t h e f3 An t i - G riinfeld

1 7 .fS ! gxfS 1 8 .exfS xfS 1 9 .tlldS d6 20 .xf6 xf6 2 1 .0-0 es 22 .l:rxfS 0-0 2 3 .xd7 l:!.ad8 24.c6 Wh8 25 .b7+- . Alekhine missed this very difficult win of a piece for a pawn.

1 7 ... 0-0-0

In case of 1 7 . . . 0-0 1 8 .tlldS xdS ( 1 8 . . . d6 l 9 . a4+-) 1 9 .xdS l:!.ad8 20 .0-0i White has a solid bishop pair advantage plus more space.

1 8.a4 Alekhine missed the difficult winning line 1 8 .tlldS ! d6 l 9 .a4 xdS 20 .l::rd l c7 2 1 .l::rxdS tlld6 2 2 .tllc3 tllf8 2 3 . .l:!.d3 Wb8 24 .tlldS f7 25 .c6+- .

1 8 .. .f5 1 9.e5 g5! 20.c4 After 20 .g3 tllfxeS ! 2 1 .fxeS tllxeS Black gets enough for the piece.

20 ... tt:Jdxe5 20 . . . tllb8 2 1 .0-0+- .

21 .xe6+ xe6 22.fxe5 tt:Jxe5 23.0-0

23 ... c4? Better was 2 3 . . . tlld3 24 .'iYbs l:rd7 -and with two pawns, center control, and a super knight outpost, Black has almost enough for the sacrificed knight.

24.b4!! xb4 Or 24 . . . cxb4 2 5 .tllbS ! xe2 (25 . . . Wb8 2 6 .xb6+- ) 2 6 .tllxa7 + Wb8 2 7 .tllc 6++- Wc7 28 . .l:Iac l tllc4 29 .tllxd8 l:!.xd8 3 0 .xb6+ and White wins.

25.c2? 2 5 .xb4! cxb4 26 .tllbS @b8 2 7 .tlled4 f4 28 .l:rad l f6 29 .l:rfe l +- . Black's three scattered pawns are no match for the piece here, as White has numerous threats.

25 ... tlld3 26JHb1 ? 26 .l::rfd l i .

26 ... c4 27.l::ra4 e6 2 7 . . . f? 28 .tllbS.

28.tllb5

28 ... WbS? After 2 8 . . . tllxf2 2 9. Wxf2 Wb8 3 0 . .l:rd 1 c8 3 1 .tllec3 l:!.hf8 3 2 .Wg l White has a slight edge. He is up half a pawn in material, but otherwise the position is balanced.

29.tlled4 e4 2 9 . . . xd4 3 0 .xd4 l:!.xd4 3 1 ..l:!.xd4 e3+ 3 2 .Wfl +- .

30.tllc3 ea 31 .xd3 cxd4

2 1

S a b o t a g e t h e Grii nf e ld

32 . .2.xd4? 3 2 .i:Wf3 ! i:Wf7 3 3 . .l::rxb6+ axb6 34.a7+ "i:Wxa7 35 . .l::!.xa7 @xa7 36 .tllbS+ @a6 3 7 .tllc 7 + @a7 3 8 .a3 + @b7 39 .tlle6+-.

32 ... i:We6? 33.i:Wf3 i:Wf7 34.xb6!

Alexander Alekhine

22

Black resigned, a s after he takes the bishop 3 5 . "iW c6 is mate in 7 , as is also 3 5 . .l:f.xb6+.

(D70) Game 1 .4 Alekhine, Alexander Bogoljubow, Efim Bled 1 93 1 (22)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tllxd5 5.e4 tllb6 6.tllc3 g7 7.e3 0-0 8.'i!Wd2 tllc6

9.d5? Only castling offers a chance for advantage, while the old 9 . .l:i.d 1 ? ! may suffice for equality.

9 ... tlle5 1 0.g5? The lesser evil was I 0 . .l:i.d I .

1 o ... c6 1 1 . .l:rd 1 ? cxd5 1 2.exd5 f5 1 3.g4 d7 14.d6? f6 1 5 . .2.h6?

1 5 ... c6

Chapter I: Hi s t o ry of t h e f3 An t i - G ri infeld

1 5 . . . .ixh6 1 6 .xh6 exd6 1 7 .e2 c6 l 8 .f4 e7-+ ; with a pawn less and a crushing deficit in development and king safety, White could reasonably resign.

1 6 . .ixg7 @xg7 1 7.f4 exd6 1 8 . .ie2 e7

This is the same position as in the previous note after l 5 . . . .ixh6, except for the position of the black king, which is not important.

1 9.@f2 White's position is still resignable, and yet somehow he doesn't lose, although Black was among the world's best players then.

1 9 ... g5 20.d4 f5 21 .h4? fxg4 22.hxg5 xg5

The computers rate this as about a rook up!

23.tt::le4 f4? 2 3 . . . .ixe4 24.xe4 hS 2 5 .l:i.xd6 l:i.ac8-+.

24.tt:JxdG xd4+? 24 . . . gS 2 5 .tt::lh3 'i:!:Yf6 26 . .l::!.hg l hS 27 . f4 .!d.ad8-+ .

25.xd4 @g6 26 . .lah4 liad8? 2 6 . . . .ixf3 2 7. tt::lxf3 hS is still clearly better for Black.

27.f 4 Black still has the upper hand here, but with his next move he definitely lets his advantage slip.

27 ... tt:Jd5? 27 . . . tt::lc8+.

28.@g3 tt:Jxf4 29.a'.xf4 lixf4 30.@xf4 .l:f.xd6 31 .@xe5 l:i.d5+ 32.eG h5 33.xg4 hxg4 34.xg4+ @h5 35.l:i.g3 .ld.d2 36.tt::lf3 l::rxb2 37.tt::ld4= 112-1/2

White's vastly superior king position makes the draw easy despite the pawn deficit. What an amazing save !

(D7 0) Game 1 .S Rodi, Ludwig Helling, Karl Bad Pyrmont ch-GER 1 93 3 (5)

1 .d4 tt::lf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tt::lb6 6.tt::lc3 .ig7 7 . .ie3 0-0 8.f 4

8 .. .f5?1 As shown in the theoretical section, the unknown move 8 . . . cS ! is quite promising, while 8 . . . tt::lc6 is normal and sufficient for equality with precise follow-up.

23

S a b o t a g e t h e G r iinfeld

9.'YWb3+ @h8?! The immediate 9 . . . e6 is better as on h8 the king is a target for h4-h5 .

1 0.e5 e6

1 1 .tllf3 l l .h4 directly is probably more accurate.

1 1 ... tllc6 1 2.h4t a5 1 2 . . . hS looks like the lesser evil.

1 3.h5 gxh5 14Jhh5 l 4.a3 looks even better, to prevent . . . a5-a4 and . . . a4-a3 .

14 ... 'YWeS 1 5.h3 a4 1 6.'YWd1 a3 1 7.b3 tllb4

1 8.d3?! 1 8. f2. ! . Why make the bishop be a hostage to the knight?.

1 8 ... d7?! 1 8 . . . h6t.

1 9.g4 c6 20.gxf5 tllxd3+ 21 .'YWxd3 xf5 22.@f2 xf3 23.Wxf3

24

23 ... J::ldS 23 . . . 'YWc6+ 24 .dS tllxdS 2 5 .tllxdS 'YWxdS+ 26 .'YWxdS exdS 27 . .l:!.ah l h6 28 .g4 l:laf8 29 .e6+-.

24.!iah1 +- h5 25.tlle2 'YWg6 26.g1 'YWf7 27.tllg3 l:lxe5 28.tlle4?

28 .@e2 eds 29 .fS l::rxfS 30 .tllxfS exfS 3 1 ..&!.gs tllds 3 2 .l::rxfS tllf6 3 3 .!ihxhS + @g8 34 . .l:!.hgS+-.

28 ... ..thG 29.@e2

29 ... xf4?? Mistakes like this give me the general impression that the standard of play was a lot lower then than now. The players who finished below 5 0% in major tournaments back then were probably below what we would call master level ( 2200) today in the USA. Kasparov remarked in the 1 980s that even the grandmaster who finished last in a certain tournament (Ljubojevic) was stronger than Capablanca had been half

Chapter 1: Hi s t ory of t h e f3 Ant i - Griinfeld

a century earlier. Although not a diplomatic thing to say, it was probably true. 29 . . . fs 30 .tt:Jg3 .

30.xf4 f5 Maybe Black missed 3 0 . . . xf4? 3 1 .dxeS l::ixd3 3 2 .xhS+ with mate next when he took on f4.

31 .e3? 3 1 .ti:Jg3+- .

3 1 ... l::rf8 32.tt:JgS g6 33.l:igh1 @g8 34.l::rg1 @ha 35.tt:Jf3 f6 1 5.tt:Jge2 36.lagS 1::1.xgS 37.xgS f7 1 s . .!d.a4;. 38.f6+ 1-0

(D70) Game 1 .6 Fine, Reuben Dake, Arthur William Detroit 1 9 3 3 ( I 0) 1 .d4 tt:Jf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tt:Jb6 6.e3 g7 7.tt:Jc3 o-o 8.d2 es 9.d5 c6

1 0.a4? Only 1 O .h4 promises advantage, as shown in the theoretical section, while 1 0 . .lld 1 is playable.

1 0 ... cxd5 1 1 .exd5 h4+? With 1 l . . . tt:Ja6 !+ Black prepares . . . tt:Jb4 while showing that 1 2 .aS was not a threat due to 1 2 . . . tt:Jc4 1 3 .xc4 h4+ .

1 2.f2? 1 2 .f2 ! .

