TWIC is 30. First issue 17th September 1994.
John Watson Book Review (77)
Good...Good...Good...Disastrous
IM John Watson - Sunday 2nd September 2007
At some point this column will settle into a once-per-month format. For the time being, I have so many 2006 books to catch up with that I will be dashing through them, mixed with 2007 releases and a few neglected publications from earlier years. With luck, the next column will follow fairly soon and finish off the openings category for a while.
In this column we look first look at a book with a fresh approach to unusual subject. The result is a work that thrives under the writer's enthusiastic consideration. Next, I turn my attention to two excellent DVDs/videos dealing with openings in very different ways.
To conclude, I'll discuss a book that I am frankly upset with. I'm generally going to stick with the positive in these reviews, and will usually insert my criticisms of a product within the context of overall approval. But sometimes a book will have egregious qualities that in themselves deserve rebuke. Hopefully, strong criticism will not only provide readers with examples of what to avoid, but nudge the authors in a better direction for future projects. Since writing a chess book is a very difficult thing to do (or rather, to do well), it's unfair to engage in radical censure without providing specific evidence. Thus you will find a long and sometimes technical discussion below, but the gist of the problems can be absorbed in a casual reading.
Tiger’s Modern; Tiger Hillarp-Persson; 216 pages; Quality Chessbooks 2005
Alexei Shirov: My Best Games in the Spanish ; Alexei Shirov; 4.5+ hours; ChessBase DVD 2006
The Scandinavian The Easy Way; Andrew Martin; ChessBase DVD 2005
Chess Openings for White, Explained; Lev Alburt, Roman Dzindzichashvili & Eugene Perelshteyn; 548 pages; CIRC 2006
Tiger’s Modern is an opening book that may well appeal to those not normally attracted to opening theory. Named after the author Tiger Hillarp Persson, the work investigates a version of the Modern Defence deriving from the moves 1...g6, 2...Bg7, and 3...d6, and (versus most traditional setups) 4...a6. Black's standard follow-up, when appropriate, is ...b5, ...Bb7, and ...Nd7. This plan is implemented in most variations where White plays e4 and d4 without c4. Hillarp Persson notes right away that 'the first great advantage of the Modern with ...a6 is that it gives White almost zero chances of playing for a draw and therefore it is an excellent choice in situations where playing for a win is essential.' The book is organized upon 69 games with a large representation of Tiger's own battles. Personal experiences abound.
Like every other author on the planet, Tiger claims that his is 'a book based on ideas more than lines', yet he is quite serious about covering all of the most obvious variations in considerable detail and with a regard for exact move orders when called for. His is easily the most in-depth treatment of the subject, although I wish that he'd engaged in a more thorough research of the literature. In a brief comment upon the history of the variation, for example, Tiger mentions games by Ivkov and Utjelky in the 1960s, followed by Suttles, Keene, and others into the 1990s. It's hard to know in what fashion he did his research; only 3 books are listed in the bibliography. In fact, Ivkov played variations with ...a6 in modern style over many years, and quite a few of his games appear in Keene and Botterill's seminal work on the Modern Defence (1972). The latter book also cites practice in lines such as 3 Nc3 d6 (3...a6!? was also played) 4 Be3 a6, 4 Nf3 a6, and 4 f4 a6. Unfortunately, very few of these games seem to have made it in MegaBase (nor into Hillarp-Persson's book) and it makes one wonder how much pre-computer data and research is disappearing, and at what rate. This is merely a lament; obviously I can't blame Tiger for not having this dinosaur on his shelf!
The other main source that I have for ...a6 ideas is Speelman and McDonald's 2000 book Modern Defence, which is listed in the Bibliography. That work applauds some move orders that Tiger condemns and raises other issues. It would have been nice to see these addressed, but again I'm not sure how often this book was referenced. For example, Speelman & McDonald give the high-profile ...a6 games Anand- Svidler, Linares 1998 and Rublevsky-Ibragimov, Elista 1998, neither of which I see cited in Hillarp-Persson, presumably because they don't have the exact fit into his repertoire. Actually, since the book is organised by games and not by variations, nothing of this sort be said with total confidence, but there are indices of games and variations, and I believe that I've looked at all plausible hiding places. In a book of ideas, the instructive courses of those two games are worthy of inclusion, especially since ...g6/...a6 is not well-represented in the games of top grandmasters. But Tiger's understanding of the subtleties is clear in the relevant variations. After the continuation 3 Nc3 d6 4 Nf3 a6 5 Be2, the Rublevsky game went 5...Nd7!? 6 0-0 b5, transposing to Tiger's rather lengthy treatment of 5...b5 6 0-0 Nd7. At this point Rublevsky played 7 a4 ('!' Speelman & McDonald; the result is not entirely clear). In his book Tiger uses two games with 7 Re1 for his analysis; this seems to be a main line, since games by several other strong players are quoted in Tiger's notes. But instead of those two moves, he also suggests the unplayed 7 d5('!'), with the simple idea of Nd4, when White has the better of it. That is true, and a fine illustration of what a perceptive author can do that a database search can't.
Many top players employ 3 c4 against the Modern, as I have, and the author grudgingly accedes to the idea that the King's Indian Defence (3...d6 4 Nc3 Nf6) may be best in that case. His other recommendation is 4...e5, which he covers well, allowing the sequences such as 5 Nf3 Nc6 6 d5 Nce7 Be2 are difficult. For example, a fascinating line goes 7...h6 8 h4 f5?! 9 h5 g5 10 g4!! from Jovanic-Jurkovic, Pula 2004, when White succeeds in closing the kingside:
Naturally, Tiger offers improvements here. In the same variation (3 c4 d6 4 Nc3 e5), he does gives short shrift to the variation 5 d5 f5 6 exf5 gxf5, stopping after 7 Qh5+ Kf8 and claiming good chances for Black after ...Nf6 followed by moves such as ...Na6 and ...Qe8-g6. This position has been debated for many years and there have been numerous games. It would have been good to go further with the analysis and perhaps feature a game, because back in my youth when 5...f5 was popular, most players thought that White got an edge. The first player can also forego 7 Qh5+ and try to exploit the looseness of Black's position with moves such as g3/Bh3 and Qf3. Of course, Black may stand quite satisfactorily, but that should be shown. When the Modern was fashionable and I joined the craze for 1...g6, 5 d5 gave me fits. There are usually ways to work around any problem, and if need be Black can play with a small disadvantage in a double-edged position.
As is my wont, I'm straying from the central points. This is a thorough and very thoughtful book. It is also easy to understand specific strategies, in part because we find extensive explanation of ideas and maoeuvres at the beginning of each chapter. Hillarp-Persson gives new analysis and improvements in nearly every line. The latter is perhaps not surprising given the fact that he is one of the few grandmasters consistently playing this ...a6 variant of the Modern. Unlike many grandmaster authors, he expends a great deal of energy poring over details that would probably not occur to the average reader. I'd be surprised if White could achieve more than a small advantage against the suggested Black repertoire, which features elastic development and chances for counterattack. A word of warning, however: the Modern is difficult to play without considerable chess understanding, so you might lose a few games while you are getting the hang of it.
In conclusion, Tiger's Modern is a first-rate book by an original thinker. If you're looking for something a little eccentric and don't mind taking on an initially cramped position, you might consider adopting the 1...g6, if only as a second defence. The Modern is a 'one-size-fits-all' solution for Black, yet it avoids the oppressive boredom of similar all-purpose White openings such as the King's Indian Attack or Colle System.
Over the next few columns, I will be reviewing and recommending several of the ChessBase 'FritzTrainer Opening' DVDs, which are both innovative and easy to learn from. Before doing so, however, I have to make a general complaint that applies to most (but not all) of them. The majority of these DVDs don't come with a database, not even of the games that the lectures themselves are organised around. Although you might discover that you can save the games one-by-one into a database as they're played on the screen (a procedure that is not immediately obvious), the resulting game notes would be incomplete and wouldn't include the verbal commentary that is the essence of making this a video lecture rather than a book. This situation leads to the awkward possibility that the viewer may have to go back and watch a section of the video again each time that he wants to remember what the presenter said. ChessBase is constantly improving their products and offering more features; here, however, they seem to have missed the boat.