12 ... b4 1 3.aS tt:Jc4 14.xc4 vtxc4

Reuben Fine

1 5 ... tt:Ja6= 1 6.0-0 tt:Jc5 1 7.xcS?! xc5+ 1 8.@h1 .ld.d8 1 9.gS

1 9 .. .f6?

25

S a b o t a g e t h e G riinfeld

1 9 . . . .id7+ with the bishop pair versus two knights and a fluid position.

20.'t!Vh4 g5 21 .'t!Va4 .if5? The pawn should go there as now the g7 -bishop is pretty bad.

22JUd1 ac8 23.t2Jg3 .ig6 24.t2Jge4;!; 't!Vc4 25.'t!Va3?

2 5 .'t!Vxc4 l:lxc4 2 6 .a6;l;. 25 ... .ifS 26.d6?

26 ... @g7? With 26 . . .fS 2 7 .tLlxgS l:rc6 28 .l:lac l l:lcxd6+ Black regains the pawn with a dominating position.

27.d5? Jl.f7? 2 7 . . . .ixe4 2 8 .fxe4 l:!.c6 wins a pawn.

28 . .l:!.d2 .l::!.c6 28 .. .fS ! 29 .tLlxgS .ig8+.

29 . .l::l.ad1 ? 29 .g4! to prevent 29 . . .fS .

29 ... .l::!.a6? 29 . . .fS ! 3 0 .tLlxgS .l:!.cxd6+.

30.'t!Va1 b6 31 .'t!Vc1 ;t .a:xa5?

32.t2Jxf6!

26

Although this was not a well played game up to here, the finish is very nice!

32 ... @xf6? 33.t2Je4+ @g7 34 . .l:lc2 't!Va4 35.'t!Vxg5+ .ig6 36 . .l:!.c7+ @g8

3 6 . . . J:[d7 3 7 .b3+- . 37.'t!Vxg6+ 1-0

On a personal note, I met both of the players in this 80 year old game about 40 years later.

(E60) Game 1.7 Yudovich, Mikhail Freiman, Sergey Leningrad ch-URS 1 934

1 .d4 t2Jf6 2.c4 g6 3.f3 e6 Amazingly, this move, which VachierLagrave has brought back from obscurity, dates back to 1 934!

4.t2Jc3 4.e4 d5 5 .cxd5 exd5 6 .tLlc3 is better, see the theoretical section.

4 ... d5

-.t[t I i i . , . ,

. . , , . . . , . .

8 .8. 8 88 .:: 'iVi. .::

5 . .ig5 5 .cxd5 t2Jxd5 ! (5 . . . exd5 ? ! 6 . e4 transposes to the recommended line for White) 6 .e4 t2Jxc3 7 .bxc3 c5 is an Exchange Grunfeld where . . . e7 -e6 looks more helpful to Black (restraining d4-d5) than f2-f3 does to White.

5 ... e7 6.'t!Vd2 In the event of 6 .e4 0-0 7 .h6 !:!.e8 8 .e5 tLlh5 9 .f4 c5oo it's hard to say who has the better chances.

Chapter 1: H i s tory of the f3 An t i-Griin feld

6 ... t2Jc6 It's usually wrong to block the c-pawn like this when the cl-pawn is advanced two squares, because neighboring pawns that are not part of the castle should work together.

7.e3 t2Jg8?! 8.f4 h4+ 9.g3 xg3+ 1 0.hxg3;t

White has space while Black has a bad bishop.

1 0 ... 'fHdG 1 1 .0-0-0 t2Jge7 1 2.g4 b6 13 .l2Jh3?!

1 3 .e4 a6 1 4.l2Jh3 +- . 1 3 .. .fG? 14.l2Jf4?!

1 4.e4 dxc4 1 5 .xc4 l2Ja5 1 6 .d3 tbec6 1 7 . tbf 4+-; the threat of dS is too strong.

14 ... b7? 1 5.l2Jb5+- d7 1 6.t2Jxe6 .l::!.c8 1 7.tbf 4 dxc4 1 8.xc4 t2Ja5 1 9.e2

1 9 .d3 ! +-. 1 9 ... gS 20.l2Jh5 0-0 21 .e4 c5 22.l2Jc3 fHe6 23.d5 'fHd6 24.f4 h6 25.e5

Black resigned, because if 2 5 . . . fxeS 2 6 .fxgS gives White a crushing attack. This was a rather one-sided game.

(E60) Game 1.8 Engels, Ludwig Opocensky, Karel Bad Nauheim 1 935 (5)

1 .d4 t2Jf6 2.c4 g6 3.f3 c5 4.d5 b5?!

Amazing, a version of the Benko Gambit before Benko ever played his gambit!

5 .e4

The problem for Black is that f2-f3 (on move 4 or 5 ) against the Benko Gambit is best met by . . . e7-e6, not by . . . g7-g6, which is already on the board.

5 ... d6 6.t2Jc3?! 6 .cxbS ! a6 7 .l2Ja3 g7 8 .l2Je2 0-0 9 .tbc3 l2Jbd7 1 0 .e2;t.

Analysis diagram

27

S a b o t ag e t h e G rii n f e ld

White has a good version of the Benko Gambit here.

6 ... bxc4 6 . . . b4!= ; White should not allow his knight to be kicked thus in the Benko.

7.xc4 g7 8.tLlge2 t2lbd7 9.0-0 t2lb6

After 9 . . . 0-0= Black is fine. White really should have grabbed the pawn when he could. Black should not invite the check by moving his knight before castling.

1 o.b5+ it.d 7 1 1 .a4 1 1 .d3;!;; the black bishop on d7 is misplaced, preventing the knights from using this vital square.

1 1 ... 0-0 1 2.f4!? Perhaps too ambitious.

1 2 ... xb5 1 3.axb5 e6 14.dxeG fxe6 15 .d3 e7= 1 6.h3 d5 1 7.e5 t2ifd7 1 8.b3 g5 1 9.it.d2 gxf4

1 9 . . . c4 20 .tLld4 tLlcS 2 1 .tLlc6 e8 2 2 .bxc4 t2lxc4 23 .it.e3.

20.tLlxf4

20 ... c4?? 20 . . . it.xeS 2 1 .tLlxe6 1:1.xfl + 22 .l:rxfl lle8 2 3 .g4+ h8 24.tLlgS tLlf6=.

21 .tLlxeG cxb3?? 2 1 . . .xfl + 2 2 .l::!:xfl it.xeS 2 3 .bxc4 kte8 24.cxdS tLlxdS 2 5 .tLlxdS xe6 26 .d3 and White is a healthy pawn up.

28

22.g5+- b4 23.tLlxfS tLixfS? 24.fG b2 2s.nab1 ncs 26.it.xg7 d4+ 27.Wh1 Wxg7 28.eG! tLlgG? 29 . .l::If7+ 1-0

Black was a rather famous player I opening innovator, so I 'm surprised to see so many tactical mistakes by him.

(D7 0) Game 1 .9 Goglidze, Viktor Levenfish, Grigory Moscow 1 935 (7)

1 .d4 tLlfG 2.c4 g6 3.f3 d5 4.cxd5 t2ixd5 5.e4 tLlbG 6.e3 it.g7 7.t2lc3 tLicG 8.b5 0-0 9.tLlge2

9 ... t2ia5 9 . . . es 1 O .dS tLlaS is an interesting piece sacrifice analyzed in the theoretical section.

1 O.b3 a6 1 1 .d3 tLicG 1 2.c2 t2lb4 1 3Jk1 t2ixc2+ 14.xc2 a5

C h a p t e r 1: H i s t o ry of t h e f3 An t i - G ri i n fe ld

15 .0-0 1 5 .dS ! a4 1 6 .0-0 transposes to the note to move 1 6 while avoiding the note to Black's 1 S th move.

1 5 ... a4 1 5 . . . c6 1 6 . .l:!.fd l a4 1 7 .dS cxdS l 8 .bxa4;l;.

1 6.tt:Jxa4? 1 6 .dS ! axb3 1 7 .axb3 d7 (after 1 7 . . . c6 l 8 .dxc6 bxc6 1 9 . .l:!.fd l d7 20 .hl tbc8 2 1 .tba4 tbd6 22 .tbd4 the pressure on c6 and on the d-file, and the option of tbcS give White a clear plus despite Black's bishop pair) 1 8 . .l:!.fd 1 tbc8 l 9 .1/Wd2 tbd6 20 .d4 f6 2 l .tlJf4;t_ White's huge edge in space and piece placement easily outweigh Black's bishops.

1 6 ... tt:Jxa4 1 7.bxa4 c6 1 8.fd1 'if as 1 9.tt:Jc3 e6

20.dS?! Better was 20 .l:rb l b6 2 1 .tbe2 d7 22 .'id2 1/Wxa4 23 . .l:!xb6 '1Wxa2 24.I:l.b7 when White's rook on the 7 th and superior pawn structure slightly outweigh the bishop pair.

20 ... cxdS 21 .tt:JxdS xd5 22.f!xd5 1/Wxa4 23.1/Wxa4 l:l'.xa4

Draw agreed in this totally equal endgame. Black in this game was one of the world's best players.

(D7 0) Game 1.10 Engels, Ludwig Richter, Kurt Aachen ch-GER 1 935 (4)

1 .d4 tt:Jf6 2.c4 g6 3.f3 d5 4.cxd5 c6

This is an interesting gambit. Probably after S .dxc6 tbxc6 6 .e3 eS 7 .bS Black doesn't get full compensation for the pawn.

5.e4 cxd5 6.e5 tt:Jfd7 7.tt:Jc3 7 .d3 ! tbc6 8 .tbe2 e6 9 .e3;!; looks like a rather good version of the French for White.