Having said that, this hardly ruins the product, and I very much enjoy the recently-released series of Alexei Shirov's DVDs on openings. In this column I'll mainly be discussing his My Best Games in the Spanish, but can also recommend My Best Games in the Sicilian, My Best Games in the Sicilian-Najdof, and My Best Games in the Petroff Defence. Like other ChessBase 'Fritztrainer Opening' DVDs (what is it with these ChessBase titles?), this is a very lengthy multimedia presentation with Shirov speaking on the screen and also poised in front of his computer while playing over his games. The latter are shown 'live' on a split screen while Shirov comments upon them. The lectures are still more edifying because Shirov plays both the White and Black sides of the Spanish (= Ruy Lopez), as he does the other openings in the series.
Let me briefly detour and talk about My Best Games in the Sicilian. The Sicilian leads to the kind of games that an attacking player such as Shirov would seem to revel in. In the Sicilian DVDs, he presents great games and an introduction to the theory of the Sicilian, which any Sicilian player indeed can learn a lot from. Shirov does tend to be almost apologetic about where he or his opponent technically went wrong in the middle of complications, with the result that the overall game can lose its dramatic effect in the midst of analytical details. Thus we have more of a best games collection (which is fine; there's a lot of excellent material!), but without a compelling narrative. In other words, the games are presented objectively, as they are in most modern games collections. Amusingly, Shirov ocassionally forgets move orders! There are some nice moments that are typical of this series. 'I cannot call myself an expert in the Sicilian', he says, 'There are lines that I know quite well...but theory is always changing my assessment, and I wouldn't dare to say "Okay, White should do this or White should do that"...(sighing).' Of course, then we get thorough lessons in the given variations. Shirov also presents some great endgame play (atypical of these DVDs), and jokes around about a 25-move forced mate that a computer comes up with in one game. There are also insights: At one point he says 'It's better to have a passed pawn with the two bishops than to have two passed pawns' (with even material, of course). When describing one newer subtlety, he says 'That was 1977. As we know, chess perception was not on the same level as it is now'. Now it's time for me to sigh: alas, 1977 seems like just yesterday.
I like Shirov's Spanish DVD the best of these videos, although at 4-5 hours apiece I haven't completely finished watching any of them. In My Best Games in the Spanish, he presents a nice selection of games and variations with different characters, thus allowing you to get a feel for the positional and tactical issues. He begins (quoting loosely) 'Is it possible to play the Ruy Lopez with just strategical ideas? Probably not. Is it possible to memorize all the lines? Probably not.' This sets the tone for a presentation that is cognizant of detail while not being slave to it. The material is extensive, and covers more than four hours. He splits his material into the Steinitz Variation, the Berlin Defence, the Open Variation, , the ...Bc5 systems, Marshall Attack and Anti-Marshall (with 8 h3), the Chigorin Defence (the latest versions), and and finally, the dynamic Zaitsev Variation. Quite a list!
One's first impression is of an almost too-shy narrator whose tempo and delivery is a bit awkward. But he is very personable and quickly grows on you. Shirov interlaces anecdotes with his commentary, adding personal bits. 'As I am a sentimental person, I often think on my past', he says, showing his first game, at the end of which he admits that it is 'not very useful', because nothing has changed in the basic ideas or theory! 'So', he concludes, 'I should concentrate ... on setting [the games] up by openings and themes.' One advantage of the DVD is that the analysis often becomes detailed about the opening itself (rather than just the game). For example, there is a great deal of theory, including some little-known ideas, about the ..Bc5 variations, specifically, 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 b5 5 Bb3 Bc5, which Shirov has promoted. In this context, I should repeat my technical complaint that there is no database. To my mind, however, the virtues of the DVD easily outweigh this drawback, which may be irrelevant to many viewers anyway.
I won't go into the chess details, but Shirov's would be a great way to get used to the ideas and selected theory of the Spanish Game as both White and Black; you certainly get a lot of viewing time for your money, including commentary on opponents and how he approaches preparation. The games are exceptionally interesting. And when all is said and done, it's fun to see and listen to one of the most important players of our time. I very much recommend this DVD and others in this series.