7 ... tt:Jb6 After 7 . . . tbc6 ! 8 .tlJxdS tbdb8 9 .c4 e6 1 O .tbe3 xc4 l l .tbxc4 bS 1 2 .tba3 '1Wxd4 Black i s slightly more comfortable.

8.d3 tt:Jc6 9.tt:Jge2 g7 1 0.0-0 0-0 1 1 .f 4 g4 1 2.h3

29

S a b o t a g e t h e G riinfeld

1 2 ... f5?! Normal would be to take the knight, since this is a rather closed position. Apparently Black provoked h2-h3 before playing . . . . f5 to avoid the rook maneuver :I.-f3 -h3 .

13 .xf5 gxf5 1 4.b3 @h8 15 .e3 ld:g8 1 6.tllg3?!

tll-c l -d3 looks better. 1 6 ... xe5?

1 6 . . . e6 1 7 .f2 f8=. 1 7.tllxf5;t g7 1 8.WVh5

l 8 .tllxg7 l:txg7 1 9 .fSt . 18 ... WVf8

1 9.tlle2? l 9 .g4t.

1 9 . . . tlld7= 20.g4 tllf6 21 .WVh4 tlle4 22.Wh2 l::rc8 23 . .l::!.ac1 e6 24.tllxg7 WVxg7 25.f5 e5? 26.h6 WVf6 27.WVxf6+ tt:Jxf6 28.e3?

After 2 8 .dxeS tllxe5 29 .@g2;\; White's better pawn structure and very strong bishop give him the edge.

28 .. Jke8= 29.f2 h5 30.h4 tlle4 31 .tllg3 hxg4? 32.tllxe4 dxe4 33.d5?

3 3 .f6+ @h7 34.hxg4 .l:!.xg4 35 . .i::lc3 @g8 3 6 .dS tlle 7 3 7 . .i::lh3 @f8 3 8 . .l::!.d l +- .

33 ... tlld4 34.d6?? 34.f6+ @h7 3 5 .hxg4 .!::l.xg4 36 . .l::!.c3 tllxf5 3 7 .l:[xfS @g6 38 . .l::!.xeS+-.

30

34 ... tllf3+ 35 . .l:!.xf3 gxf3 36.d7 .l:!.a8 37 . .l::!.c4? @g7 38 . .l::!.xe4 f6

White resigned, as after the coming exchange of f3 for d7 he will be down the exchange for nothing.

(D7 0) Game 1.1 1 Bondarevsky, Igor Aronin, Lev Moscow ch-URS 1 948 ( 1 8)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tllb6 6.tllc3 g7 7.e3 tllc6 8.b5 0-0 9.tllge2 e5 1 0.d5 tt:Je7

This concedes White some advantage, whereas 1 O . . . tlla5 is a speculative piece sacrifice. I have analyzed this idea deeply in Chapter 3 , in Game 3 . 1 between Khismatullin and Yandemirov, after Black's 9th move.

1 1 .0-0 1 1 .cS tlld7 1 2 .f2 c6 1 3 .dxc6 tllxc6 1 4.0-0;\; appears to be a slight refinement over the game as the bishop is brought to a less vulnerable square with no loss of time.

1 1 ... c6 1 2.dxc6 tt:Jxc6 1 3.c5 .l:!.e8 14.tlld5

1 4.WVe l e6 1 5 .WVf2;t looks a bit better, as Black's queen is less active than White's. White will keep a lead in development.

Chapter 1: H i s t ory o f the f3 An t i - Gri i n feld

14 . . . fS 1 5.e3 d7? 1 5 . . . e6 1 6 .tl:lec3 tl:ld7 1 7 .e2 a6 1 8 .d3 - White has moderately superior development and is slightly better after, say, 1 8 . . . .ld.c8 .

1 6.tl:lec3 tl:lxd5 1 7.tl:lxd5 g7 18 . .l:Ic1 .l:Ic8 1 9.d2 a6

24 .. .f5?? After 24 . . . tl:ld4 2 5 . .l:Ixc8 xc8 26 .xd4 xdS 2 7 .xdS exd4 28 . f4 cs 29 .xd4 xd4+ 30 .l:lxd4;\; the opposite-colored bishops give Black fair drawing chances despite being a pawn down.

25.tl:lb6 xb3 26.axb3 xd2 20.a4 27 . .ld.xd2 f4 28.f2?

20 .b6 ! h4 2 1 .a4 was better, to 2 8 . tl:lxc8 ! fxe3 2 9 .l::!.d7 ( 29 . .l:Id6 avoid the useful 20 . . . bS . tl:ld4) 29 . . . l::!.xc8 3 0 . .l::!.d6 was win

Igor Bondarevsky

20 ... b5 21 .b3 e6? Black should not have missed the chance to play 2 1 . . . tl:ld4! , even though it doesn't equalize.

22.g5! f6 23.e3 @h8 24 . .l:Ifd1

ning for White. 28 ... .!::!.cdS 29 . .l::i.xdS

29 ... .l:IxdS? 2 9 . . . tl:lxd8 would avoid the loss of a piece, but still it loses a pawn: 3 0 .l::!.c8 .l:If8 3 1 .l:la8 tl:lc6 3 2 .xa6 tl:lb4 3 3 .l:laS tl:ld3 34 .@fl tl:lxf2 3 S .@xf2 b8 3 6 .tl:lds f8 3 7 .b4+- .

30 . .!::!.xc6 .l:f.d1 + 31 .e1 .l:Ixe1 + 32.@f2 .lab1 33..!kS+ f8 34 . .ld.xfS+ @g7 35 . .ld.eS .l:Ixb2+ 36.@e1 .ld.b1 + 37.@d2 i::txb3

3 1

S a b o tage t h e G r ii n f e ld

38.ld'.xe5 ld'.b2+ 39.\t>d3 l:rxg2 40.ld'.e7+ 'itif6??

40 . . . 'it>h6 4 1 .eS J::l.f2 42 .'itie4 ld.xh2 43 .'itixf4+- .

41 .tlld5+ 'itig5 42 . .ld.xh7 1-0 Black must lose his rook to delay mate. The opponents were both top Soviet players, especially Bondarevsky who was a candidate for the World Championship.

(D7 0) Game 1 . 1 2 Enevoldsen, Jens Heinicke, Herbert Oldenburg 1 949 (3) 1 .d4 tllf6 2.c4 g6 3 .f3 d5 4.cxd5 tllxd5 5.e4 tllb6 6 .tllc3 g7 7.jie3 0-0 8.'iYd2 tllc6 9.0-0-0 e5 1 O.d5 tlld4 1 1 .f4 c5 1 2.fxe5 g4 1 3.l::re1 .ixe5 14.h3 'iVh4

This line is still being played, but I think White is significantly better.

15 .'itib1 1 5 .d3 also favors White: 1 5 . . . ld'.ac8 1 6 .gS 'iYhS 1 7 .'itib l .id7 1 8 .ld.fl f6 1 9 .h6:t .

1 5 .. .f6?! 1 s . . . fs 1 6 .f2 'iVhs 1 7 .hxg4 'iYxh l 1 8 .gxfS 'iVhS 1 9 .tllf3 tllxf3 20 .gxf3 'iVh2 2 1 .d3 c4 22 . .ic2 ae8 2 3 . .l:!d l gxfS 24.exfS:t .

3 2

Analysis diagram

White is down 1,4 pawn and has a bad pawn structure, but the dangerous passed d-pawn combined with threats to the black king will give him the edge.

1 6.!k1 1 6 .tllbs hs 1 7 .f2 .ig3 1 8 .xg3 'iYxg3 1 9 .tllxd4 cxd4 20 .tlle2 .ixe2 2 1 .xe2 'iYes 2 2 .'iYd3 tlld7 2 3 . .l::!.c l .l:i.ac8 2 4 . .l::rhd l tllcS 25 .'iYxd4 tllxe4 26 .l::rxc8 l::rxc8 2 7 .'iYxeS fxeS 2 8 . .ig4.

Analysis diagram

White should be slightly better in this endgame, with rook and bishop vs. rook and knight and the more advanced passer.

1 6 ... a6 1 7.f2 1 7 . .id3 l::rac8 1 8 .tllge2 tllxe2 1 9 .tllxe2 .ixe2 20 .xe2 'iYxe4+ 2 1 . .id3 'l!Ha4 2 2 .xcS .if4 23 . .ie3 .ixe3 24.'iYxe3 'tWb4 2 5 .a3 'iYd6 26 .'iYe6+ 'iYxe6 2 7 .dxe6 l::rfe8 2 8 .e4:t .

Chapter 1 : H i s tory of the f3 An t i - Gri i n feld

Analysis diagram

The rook and bishop will be stronger than the rook and knight here.

1 7 ... 'tWh5 1 8.tt:Jce2 1 8 .d3 ! fS 1 9 .tllge2 xe2 2 0 .xe2 tllxe2 2 1 .tllxe2 tlld7 2 2 .xcS tllxc5 23 ..lhcS 'tWh4 24.exfS 'tWe4+ 2 5 .'tWc2 'ixf5 26 . .ld.d l .

Analysis diagram

Black has only partial compensation for the pawn.

21 .tllf3? 2 1 ..l:i.e l f4 2 2 .'tWc3 es 2 3 .'tWd2 leads to a draw by repetition.

21 ... tlla4 22.e2?? A) 2 2 .tllh4 'tWd7:+; B) 2 2 .l:Ie l xb2 2 3 . .ld.xe4 VWxe4

24.d3 'tWxdS 2 5 .tlle3 'tWc6 2 6 .c2 a3+.