In Andrew Martin's Fritz Trainer Opening DVD called The Scandinavian The Easy Way, Martin promotes the move 3...Qd6 after 1 e4 d5 2 exd5 Qxd5 3 Nc3. He doesn't give it a name, but I think that 'Melts Variation' is a fair stab in view of Michael Melts' lengthy treatise on the subject a few years back (reviewed in this column). The main line is 4 d4 Nf6 5 Nf3 a6, which has enjoyed some popularity over the last 5 years or so (the DVD was made in 2005), although it still hasn't had enough success to attract much interest at the very highest levels, accounting for its relative obscurity. Incidentally, I think that the move 5...c6 is underrated here, since Black's queen often comes to d6 after 3...Qa5 and 3...Qd8 anyway. Martin is a real expert on the Scandinavian (he prefers the name 'Centre Counter'), having written two books about it, concentrating upon 3...Qa5 line. This is one of the best videos that I have seen from him, in part because he fleshes out a complete repertoire including a close look at the most important subvariations and less popular responses. This degree of creative coverage is not reached in some of his DVDs, although there are others that I like and will hopefully get around to discussing.
In these 3...Qd6 lines he has several original ideas, and the solutions he chooses in response to the most challenging White approaches are often not the main lines. Martin has always had a sharp idea for the slightly offbeat, and you feel that some of what he is telling you he just happened upon in his research. The resulting remedies should therefore be less familiar to your opponents. On top of that, the Scandinavian is still a bit of a mystery to your average 1 e4 player.
I'm always interested in what advocates of the Scandinavian will say about the 1 e4 d5 2 exd5 Qxd5 3 Nf3 variation, skipping Nc3 temporarily in favour of a potential d4 and c4. Martin suggests the unusual move 3...g6!?, as played by Epishin, intending 4 Nc3 (now a d4/c4 setup leaves d4 a little exposed, although it's still an important idea) 4...Qd6 and simply ...Bg7, ...Nf6, and ...0-0. It's hard to make progress against this solid setup.
Throughout the video Martin is focused and engaged; he has a firm and precise speaking style that allows of no misunderstanding. Although there is little if any humour in the lectures, his intelligent commentary ensures that things will not be dull for the serious student.
While on the subject of the Scandinavian in electronic format, I should note that, Curt Hansen's 2002 'Scandinavian' ChessBase CD (not a DVD/video) is still an excellent source of ideas in that opening, especially for someone who is looking for 3...Qa5 material. For one thing, it includes a nice database of both annotated and unannotated games. Presumably a 1 e4 d5 player might want both products, but naturally he would need to be interested in 3...Qd6 in order to get the most value out of Martin's disc. As in the last column, we get very different advantages. The DVD, apart from being more recent, offers ease, a visual mode of learning, and a sharply focused repertoire. The CD analyses a much broader set of material with many more details and annotations.
Over the years, I have occasionally come across one of Roman Dzindzichashvili's openings videos, and if it dealt with some opening that I was interested in I'd look it over for new ideas. Sometimes the recommended lines were initially attractive, but soon I figured out what was wrong, usually just a good move or two not mentioned at the appropriate juncture. I never found anything usable, figured that the tapes were something for inexperienced players to goof around with, and didn't pay much attention thereafter. A couple of times a student took up one of the systems and I'd go through pretty much the same experience.
Because of good reviews, I bought my own copy of Chess Openings for White, Explained (generally I receive review copies, and indeed, CIRC had generously sent me the first volume for Black). The book offers a 'complete' repertoire for White based upon the move 1 e4. The authors are listed as GM Lev Alburt, GM Roman Dzindzihashvili, and IM (now GM) Eugene Perelshteyn; but I suspect that most of the concrete work was done by Dzindzihashvili and Perelshteyn, (apologies to Alburt if I'm wrong, although he may consider that a good thing). I'll refer to the them collectively as 'DPA'.
Most of the first chapter is essentially a marketing spiel, and in some ways a promotion of same flavour as advertisements for Dzindzihashvili's videos mentioned above. The reader is basically offered the universe with dressing on top. Using the kind of hyperbole normally limited to publishers' blurbs on back covers, we find that 'This volume contains every secret the authors have compiled over decades of research in the recommended openings. No theoretical novelty (TN) will be withheld from you. Such information is normally revealed only to world championship contenders, who hire top theoreticians, like Roman Dzindzichashvili, for that purpose.' Chess players are used to heavy-handed marketing, but something this crass is off-putting.