22 ... xb2 23.g4 'tWd7 24.xc5 xc1 25.xc1 tt:Jxc5 0-1

(D70) Game 1 . 13 Enevoldsen, Harald Bolbochan, Julio Dubrovnik Olympiad 1 95 0 (7)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tllb6 6.tllc3 g7 7.e3 0-0 8."id2 tt:Jc6 9.0-0-0 e5 1 O.d5 tlld4 1 1 .f4 g4 1 2.e1 c5

1 8 ... f5 1 9.tllxd4 xe4+ 1 3.h3 20.tllc2 'iif5 1 3 .fxe5 xe5 l 4.h3 is the more accu

rate move order. 1 3 ... d7

With 1 3 . . . exf4 1 4.xf4 d7 1 5 .tllf3 l:le8= Black retains an outpost on d4.

14.fxe5 xe5 1 5.tllf3 tt:Jxf3 1 6.gxf3 'tWe7

If 1 6 . . . .ld.e8 1 7 .d3 .i::rc8 1 8 .@b l ;l; and with ideas of f3 -f4 and h3 -h4, White must be better.

1 7.h4

33

S a b o t a g e t h e G riinfel d

1 7 ... h5 1 7 . . . fS was relatively best, but 1 8 .hS or 1 8 .gS give White a clear advantage.

1 8.d3 tt:Ja4 1 9.tt:Jxa4 xa4 20.f4 c4 21 .fxe5 cxd3 22.d4 l::tac8+

23.@b1 ? 2 3 .c3 ! bS 24.'b l c2+ 2 S .Wa l aS 2 6 .ld.hg l t. White's attack is stronger than Black's.

23 ... c2+ 24.@a1 1d.c4 25.c3 .l:!.a4 26J1e3??

26 . .l:!.h2 cs+ .

34

26 ... b5?? After 2 6 . . . cS ! ! there is no reasonable defense to the threat of 2 7 . . Jha 2 + with mate to follow soon.

27 . .l:!.xd3 xd3 28.xd3 b4-+ 29.b3

After 29 .d2 xeS Black is just up the exchange.

29 ... bxc3 30.bxa4 xe5 31 .c2 .l:!.b8 32.h3 b4 33.d3 xe4 34.xc3 xc3+ 35 . .l::!.xc3 ld'.xh4 36.l'.ld3 .l:!.b4 37.dG J::rb8 38..l:k3 @ta 39Jk7 h4 40.d7 1:1.da 0-1

(D70) Game 1 . 14 Evans, Larry Melvyn Byrne, Donald New York ch-USA 1 963/64 ( 1 1 )

1 .d4 tlJfG 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tt:Jb6 6.tt:Jc3 g7 7.e3 0-0 8.d2 tt:Jc6 9.0-0-0 a5?!

This is reasonable, but doesn't seem to be satisfactory.

1 0.h4 1 o .@b l a4 1 1 .tLlbS tlJaS 1 2 .tlJa3t was a fancy way to neutralize Black's counterplay.

1 0 ... eG? This is just too slow. After 1 O . . . a4 1 1 .hS tlJaS 1 2 .hxg6 fxg6 1 3 .gS 'iYd6 1 4.Wb l tlJbc4 l S . .ixc4+ tlJxc4 1 6 .c l t White is for choice with his strong center and open h-file

Chapter 1: H i s t ory of t h e f3 An t i - Gri i n fe ld

in return for the bishop pair, but at least Black has counterplay.

1 1 .h5 a4 1 2.g4+-White is already winning due to the threat of 1 3 .hxg6 and 1 4.WVh2 .

1 2 ... a3 13 .b3 WVe7 14.hxg6 fxg6 1 5.WVh2 h8 1 6.f4 .!:lf7 1 7.tt:lf3 b4

1 8.@c2 After 1 8. l::!'.d3 ! +- Black has no attack and no defense against 1 9 .tllg S .

1 8 ... l::raS 1 8 . . . es ! 1 9 .dxeS xg4 2 0 . .ixb6 .ixf3 2 1 .c4 xd l + 2 2 .tllxd l @f8 23 . .ie3 and White should still win, but at least this line is messy.

1 9.e5+- WVe7 20.tt:lg5 .ld.g7 21 .tllce4 tt:Jd5 22.tllfG+ @f8 23.tllxd5 .ld.xd5 24.tt:lxh7+ .ld.xh7 25.WVxh7 WVxh7 26 . .ld.xh7 g7 27.g2 tt:lb4+ 28.@b1 'it>g8 29.xd5 tt:Jxd5 30 . .l::!.h3 tt:lc3+ 31 .@c2 tt:Jxd1 32.@xd1

Larry Evans

With the exchange and a pawn for the bishop pair (plus 2 1/4 pawns) , the win is almost in the bag.

32 ... .ifS 33 . .id2 d7 34.b4 b5 35 . .l::!.xa3 c6 36.@e2 d5 37.a7 c6 38.a3 g7 39.e1 .ih6 40.@e3 b3 41 .h4 d5 42.fG f8 43.@12 h6 44.@g3 f8 45.I;lc7 e4 46.@h4 d5 47.f5 1 -0

(A5 7) Game I.I S Zaitsev, Alexander Adorjan, Andras Polanica Zdroj I 9 7 I ( 3) 1 .d4 tllf6 2.c4 g6 3.f3 c5 4.d5 d6

4 . . . bS 5 .e4 d6 6 . cxbS transposes to the game while avoiding 4 . . . d6 5 . tllc3 , which prevents . . . b7-b5 .

5.e4 b5 6.cxb5 a6

35

S a b o t a g e t h e G riinfeld

7.bxa6 7 .tlla3 ! ilg7 8 .tlle2 0-0 9 .tllc3 . White scores 67% in 69 games from here in the database. It's a bad Benko for Black because White will keep a piece on bS to block the usual pressure down the b-file, for example: 9 . . . tllbd7 1 0 .ile2 tlle8 1 1 .0-0 tllc7 1 2 .c2;t . This is perhaps even better than the usual 1 2 .gS . In neither case does Black have much to show for the pawn.

7 ... g7 8.ilb5+ tllbd7 8 . . . tllfd7 ! 9 .tlle2 ilxa6= and Black seems to have normal Benko Gambit compensation here.

9.tlle2 a5+ 1 0.tllbc3 xa6 1 1 .xa6 xa6 1 2.0-0 0-0

1 3.ile3 1 3 .ilgs;t - in the Benoni it's usually better to play ilgS to provoke . . . h7-h6 and then retreat to e3 than to play the bishop directly to e3 . Once you have met . . . c7-c5 by d4-d5 , the bishop on e3 is attacking a strong pawn on cS ; it's better to at least have one target on h6 that is not pawn-protected.

1 3 ... llfbS 14.c2 tlle5 1 5.b3 tlle8 1 6.r.!.ad1 tllc7 1 7.h3

1 7 .a4 to prevent Black's next move might be best, though Black has good compensation.

1 7 ... tllb5 1 8.tllxb5 .!::rxb5 1 9.g5 llb7 20.f4

36

20 ... h6 Or 2 0 . . . tlld7 2 1 .ilxe7 xa2 2 2 .xa2 .l::rxa2 2 3 .tllc l .l:i.a6= and Black has full typical Benko Gambit compensation.

21 .ilh4 g5?! 22.fxe5 gxh4 23.exd6 exd6 24 . .l:i.d2 ba7

25.e5?! After 2 5 .tllc l t Black's compensation looks thin in view of his own weak pawns.

25 ... xe5 26.e4 ca 27.xh4 'i!Yd8?

2 7 . . . .!:rxa2 28 . .!:rxa2 .l:!xa2=. 28.xh6 l:rxa2??

Chap t er I : H i s t o ry of t h e f3 Ant i - G riinfeld

29 . .l::!.xa2?! 29 . .l:rf5 ! +- .

29 ... .l:!.xa2 30 . .ld.f5 .ld.xe2 31 . .l::!.g5+ 'iWxg5 32.'iWxg5+ Wf8 33.h4+.l::i.b2 34.h5 .l:!.xb3 35.h6 l::tb1 + 36.wf2 wea 37.'iWh4

3 7 .h7 .id4+ 38 .We2 .l:i.hl 3 9 .Wd3+-. 37 ... .l:tb2+ 38.We3 .l::i.b3+ 39.

S a b o t a g e the G riinfe ]d

1 0 ... g4 1 0 . . . e s != ; see the theoretical section.

1 1 .e2! e5 l l . . . .ixe2 l 2 .tLlgxe2;\;.

1 2.fxeS?! It's better to exchange the important g7-bishop: 1 2 .xeS ! .ixeS 1 3 .fxeS .ixe2 1 4.'iYxe2 tLiac4 l S .tlJf3 'iYe7 l 6 .h4 tLixeS 1 7 .hS c6 1 8 . 0-0-0;\;.

Analysis diagram

White's h-file play should give him some edge.

1 2 ... .ixe2 1 3.'iYxe2 tt:Jac4 ?! 1 3 . . . 'iYh4+ 1 4.g3 'iYe7 l S .tlJf3 tLiac4 1 6 . 0-0-0 'iYb4= .

14.tt:Jf3 'iYe7

Emil Ungureanu

38

1 5.e6?! After l S . 0-0-0 tLlxeS 1 6 .@b l ;\; White's center is his edge.

15 .. .fxe6 1 6.xg7 'iitxg7 1 7.b3 tt:Jd6?

Better was 1 7 . . . tlJeS ! 1 8 .tLlxeS 'iYf6 1 9 .dxe6 'iYxeS 20 .'iYb2 :lae8 2 1 . 0-0-0 .l:!.xe6=i= and Black has the better pawn structure and better piece placement.

1 8.dxe6 'iYxe6 1 9.0-0 ae8 20.'iYb2 'iitg8 21 .11ae1 ;t; tt:Jd7 22.'iYf2?!

2 2 .eS tlJfS 2 3 .tlJe4t ; White should not allow the passer to be blockaded on e4 when it can safely reach eS .

22 ... tt:JeS 23.'iYxa7 b6 24.'iYa6 tt:Jxf3+ 25.gxf3 'iYh3 26.'iYe2 es 27.@h1 .lags 28.l:rf2 :!d:g3??

29 . .l::!.ef1 ? 29 .eS ! tt:Jfs 3 0 .tLle4 'iYh4 3 1 .'iYc4+ @h8 3 2 .tlJxg3 tLixg3+ 3 3 .@g2+- and White is just up the exchange and a pawn.

Chapter 1 : H i s t ory o f the f3 An t i - Grii n f e ld

29 ... tllf7? 30 . .l:Ig2? 30 .e S +- .

3 0 ... l::rxg2 31 .'ixg2 'ie6 32.tlld5 32 ... c6?

Black needlessly loses a second pawn. He should have tried 3 2 . . . tlld6 3 3 .tllxc7 'if7 34 .tlldS tllxe4 3 5 .tllxb6.

33.tllxb6+- tlle5 34.f4 tlld3 35.f5 'ie7 36.tllc4 g7 37.e5 'iY c5 38.f6+

Black resigned, as 3 9 .e6 next is crushing.

(D70) Game 1. 1 7 Ungureanu, Emil Ftacnik, Lubomir Bucharest 1 97 8 (2)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tllxd5 5.e4 tllb6 6.tllc3 g7 7.e3 0-0 8.f4 tllc6 9.d5

9 ... tllb8?!

9 . . . tllas ! 1 O .d4 eS=. 1 0 . .id4?!

1 O .a4 !t: - it's quite helpful to weaken the knight on b6 this way.

1 o ... .ixd4 1 1 . xd4 c6 After l l . . . e6 1 2 .tllf3 exdS 1 3 .exdS g4= Black has no development problems, and f3 -f4 is just a weakening.

1 2.tllf3 cxd5 1 3.exd5

1 3 ... tll8d7 1 3 . . . e6 1 4.d6 tllc6 1 5 .Wk'cs eS= is complex but Black can equalize.

14 . .ie2 tllf6 1 5.a'.d1 e6 1 6.dxe6 jLxe6 1 7.0-0 'ixd4+ 1 8 . .l:lxd4 tllfd5

Other things being equal, it makes sense to exchange off the less centralized knight : 1 8 . . . tll bdS 1 9 . tllxdS .ixdS=.

1 9.tllxd5 .ixd5 20.b3 ac8 21 . .l:Ifd1 llc2 22.a'.1 d2 a'.fc8 23.f2 .!axd2 24 . .&rxd2

39

S a b o t a g e the G r iinfeld

24 ... f6 24 . . . 'it>f8=. 24 . . .f6 is an unnecessary weakening.

25.tlld4 a6 26.tllc2 @fa 27.tlle3 jic6 2a.f3 xf3 29.@xf3 @e7 30.f5 .lk5 31 .g4 tlld 7 32.tlld5+

White is obviously for choice, but it's pretty drawish.

32 ... @f7 33.@e4 gxf5+ 34.gxf5 lk1 35.tlle3

35 ... tlle5?! 35 . . . @e?=.

36 . .l:;td6t The rook breaks into the enemy camp.

36 ... ld.e1 ? 3 6 . . . ld.c6 3 7 . .l:;td8t.

37.h3 ld.e2 3a.a4 @ga

39.Jada+?! It's hard to guess why White didn't take the pawn with 3 9 .!txf6.

39 ... wg7 40.nba h5?

40

40 . . . bS 4 1 .ltb7+ @f8 42.axbS axbS 43 .xbSt.

41 .nxb7+ 'it>h6 42.l::rb6? 42 .kla7+-.

42 ... tlld7 ! 43.l::!.d6 tllc5+ 44.'it>f3 .l::!.h2 45 . .l:Ixf6++-

Now White has a winning advantage. The rest needs no comment.

45 ... @g7 46 . .l::!.g6+ @f7 47.@g3 ld.b2 4a.nc6 .l:!.xb3 49.'it>f4 .l::rb4+ 50.@g5 tlle4+ 51 .@h6 .l::!.xa4 52.tllc4 h4 53.tlle5+ @e7 54.l::f.e6+ @da 55.f6 tllxf6 56 . .l:i.xf6 .i::ta3 57.tllf3 a5 5a.@g5 @e7 59.ld.f4 .l:Ib3 60.@g4 nb4 61 .J::l'.xb4 axb4 62.tlld2 @f6 63.@xh4 @g6 64.'it>g4 @h6 65.tllb3 @g6 66.tlla5 @h6 67.h4 @g6 6a.h5+ @h6 69.@h4 @h7 70.@g5 @g7 71 .h6+ @h7 72.@h5 @ga 73.'it>g6 @ha 74.tllc6 b3 75.tt:Jda b2 76.tllf7+ @ga 77.h7+ 1 -0

(D70) Game 1 . 1 8 Bjarnason, Saevar Svenn, Matti Stockholm 1 97 8 ( 6) 1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tllxd5 5.e4 tllb6 6.tllc3 g7 7.jie3 tllc6

Chapter I : H i s to ry of t h e f3 A n t i - G ri infeld

8.d5 This is the safer route to a plus, 8 .bS being the alternative.

a ... tlle5 9.f4 tllg4 1 o.b5+ c6 1 0 . . . d7 1 1 .d4 xd4 1 2 .'i1'xd4 0-0 1 3 .xd7 'i1'xd7 1 4.h3 tllf6 1 S .a4t. This is often a good move against a knight on b6. The reply . . . a7-a5 makes . . . c7-c6 unplayable.

1 1 .dxc6 0-0 1 2.'iVxdS .l::!.xd8 1 3.c7t

Black will have to struggle to regain the pawn.

1 3 ... fS 14 . .itc5

14 ... d7? Better was 1 4 . . . xc3+ 1 5 .bxc3 d7 1 6 .a4 .l::!.fc8 1 7 .aS xbS 1 8 .axb6 axb6 1 9 ..l:ha8 l:f.xa8 20 .xe7 fS 2 1 .tlle2 fxe4 22 .tlld4 d3 23 .@d2t , but the 'bomb' on c7 gives White a plus.

1 5.0-0-0+- xb5 1 6.tllxb5 xb2+ 1 7.@b1 tlla4 1 8.xe7

1 8 ... tllf2? But also after 1 8 . . . .l:!.fe8 1 9 .l::!.d8 l:texd8 2 0 . cxd8'Y!!V+ .l::!.xd8 2 1 .xd8 tllf2 22 .tllxa7 tllxh l 2 3 .tllbS tllf2 24.eS tllg4 2 5 .tllf3 +- , with a healthy extra pawn and better piece placement, White should win.

1 9.xfS? 1 9 .tllh3 ! tllxd l 2 0 .l:!'.xd l .l:!.fc8 2 1 . .l:!.d8+ @g7 22 .tlld6 J:lxc7 2 3 .tlle8+ @g8 24.tllxc7 + .l:!.xd8 2S . .itxd8+and White is a clear knight ahead.

1 9 ... tllxd 1 20.tllh3

20 ... tlle3?? He should have played 2 0 . . . tlldc3 + 2 1 .tllxc3 xc3 22 .d6 as 23 . .ld.c l l::tc8 24.eS. White is a pawn up, but the win is not yet certain.

21 .e7 .l:!.c8 22.ne1 tt:lc4 23Jid1 @g7 24 . .:bi.dS tllab6 25.e5 f5 26.ta+ 1 -0

It's mate in three.

4 1

S a b o t a g e the G r iinfeld

(D70) Game 1 . 1 9 Gheorghiu, Florin Jansa, Vlastimil Warsaw Zonal 1 9 79 ( 4) 1 .d4 tllf6 2.c4 g6 3.f3 dS 4.cxdS tt:JxdS 5.e4 tllb6 6.tllc3 g7 7.e3 0-0 8.f4 tllc6 9.dS tllaS 1 0.d4

1 0 ... g4?! 1 0 . . . e S != .

1 1 .'id3?! 1 l .e2 ! xe2 (for 1 1 . . . eS see the game Ungareanu-Pavlov) l 2 .tllgxe2;\;.

1 1 . . . eS!= 1 2.fxeS tt:Jac4 1 3.'ig3

1 3 ... 'iYgS?? 1 3 . . . hS 1 4.xc4 tllxc4 1 5 .tllf3 c8 1 6 .e6 xf3 1 7 .xg7

C h a p t e r I : History of the f3 Anti - G ri i n fe ld

23 ... bxc4 24.h4 i:i.ab8 25.laf1 V.Wh6 26.f4 V.Wf8 27.d6 c6 28.f5 V.Wd8 29.V.Wf2 V.Wd7 30.h5 I:te8 31 .V.Wg2 g7 32.'itd2 h6 33.hxg6 fxg6 34.fxg6 nta 35.lah1 h5 36.a'.xh5 J:rf1 + 37.l2Jd1 J:rh8 38 . .l:!.xhS @xh8 1 -0

l

After 3 9 . .!:lxc4 White will be four pawns up with many threats.

(D70) Game 1 .20 Hausner, Ivan Pribyl, Josef Trnava ch-CSR 1 980 ( 1 0)

1 .d4 l2Jf6 2.c4 g6 3.f3 d5 4.cxd5 l2Jxd5 5.e4 l2Jb6 6.l2Jc3 g7 7.e3 e6

8.V.Wd2 8 .f4i is also sensible, since the knight can't be pinned on f3 now.