We are told that the book can even be used by 'Super GMs', and that Dzindzichashvili has been one of the prime movers and creators of modern opening theory for the past 40 years', with his 'advice and help... sought, with rewarding results, by such greats as Boris Spassky, Victor Korchnoi, Anatoly Karpov and Gata Kamsky.' To me, taken as a whole, the first chapter implies that at least a few of these brilliant ideas will be laid at our disposal, and nothing that the authors know will be hidden from us. Whether this is so you can decide for yourself as you read this review. Finally, 'We want this book to be your complete openings reference for your entire playing career.' [see also the end of this column].
Before delving into its depressing contents, I should say that the book is wonderfully laid out in terms of emphasising key ideas and variations in colour, and using large type with many diagrams (four to a page is typical), so that one can read the book without a board, albeit superficially. CIRC did something similar with the excellent book on the Pirc by Alburt and Chernin.
Anyway, I decided to look at the major openings (the Sicilian and 1...e5) to see what solutions were provided. But as someone with experience in the French, I naturally went to that section first. Against the Winawer, I found their suggested system covered in no less than 35 pages of analysis:
1 e4 e6 2 d4 d5 3 Nc3 Bb4 4 e5 c5 5 Bd2
This variation is generally not feared in Winawer circles or used by many masters, but I particularly wanted to see the lines after
5...Nh6
, a move with which I've had great success, and which has done well since its introduction into grandmaster play. DPA recommend the line
6 Nb5 Bxd2+ 7 Qxd2 0–0 8 f4
Here I think that 8...a6 is a good move - see below. But there's also nothing wrong with
8...Nc6 9 Nf3
, and at this point the authors say merely 'Compare 5...Ne7' and stop! What a strange way to avoid the issue: it takes almost no thought to see how different the two lines are. A knight on h6 defends against the tactics involving Bxh7+ as well as the move Nf7 that DPA feature in the lines after 5...Ne7. Furthermore, Black's knight can help attack the centre from f7 after he plays ...f6, and ...Nh6 also leaves Black's queen in contact with the kingside. Finally, the knight can (and does) go to g4 with great effect when a pawn on d4 anchors e3 for occupation via ...Ne3 and/or when Black's queen on b6 supports that move. A knight on e7, on the other hand, has its own advantages in that it can go to c6 (often with tempo), or to g6, or c8 (from which square it can challenge a knight on d6).
Did the authors simply not understand this? Well, here's a game to answer that question:
9...f6!?
The authors look past another obvious and natural move, namely, 9...a6!: 10 Nd6 cxd4 (the less interesting 10...Nxd4 11 Nxd4 cxd4 12 Qxd4 f6 seems equal after 13 Be2 fxe5 14 Qxe5 Qa5+ 15 c3 Bd7) 11 Bd3 (DPA's main idea. Instead, 11 Nxd4 f6 12 0–0–0 fxe5 13 Nxc6 bxc6 14 fxe5 Ng4! is good for Black, and 11 0–0–0 f6 will probably transpose) 11...f6.
Here, in the 5...Ne7 line that DPA refer to, White plays 12 0–0 and Black responds 12...fxe5 (which is risky). But after 12 0-0 in the diagram, Black can play the strong move 12....Ng4! (even 12...fxe5!? 13 fxe5 Nxe5 – which DPA show fails by force with the knight on e7 – isn't clear here after 14 Nxe5 Qxd6 15 Rxf8+ Qxf8 16 Rf1 Qe7; this takes some analysis). After 12...Ng4!, for example, play might continue 13 Qe2 Ne3 14 Rf2 Qb6 15 b3 g6, when Black threatens to capture on e5, but then 16 exf6 Rxf6 17 Ne5 Qc5 is uninspiring for White, to say the least.
So already White's 5 Bd2/ 8 f4 strategy is looking very shaky!
10 c3 cxd4 11 cxd4 fxe5 12 fxe5 Nf5 13 Bd3 Qb6!?
An option is 13...Nh4!?, for example, 14 Nxh4 (14 Ng5 h6 15 Nh7 Rf7 16 Nd6 Rf4=) 14...Qxh4+ 15 g3 Qd8 16 0–0–0 Qb6.
14 0–0 Bd7!? 15 g4?
Best was 15 Bxf5! Rxf5 16 Nd6 Rff8, when White may have a little something after 17 Kh1 (17 Rf2 Be8 18 a4 a5 19 Raf1 Bh5) 17...h6 18 a4 a5 and ...Be8, but it isn't much, if anything.
15...Nfxd4! 16 Nfxd4 Nxe5 17 Kg2 Nxd3 18 Qxd3 e5
and Black was winning.
The game? Perelshteyn-Shaked, Bloomington 1997. .
Okay, that's embarrassing, and surely given this personal experience Perelshteyn shouldn't have hidden behind the 'see 5...Ne7' comment. Just as importantly, if one is to advocate this line for White and even superficially analyse it, one would have the obligation to justify playing into the most difficult positions which directly arise from it, especially right away on the ninth move!
But it gets worse: What about the simplest move after 8 f4:
8...a6? Then the line they recommend (and the only logical line) is
9 Nd6 cxd4 10 Nf3
Here, again DPA, say: 'See 5...Ne7' ! They don't even use the word 'compare' this time. You will see how consciously avoidant that is when you look at the following combination of known theory and some analysis:
A 10...f6
11 Nxc8
The alternatives all have drawbacks as well.
11...fxe5!
Not that 11...Qxc8!? is bad, but it's only about equal after 12 exf6 Rxf6 13 Nxd4 Nc6 14 0–0–0 Nxd4 15 Qxd4 Nf5 16 Qe5 Qc5. Black needn't be satisfied with that.
12 fxe5
12 Nxe5 doesn't help much after 12...Qxc8 13 Bd3 Nc6 14 0–0 (14 Nf3? Ng4) 14...Nxe5 15 fxe5 Qc7 16 Qe2 g6 17 Rae1 Rxf1+ 18 Rxf1 Rf8. The e-pawn is weak and White isn't getting his pawn back.
12...Qxc8 13 Nxd4?
Black's advantage is limited after 13 0–0–0 (or 13 Bd3 =+) 13...Nc6 14 Bd3, but he is still on top with 14...Nf5 (or 14...Nf7), for example, 15 Bxf5 (15 Rde1 Qd7 16 g4? Ne3 17 Nxd4 Nxg4) 15...Rxf5 16 Rhe1 Qc7 17 Nxd4 Nxd4 18 Qxd4 Raf8, etc.
13...Nc6
Also strong is 13...Ng4.
14 Nxc6 Qxc6 15 0–0–0 Ng4!
and with ideas of ...Nf2, ...Rf2, ...Nxe5, and ...Rac8, Black has a winning advantage. Obviously Black isn't winning after 10...f6, but White can't be happy, and not even the first moves are covered by the authors.
B 10...Nc6
This is a good response, for example,
11 Nxd4
11 Bd3?! f6! is a known line which works out badly after 12 0–0 Ng4!
11...f6 12 Nxc6
12 0–0–0 fxe5 13 Nxc6 bxc6 14 fxe5 transposes.
12 ...bxc6.
This position is not easy for White, e.g., 13 0–0–0 fxe5 14 fxe5 Ng4 (still safer is 14...c5, for example, 15 Be2 Nf5 16 Nxf5 Rxf5 17 Qc3 Qg5+ 18 Kb1 Rxe5) 15 Re1 Rb8! 16 h3 Qb6 17 c3 Rf2 and so forth.
Finally, in the variation with:
5...Ne7
[after 1 e4 e6 2 d4 d5 3 Nc3 Bb4 4 e5 c5 5 Bd2], DPA's main line goes
6 Nb5 Bxd2+ 7 Qxd2 0–0 8 f4 Nbc6 9 Nf3
Now Black can play simply by
9...a6
DPA also give 9...Qb6 10 dxc5 Qxc5 11 0–0–0, and now 11...a6 12 Nbd4. But they don't mention 11...f6, which is the natural move you will likely run up against, or the more complicated 11...Bd7!? (threatening ...Nxe5).
10 Nd6
10...Nxd4 11 Nxd4 cxd4 12 Bd3 Qb6 13 0–0 Bd7 14 Qf2 f5 15 c3 Nc8!?
15...Nc6 also looks logical.