8 ... l2Jc6 9.b5 d7 1 O.l2Jge2 a6 1 1 .d3 l2Jb4 1 2.0-0 l2Jxd3

1 3.V.Wxd3t

Black has no play against White's strong center, which together with his space advantage mean more than Black's bishop pair.

1 3 ... 0-0 14.i:i.ac1 l2Ja4 1 5.l2Jxa4 I S .'iWd2 l2Jxc3 1 6 .l2Jxc3i and if White exchanges the dark-squared bishops he will have an undisputed advantage.

1 5 ... xa4 1 6.l2Jc3?! l 6 .iVd2t .

1 6 ... b5 1 7.l2Jxb5 axb5

1 8.xb5 White need not take here, but with both a2 and d4 under pressure he has no advantage.

1 8 ... xd4 1 9.xd4 'ifxd4+ 20.'it>h1 ?!

20 . .l::tf2 ! equalizes by guarding b2 . 20 ... .l:!.xa2 21 .xb7 'ifxb2 22.'ifxb2 1:!.xb2 23.J:rxc7 .i:::i.d8 24.h4 ld.dd2 25Jg1

43

S a bo t a g e the Griinfe ld

Black is obviously for choice, but the position is a draw. Grandmaster Nunn stated that in general double rook endings were the most drawish of all; this example might illustrate his comment.

25 ... l::rf2 26.l::!:c3 @g7 27.@h2 hS 28.@g3 f6 29.c7 fd2 30.1::!.a7 ndc2 31 .@h3 1f2-1/2

(AS 7) Game 1 .2 1 Pytel, Krzysztof Pein, Malcolm Manchester 1 980 (7)

1 .c4 g6 2.d4 tl'if6 3.f3 c5 4.dS bS 5.e4 d6

S . . . bxc4 6 .eS;!;. 6 .cxb5

White is better off than in the Benko Gambit, where this f2-f3 strategy would be met by . . . e7-e6 rather than . . . g7-g6 .

6 ... a6

44

7.tl'ic3 7 .tl'ia3 g7 8 .tl'ie2 0-0 9 .tl'ic3;!;.

7 ... g7 8.a4 0-0 9.na3 axbS 1 O.xbS tl'ia6 1 1 .tl'ige2

11 ... tt:Jc7 1 1 . . .tl'ib4 1 2 . 0-0 a6 1 3 .e3 tl'id7 1 4.d2 tl'ieS 1 S .xa6 tl'ixa6 1 6 .b3;!;.

1 2.c4 1 2 .c6 as 1 3 .0-0 tl'id7 1 4.f4 tl'ib8 1 S .bS;!; .

1 2 ... a6 1 3.b3

1 3 ... e6?! Better was 1 3 . . . tl'id7 1 4.0-0 .l:Ib8 l S .aS c8 1 6 .xa6 xa6= and Black seems to have normal Benko Gambit play for the pawn here.

14.gS 1 4.dxe6 fxe6 1 S .xa6 tl'ixa6 1 6 .0-0 tl'ib4 1 7 .gs;!;. Since both sides have weak pawns here, it's hard to argue that Black has enough compensation for his missing pawn.

C h apter 1 : H i s t o ry of the f3 An t i - Gri i n fe l d

14 ... d7?! Probably both players missed the trick 1 4 . . . exd5 1 5 .tllxd5 tllcxd5 1 6 .xd5 b6 1 7 .xa8 b4+ , with which Black could have equalized.

1 5.0-0 1 5 .dxe6 tllxe6 1 6 .c 1 t .

1 5 ... exd5 1 6.exd5 fe8 1 7.d2 'fk'f5 1 8 . .l:!.a2 b7 1 9.tllg3 e5 20J1d1

20 ... tllg4?? After 20 . . . h6 2 l .f4 d4+ 22 .'fk'xd4 cxd4 2 3 .tllb5 tllxb5 24.xb5 l:!.ec8 2 5 .tlle2 tllxd5 2 6 .xd6 tllc3 2 7 .tllxc3 dxc3 28 .f4t White should retain an extra pawn.

21 .tllce4+- f5 22.f 4 22 .tllxc5 ! dxc5 23 .d6+ d5 24.fxg4 xc4 2 5 . bxc4+- .

22 ... d4+ 23.xd4 xd4+ 24.xd4 cxd4 25.tllxdG

25 ... .l:!.ebS?

25 . . . l:re l + 26 .tllfl xd5 2 7 .fxg4 xc4 28 .tllxc4 tlld5 29 .g3 l:rb l 30 .gxf5 .l::!.xb3 3 l .fxg6 hxg6 3 2 .tllfd2+-.

26.tllxb7 .l:l.xb7 27.dG+ 1 -0

(D70) Game 1 .22 Speelman, Jonathan Gavrikov, Viktor London 1 985 (4)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tllxd5 5.e4 tllb6 6.e3 g7 7.tllc3 0-0 8.f4

8 .. .f5?! 8 . . . tllc6=; 8 . . . c5 ! ? .

9.b3+ e6 ! 1 0.e5 tllc6 1 1 .tllf3 a5 1 2.a3 tlle7 1 3.'it>f2 a4 14.'fk'a2 tlled5 1 5.tllxd5 xd5?!

l 5 . . . tllxd5 l 6 .c4 c6=. Probably White can't break through against this defense.

1 6.xd5 tllxd5 1 7.c4 d7 1 8.ac1 .l:i.fc8

45

S a b o t a g e t h e G rii n feld

1 9.h3 After 1 9 .xdS exdS 20 .d2;!; White has all the active play, but perhaps Black can defend successfully.

1 9 ... h6 20.g3 c6 21 .h4 fa 22.tllg5 .ld.ea 23.h5 gxh5 24.1:!.xh5 h6 25.tllf3;!; ;g;ada

26.e2 26 .tllh2 ! and the threat of g3 -g4 is hard to meet.

26 ... g7 2 6 . . . l:!.d7 ! . Black needs to combat the g3 -g4 plan.

27 . .l::ig 1 @f7 2a.g4 fxg4 29 . .l:.xg4 .J::rga

30.l:th1 With 30 .tllh4! White should prepare f4-f5 .

30 ... tl:le 7 31 .c1 .ifa 32.l:txga @xga 33.tllh4 g7 34.g4 @f7 35.l:tg1 d5?! 36.h5+ wfa 37.d2

46

Jonathan Speelman

The threat of 3 8 .b4 forces Black to surrender a pawn.

37 ... c5 3a.dxc5 e4 39.a5 d3 40.r:!.d1 tlld5 41 . .l:!.xd3 xd3 42.d2 'it>g8 43.ea i.fa 44.c6 c5+ 45.@f3

45 ... b5? 45 . . . bxc6 46 .xc6 d4 47 .xa4 xb2 48.d7 @fl 49 .a4 leaves White a solid pawn up, but the limited material gives Black some hope for drawing.

46.a5 @fa 47.d7 we7 4a.f5 exf5 49.tllxf5+ .ixf5 50.xf5

And Black resigned, as White wins a piece.

Chapter 1: H i s t o ry of t h e f3 An ti - G ri infeld

(D70) Game 1.23 Karasev, Vladimir Zdrojewski, Wojciech Leningrad 1 990

1 .c4 tt:Jf6 2.d4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tt:Jb6 6.tt:Jc3 g7 7.e3 0-0 8.'ifd2 e5 9.d5 c6 1 0.h4 cxd5 1 1 .exd5 ttJ8d7 1 2.h5 tt:Jf6 1 3.hxg6 fxg6 1 4.0-0-0 d7

See the theoretical section for this position.

15 .g4? 1 5 .b l ;l; ; 1 5 .d6;!;; 1 5 .tbh3;!; .

1 5 . . JkS After 1 5 . . . e4! 1 6 .g5 tbh5 1 7 .tbxe4 f5 1 8 .h4 tba4 1 9 .d4 l:rc8+ 20 .@b l 'ifxd5+ Black has better pawns and the initiative.

1 6.@b1

1 6 ... tt:Jc4 Better was the exchange sacrifice 1 6 .. Jhc3 ! 1 7 .'ifxc3 tbfxd5 1 8 .'ifd2 tbxe3 1 9 .'ifxe3 'iff6 20 .h2 'iff7 2 1 .d3 e6=. Black is down just 1/4 pawn and has the initiative and better piece placement.

1 7.xc4 l:i'.xc4 1 8.tt:Jge2;!; e4? 1 8 . . . b5 1 9 .tbg3 'ifc7 20 .hfl b4 2 1 .tbce4 tbxe4 22 .tbxe4t. White's strong knight outpost and passed pawn give him the edge.

1 9.g5 exf3 20.gxf6 'ifxf6 21 .tbd4 h5 22.J::rhf1

White has a knight for two pawns and the bishop pair, so one pawn plus by my count. Although Black has three connected passers, they cannot advance easily as the g-pawn cannot reach g4 to support the advanced f-pawn.

22 ... g4 23.'if d3 M.cc8 24.@a1 I!.ce8 25.d2 .l2:e5 26.tt:Jc2 'iff5 27.tt:Jd4 'iff6 28 . .!d.f2 fe8 29.tt:Jc2 'if5 30.d6 'iff7 31 .d4 :!'.5e6 32.xg7 'ifxg7 33.d5 @h7 34.tbd4 lle5 35.'ifc4

35 ... f6? More tenacious was 3 5 . . . b5 3 6 .'ifb4 'ifb7 3 7 .d7 xd7 3 8 . tbxf3 e7 3 9 .'ifd4 xf3 40.l::rxf3 :!:!e l 4 1 .l:!d3 . With a knight for two connected but not advanced passers, White should win but it is not yet certain.