16 Qxd4 Qxd4+ 17 cxd4 Nxd6 18 exd6
They say: 'The endgame is pleasant for White because, even if Black wins the d-pawn, the best Black can hope for is a draw, since White's d4- and f4-pawns hold Black's d5, e6, f5 triad and White's bishop is superior to Black's.'
'Even if'? A sneaky interpretation of a position on which White's d-pawn will fall, whether or not rooks come off (in some cases Black's king comes to the centre), and the best that White can hope for is a draw! I don't think either side can win with proper play, but Black, a pawn ahead, can at least consider playing for ...g5. In this sense, 18...Rac8 and exchanging rooks on the c-file would demonstrate the lack of any problem, whereas 18...Rfd8 with the idea ...Bc6 and ...Rxd6 is more ambitious. To be clear: Black very likely can't exploit his extra pawn. But the optimistic tone of DPA's assessment is deceptive.
Let's move on to the DPA's treatment the oldest and second most popular response to 1 e4, 1...e5. (Curiously, the book devotes 139 pages of coverage to the French Defence, 131 pages to all 1 e4 e5 openings, and 41 pages to the Sicilian Defence. Since so little research is done, I'm not sure which section that favours).
A keystone of their repertoire is the grand old Giuoco Piano main line, which the authors reach via the following moves:
1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Bc5 5 c3 Nf6 6 cxd4 Bb4+
Most readers will recognise this position from the move order
1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 c3 Nf6 5 d4 exd4 6 cxd4 Bb4+. Thousands of games have continued:
7 Bd2 Bxd2+ 8 Nbxd2 d5 9 exd5 Nxd5 10 Qb3
I personally think that this variation deserves more respect than it gets, but not for the reasons given. They continue
10...Na5 11 Qa4+ Nc6 12 Ne5!?
I don't like this move much and prefer 12 0-0.
12...0–0 13 Bxd5
13 Nxc6?! Qe8+ followed by ...Nb6 favours Black according to both DPA and theory.
13...Qxd5
Right away anyone about to play this system as White is going to wonder: 'What about 13...Nxe5 ?'
Not a word about this from the authors! In fact, Black has a clear advantage unless White plays 14 Be4! (14 Qb3?! Nd3+), and I think that Black stands better in that case as well: 14...Ng4 (this forces the action for the next few moves, making the resulting position critical. Also promising is 14...Ng6 with the idea ...c6, ...Be6 and in some cases ...Nf4; that has been played) 15 0–0 Qd6
16 Nf3 (16 g3 Nxh2! 17 Kxh2 Qh6+ 18 Kg2 Qxd2) 16...Nf6 17 Bc2 Bd7 (Good, as are other moves. Instinctively, one feels that Black stands better in this simplified IQP position, for example, 17...Bg4 with the idea 18 Ne5 c5 19 Bd1 Be6, or simply 17...Be6, when ideas of ...c6, ...Qf4 and ...Bd5 might appear. Black got the advantage by playing this way in one correspondence game) 18 Qb3 Bc6 19 Rad1 Bd5 20 Qd3 Be4 21 Qb3 Bd5 (21...Bxc2! 22 Qxc2 Nd5 or 22...Rfd8 is safely in Black's favour) 22 Qd3 Be4 23 Qe2 Bxc2 24 Qxc2 c6 ½–½ Jablonski,J-Sapa,W/Poland 1987/Corr 2000. This whole line looks better for Black, even when he doesn't make ideal moves.
14 Nxc6 bxc6 15 0–0 c5 16 Rac1 cxd4
Why not 16...Qxd4 ? Again, that's a strange move to pass by. Then it's hard to believe in any advantage, for example, 17 Qxd4 (probably not the best try; perhaps 17 Nc4!?, although White's still a pawn down) 17...cxd4 18 Rxc7 Be6 or 18...d3.