36.'ifc7+ 3 6 .tbxf3 3 7 .'ifc7+ @h6 38 .d7+-.

36 . . . @h6 37.d7 l:!.d8 38.tt:Jxf3 .l2:f5 39.tt:Jd5 'iff8 40.'if c1 + g7

47

S a b o ta g e the G riinfe ld

41 .'tflc3+ @g8 42.tllc7 .l:i.xf3 43 . .l:i.xf3 xf3 44.tlle6

Black resigned, in view of 44 . . . VWe7 4 5 .tllxd8 xd l 46 .'tflc4+ @h7 47 .'tflf7+ .

(D70) Game 1.24 Ward, Christopher Liss, Eran Port Erin 1 99 7 (6)

1 .d4 tllf6 2.c4 g6 3.f3 d5 4.cxd5 tt:Jxd5 5.e4 tllb6 6.tllc3 g7 7.e3 0-0 8.'tfld2 tllc6 9.0-0-0 f5

1 0.h4 1 0 .eS ! tllb4 l 1 .tllh3 !t .

1 0 .. .fxe4 1 1 .h5 exf3? 1 1 . . . gxhS ! 1 2 .xhS fS=.

1 2.hxg6 hxg6 1 3.tllxf3

White's attacking prospects against the denuded black king are clearly worth more than a pawn.

1 3 ... 'id6 14.h6 f5?

48

1 4 . . . ld.xf3 1 S .gxf3 'tflxd4 l 6 .'tflxd4 xd4 was the lesser evil.

1 5.xg7 @xg7

1 6.d5 Also winning is l 6 .'ih6+ @f7 1 7 .bS tlldS 1 8 .c4 e6 1 9 . tlle S + tLlxeS 2 0 .dxeS 'tfle7 2 l .g4 .l::[h8 22 . 'iVh7 + ! +-.

16 ... ttJeS 1 7.'tflh6+ @f6 1 8.'tflh4+ @g7 1 9.'tflh7+ @f6 20.'iVh4+ @g7 21 .'tflh6+ @f6 22.'iYgS+ @g7 23.ttJxeS 'ixe5 24.g4+-

24 ... 'tflf6 25.'tflh6+ @f7 26.'tflh7+ @es 27.gxf5 gxf5 28.b5+ @d8 29 . .tthf1 a6 30.d3 f4 31 .'tfle4 tlld7 32.'iYe6 tllc5 33.'tflxf6 J::!'.xf6 34.tlle4 tt:Jxe4 35.xe4 @d7 36.@d2 @d6 37.@d3 .i::l:af8 38.f3 .l:i.f5 39.de1 a'.98 40.l::!'.g1 xg1 41 .xg1 .l:!.e5 42 . .l:!.g6+ @c5 43 . .l:!.e6 1-0

Chapter 2

Third Move Offshoots 1 .d4 tt:Jf6 2.c4 g6 3.f3

Aside from the Neo-Griinfeld 3 . . . d5 and the King's Indian 3 . . . g6 (or 3 . . . d6, which should transpose) , Black has several interesting tries on move three, some of which are serious options adopted by elite players. We'll explore them in this chapter.

First we try the gambit move 3 . . . e5 , attributed to Adorjan and adopted by Leko. It aims at the elementary tactic 4.dxe5 tt::lh5 5 .e4? h4+ followed by . . . tt::lg3 , winning the exchange. Of course White need not cooperate, with 5 .tt::lh3 apparently being the best way to avoid this trick, in which case Black retains some but not full compensation for the pawn. See Game 2 . 1 . The related 3 . . . tt::lh5 is also covered there; it is no better.

Next we look at 3 . . . tt::lc6 , my recommendation for Black in KRBW It is quite logical to attack d4 since White has played f2-f3 , which strengthens the light squares but weakens the dark ones. Nevertheless White gains time kicking the knight around, and it seems that the line I recommended in KRBW is not quite equal for Black. Whether White should develop his knight to f3 or h3 (after playing f3 -f4) is a tough choice; I currently lean towards f3 . See Game 2 . 2 . This line remains quite playable for Black, if not fully equal.

Now we come to the latest try, Vachier-Lagrave's 3 . . . e6 (although it was actually first played in 1 934! ! ) . Black aims for . . . d7-d5 intending to take back with the knight only if White plays 4.tt::lc3 , and otherwise with the pawn. Rather a clever idea, I would say! 4.e4 d5 looks best, then White can choose between 5 .e5 tt::lh5 6 .e3 ! (not 6 .f4?, which is also examined in Game 2 . 3 ) , or 5 .cxd5 exd5 6 .tt::lc3 , Svidler's recent choice in Game 2 .4. Both should give White a pull, but I currently

49

S a b o t a g e the Griin f e ld

favor Svidler's line. Probably we'll see more games with 3 . . . e6 , i t doesn't look too bad.

Finally we examine the Benoni move 3 . . . cS . This seems to be just a transposition to the King's Indian Sarni.sch, and indeed after 4.dS g7 S .e4 d6 6.tlJbc3 0-0 7 .tt:Jge2 we reach the same position as in the 'Sarni.sch with . . . c7-cS ' chapter after 3 . . . g7 4.e4 d6 S .tlJc3 0-0 6 .tlJge2 cS 7.dS. However there are some important subtleties here. With the King's Indian move order, White can choose 7 .e3 , which 3 . . . cS avoids. On the other hand, when Black commits to . . . c7 -cS on move three, White can try to do without tt:Jge2 and play .tgs instead, which is considered a rather good line against . . . c7-cS but slightly dubious when Black can play . . . a7 -a6 before . . . c7 -cS , so as to meet d4-dS by . . . b7-bS a la Benko. This gS line is examined in Game 2 . S .

Furthermore, White can opt for bringing the king's knight to c 3 instead of the queen's knight, as White did in the World Championship game Anand-Gelfand (Game 2 . 6) . I think this is rather logical, because the e2-knight is something of a problem piece for White in the Samisch. My conclusion is that while 3 . . . cS is quite playable, on balance I think it's objectively better just to play the King's Indian with 6 . . . cS . However since the Samisch without d4-dS may be a bit drawish, I can recommend 3 . . . cs to the player who very much wants to avoid a draw.

In Game 2 . 7 we look at 4 . . . bS S .cxbS a6 , which is a version of the Benko Gambit. I think White's chances are better than in the Benko proper, if White plays as shown in Game 2 . 7 . White plays the same plan as might have occurred in Game 2 . 6 , namely posting the knights on a3 and (via e2) c3 . This seems to deprive Black of much of his typical Benko counterplay. Consequently this cousin of the Benko Gambit is rarely seen or recommended.

(E60) Game 2 . 1 Grover, Sahar Grandelius, Nils Chennai Wch-jr 2 0 1 1 ( 1 0 . 7 )

1 .d4 tt:Jf6 2.c4 g6 3.f3 e5?!

This logical but somewhat dubious gambit is credited to Adorjan. Leko beat Kramnik once with it.

s o

A similar idea is 3 . . . tt:JhS 4.e4 eS S .e3 exd4 6 .xd4 ( 6 .xd4 g7 7 .!#_xg7 tbxg7 8 .tlJc3 0-0 9 .d2 d6 I 0 . 0-0-0 tbc6 1 1 .'it>b I ;\; +0 . 29) 6 . . . .tg7 7 .d2 h4+ 8 . .tf2 f4 9 .Vi'c2;!;.

Analysis diagram

Black's queen and knight are rather misplaced and will lose time retreating , for example 9 . . . tt:Ja6 I O .tt:Jc3 +0 . 3 7 .

4.dxe5 tt:Jh5 5.tt:Jh3! ,.n Never overlook the option of ltJh3 when it's safe and the

f-pawn is on f3 . A) 5 .f4?! h4+ 6 .g3 ltJxg3 7 .ltJf3

hS 8 . .l:rg l lLixfl 9 . .l:rxfl = g7 0 .00 ; B) After S .g3 ltJc6 6.f4 d6 7 . .ig2

White may be marginally better ( +0. 1 S ) . 5 ... tt:Jc6

6 . .ig5 A good quiet positional continuation is 6 .lLlc3 lLixeS 7 .ttJf2 .icS 8 .e3 d6 9 . .ie2 ttJf6 1 0 .0-0 as l l .b3 ( 1 1 .e4 +0 .44) 1 1 . . . 0-0 1 2 .d2 d7 1 3 . .ib2;!;.

6 ... e7 7.ii.xe7 xe7 8.tt:Jc3 8 .g4? ! is recommended by Svetushkin.

Analysis diagram

But it seems he overlooked (or underestimated) the following novelty sacrifice: 8 . . . lLixeS ! 9 .gxhS d6 ! 1 0 .f4 xh3 1 1 .fxeS h4+ 1 2 .Wd2 .ifs 1 3 .e3 dxeSg? -0 . 76 . Black has development,

Chapter 2 : T h ir d Move O ffs hoots

an attack, the better pawn structure, and probably two pawns for a knight.

8 ... xe5 A) In case of 8 . . . 0-0 9 .f4 d8 (9 . . . d6

1 0 .exd6 cxd6 1 1 .lLldS d8 1 2 .g3;!;) 1 0 .g3 d6 l l .exd6 .ixh3 1 2 .xh3 lLixf4 1 3 . .ig4 tt:Je6 1 4.ds xd6 1 S .xd6 cxd6 1 6. 0-0-0 .l:rad8 1 7 .f3;!; White has pressure on the weak d6-pawn;

B) Not 8 . . . tt:Jxe S ? 9 .ltJdS d8 1 O.d4 d6 l l .f4 c6 l 2 .ltJc3 .ixh3 1 3 .fxeS cS 1 4.e3 lii.e6 l S .0-0-0 0-0 1 6 .exd6 .ixc4 1 7 .Wb l e6 1 8 .xcS .l:rc8 1 9 . xa 7 and Black does not have enough for two pawns.

9.'iYd2 This and Kramnik's 9 .g4 are both good. I would opt for the game move to avoid having to gambit a pawn in the note to 9 .g4 ltJf6 . If 9 .g4 tLlg7 (9 . . . ltJf6 1 0 .gS lLlhS l l .d2;!;; now l l . . . d6 leads back to the game, but 1 1 . . . 'iYd4 1 2 .0-0-0 xc4 1 3 .e4 (+0 . 3 7) at least forces White to sac a pawn to get a good position) 1 0 .f4 and now:

A) 1 O . . . e7 ? 1 1 .lLldS d8 1 2 .d3 (Adorjan and Vegh like 1 2 .ttJf2 0-0 1 3 . .ig2 (+0 . 7 2 ) , which is also fine) 1 2 . . . 0-0 1 3 .c3 lLie8.

Analysis diagram

This was Kramnik-Leko, Tilburg 1 998 . After l 4 . .ig2 White's huge space and development advantages are obvious

5 1

S a b o ta 9 e t h e G rii n feld

(+0 .80) . Amazingly, Black won the game ! ; B ) 1 O . . . Wk'as 1 1 .Wk'd2;!; +0 . S2 ; C ) Svetushkin recommends 1 0 . . . Wk' cS

for Black, giving only the unclear 1 1 .'l?HdS . White should gambit the c-pawn here: 1 1 .Wk'd2 ! 0-0 1 2 .tlldS 'l?Hc4 1 3 .tllc7 nb8 1 4.tlldS . White has regained his pawn and has a clear edge due to his better pawn structure and the powerful knight outpost ( +0 .8S ) .

9 ... d6 1 0.g4 tllf6 1 1 .g5

E .i. E l i l . , . ,

i 8

.8 . .8.tlJ 88 8 8 n n

1 1 ... tllh5?! 1 I . . .tlld7 1 2 .tt:Jds t>d8 1 3 .dl tllb6 1 4.b3 tllxdS l S .cxdS tlle7 1 6 .f4 +0 .4 1 . White's space advantage together with Black's loss of castling rights define White's advantage.

1 2.tlld5 0-0?! 1 2 . . .@d8 +0 .48 .

1 3 .f4 Wk'e4?! 1 3 . . . Wk'd4 1 4.e3 +0 .80 . c7 falls.

14.tllf2 ( + 1 . 2 7 ) 14 ... Wk'xc4 1 5.e3 'l?Ha4

S 2

1 6.1le2?! 1 6 .b3 ! Wk'a3 1 7 .tllxc7 .ld.b8 1 8 .1le2 tllg7 1 9 .tlle4+- ; White wins a pawn while keeping the superior position.

1 6 ... e6 1 7.b3 Wk'a5 1 8.'i:Vxa5 tllxa5 1 9.tllxc7 .ld.ac8 20.tllb5 d5 21 .flxh5 gxh5 22.tlld4?

22 .@d2;!;. 22 ... tllc6= 23.@d2 tllxd4 24.exd4 ilf5 25.l::rhc1 h6 26.h4 hxg5 27.fxg5 f6 28.gxf6 .l::!.xc1 29.l::rxc1 nxf6 30.l:rg1 + @f7 31 .@e3 l!e6+ 32.@f4 1lg6 33.@f3 1/2-112

(E60) Game 2.2 Ganguly, Surya Shekhar Gupta, Abhijeet Kavala 20 1 2 ( 6) 1 .d4 tllf6 2.c4 g6 3.f3 tllc6

This was my choice for Black in KRBW; it makes f2-f3 look a bit silly.

4.d5 4.tllc3 is probably the second best move: 4 . . . dS . So we get a Grunfeld with each side having made a questionable move (3 .f3 and 3 . . . tllc6) . S .cxdS tllxdS 6 .e4 tllxc3 7. bxc3

Analysis diagram

7 . . . 1lg7 (7 . . . es 8 .ilbs 1ld7 9 .tlle2 g7 1 O .e3 0-0 1 1 .dS;!; +0 . 1 8 . This is similar to the 7 . . . g7 line) 8 .h4 hS 9 .d3 eS 1 0 .dS tllb8 1 I .nb 1 tlld7

1 2 .tlle2 0-0 1 3 .e3 f6 1 4.g3 e7 1 5 .0-0

Analysis diagram

+0. 1 8 . White has better development and more space for free.

4 ... tt:Je5 5.e4 d6 6.tllc3 g7 6 . . . c6 7 .f4 tlled7 8 .tllh3 cxdS 9 .cxdS 'lWb6 1 O .tllf2 g7 l l . .te2 0-0 1 2 .0-0 tbcS 1 3 .c2;!; +0 . 3 7 (Svetushkin) .

7.f 4 7 . .te2N may be the best move, postponing the choice between tllf3 and tllh3 for a move to avoid the possibility of Black playing . . . tbcS before castling. 7 . . . c6 (7 . . . 0 -0 8 .f4 tlled7 9 .tllf3 transposes to the next note, while the voluntary retreat 7 . . . tlled7 is met by 8 .tllh3 ) 8 . f4 tlled7 9 .tllh3 cxdS 1 O .cxdS 0-0 l I .tllf2 bS 1 2 .0 -0 b4 l 3 .tlla4 as 1 4.a3 tllb6 1 5 .axb4 xb4 1 6 .d2 d4 1 7 .c3 xd l 1 8 .fxd l ;!; .

Chapter 2 : Third Move O ffs h o o ts

Analysis diagram

White's space advantage gives him the edge, as there are still many pieces on the board. +0 . 29 .

Surya Shekhar Ganguly

7 ... tt:Jed7

8.tllh3

5 3

S a b o t a g e the G riinfe ld

8 .tL'if3 and now: A) 8 . . . 0-0 9 .e2 (Svetushkin men

tions only the inferior 9 .d3 as leading to equality) 9 . . . tLlcS 1 0 . e S tLife4 ( 1 0 . . . ttJfd7 l 1 .e3t) l l .tL'ixe4 tLixe4 1 2 .e3 c6 1 3 .0-0 fS 1 4.iYb3 'ic7 1 S .ad l llfd8 l 6 .d4t + 0 . 2 2 . White's space advantage is obvious, while Black's compensation for it, the bishop pair, may disappear shortly;

B) 8 . . . c6 9 .e3 iYaS 1 0 .iYd2 tLlcS 1 1 .l:Id l t +0 .26 ; C) 8 . . . tLlcS ! ? 9 . eS +0 . 22 .

8 . . . 0-0 A) 8 . . . eS (or 8 . . . e6 9 .dxe6 transposing.

I failed to capture in a similar position against Shabalov recently and I lost the game) 9 .dxe6 fxe6 1 O .e2 (Svetushkin thinks 1 O .d3 to prepare f4-fS is even better, but anyway Black can avoid this by waiting for ite2 before playing . . . el -eS) 1 0 . . . 0-0 1 1 . 0 -0 b6 1 2 .e3 itb7 ( 1 2 . . . tLlcS 1 3 .iYc2 aS 1 4.f3 1:!'.b8 l S .i::rae l tLlfd7 1 6.tLlgS b7 l 7 .b3 'ie7 1 8 .a3 eS 1 9 .fS h6 20 .tLlh3 (20.b4!t) 20 . . . gxfS 2 1 .exfS e4 22 .itxcS itxc3 23 .'ixc3 bxcS was Grischuk-Navara, Rhodes 20 1 3 . Now White played 24.tL'ifl and eventually won, but 24.hS ! +0 .70 was much stronger) 1 3 .itf3 l:Ib8 1 4.iYc2 'ie7 l S . .l:Iae l t

Analysis diagram

+0.44;

54

B) 8 . . . c6 9 .e2 (Svetushkin prefers 9 .e3 but 9 . . . iV aS 1 0 . tLlf2 tLlcS l l .itd2 seems an unnecessary concession although White is still for choice) 9 . . . 0-0 1 0 .tL'if2 cxdS 1 1 .cxdS bS ! ? 1 2 .0-0 (Svetushkin considers only accepting the gambit, which he prefers to avoid by 9 .ite3 . But it seems that 1 1 . . . bS can just be ignored) 1 2 . . . b4 1 3 .tLla4 iYaS 1 4.a3 tL'ib6 1 S .axb4 'ixb4 1 6 .d2 iYd4 1 7 .c3 iYxdl 1 8 . .!d:fxd l t

Analysis diagram

+0 .29 . White has an advantage in piece placement and space.

9.e2 ttJcS A) 9 . . . l:re8 1 0 .e3 e6 l l .dxe6 xe6

l 2 .tL'if2t +0 .29 . White's space advantage is significant;

B) 9 . . . e6 Don't fail to capture on e6 when

legal in this line. 1 O .dxe6 fxe6 1 1 .0-0t; see the note to Black's eighth move;

C) 9 . . . c6 l O . tLlf2 ( 1 0 .ite3 cxdS 1 1 .cxdS bS 1 2 .a3 a6 +0. 1 0 , WallaceGormally, 4NCL 20 1 3 / 1 4) 1 0 . . . iYb6 1 1 .0-0 as 1 2 .iYc2 tLics 1 3 .e3 'ib4 1 4.eS (the surprising 1 4.f3 is also good) 1 4 . . . dxeS l S .fxeS tLlfd7 1 6 .e6 fxe6 1 7 .tL'ig4 exdS 1 8 . cxdS Wh8 1 9 . .i::!.ad 1 t. White's large advantage in development is more significant than Black's extra pawn.

Analysis