17 Rxc7
Now DPA suggest 17...Bb7?! 18 Nf3, with the idea Rd7, and after a brief analysis (18...Rfd8 19 Rfc1!) they say that White has the better chances. Fair enough, but that's a key tempo that Black has lost by playing the committal 17...Bb7?! in a position characterised by a race between the passed pawn and blockade. I see no less than four alternatives that challenge the authors' assessment of advantage for White, and one that is especially difficult to meet. The first is 17...Rd8, since 18 Rxa7? (slow; better 18 Rfc1, when 18...Be6 19 b3 a6! might follow) 18...Bb7! (only now) 19 Nf3 d3 20 Rd1 Rxa7! 21 Qxa7 Qb5! is very strong. Second, 17...Be6 might lead to 18 Rxa7 Rxa7 19 Qxa7 Ra8 20 Qb6 d3 21 a3 Bh3! 22 Nf3 Bg4 23 Nd2 Bh3= etc. Thirdly, one game featured 17...a5 18 Rfc1 Be6 (at least equal), with Black gaining an early advantage but failing to convert the full point. No games are cited by DPA in this entire variation, by the way.
Finally and perhaps most importantly, I suggest that Black can go for more with 17...d3!, the most flexible move of all, gaining a rook move in the line 18 Rfc1 (18 Rd1 Qe5) 18...Bb7 19 Nf3 Re8! with the idea ...d2. That is, the rook went directly to e8 and not to d8 first. In conclusion, White has certain advantages in the diagrammed position, but I don't believe that giving Black the d-pawn, space, and bishop-versus-knight is worth it.
It's really lazy to write a book presenting such specific variations in so old a line without researching or citing games. And the cooperative 17...Bb7?! is certainly not the move that a grandmaster would find appealing. The most suspicious part of this section, however, is that 13...Nxe5(!) is passed over, when any reader would want to know about it, and any engine (or Grandmaster) would find it instantly.
You may note in the above that the authors partake of sins of omission more often than sins of commission. I mistrust the reasons behind the former more than the latter (which can be more easily chalked up to carelessness), and this unfortunate habit pervades the book.
Incidentally, a full-page picture of Michael Adams is featured in this section, saying that he has 'shown his willingness to play the Giuoco Piano from either side'. They neglect to mention that as White, he plays the opening with d3 and not d4.
What is probably the other most important 1 e4 e5 line in the book is the following variation from the Two Knights Defence:
1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Nf6 5 e5 d5 6 Bb5 Ne4 7 Nxd4
This well-known position more commonly stems from 1 e4 e5 2 Nf3 Nc6 3 Bc4 Nf6 4 d4 exd4 5 e5 d5 6 Bb5 Ne4 7 Nxd4. One of the main lines proceeds
7...Bc5 8 Be3
8 Nxc6!? Bxf2+ is an infamous line with immense complications that, according to years of analysis, seems to end in one forced draw or another.
8...Bd7 9 Bxc6 bxc6 10 Nd2
Now the authors analyse 10...0-0 and 10...Nxd2 to White's advantage, which is also what theory concludes. See Pinsky's Two Knights Defence, for example, (Everyman 2003). Pinsky also gives an interesting game with 10...Ng5. Still, the move widely accepted as best is
10...Qh4 11 0–0
Here a common move order that generally transposes is 11 Nxe4 Qxe4 12 0–0, and now 12...0–0 13 Re1 is DPA's main line, but 12...Bb6! is more common, which one of the authors surely must know. We'll get back to that position.
Returning to DPA's 11 0-0 (the diagrammed position), the authors give 11...0-0 12 Nxe4 Qxe4 13 Re1 and analyse 13...Qxe5? and 13...Qh4, both leading to an advantage for White (in fact, they're too nice to Black after 13...Qh4?; In fact, White has a large advantage after both moves).
There are two huge things wrong with this sequence, as a mere glance at opening books and/or years of practice show. Mind you, this not a trivial technical objection, because we are looking at one of the two most important lines after 1 e4 e5 (the other was the Giuoco Piano above), in the key position of the whole line. First, in almost every game that has reached this position (after 13 Re1), whether over-the-board or correspondence, 13...Qg6 has been played, not the other (weak) queen moves. In fact, two of the main books that I have don't consider any other move, with good reason! But 13...Qg6 is not found in DPA's book at all. Black has done perfectly well with it, with an even performance rating, so in any book purporting to be helpful it at least needs to be addressed. Interestingly, only Pinsky doesn't like 11...0-0 because now White has the move 14 Ne6!, leading to 14...Bxe6 15 Bxc5. But that move (14 Ne6) has only been played in two master games that I've found, probably because of the presence of opposite-coloured bishops, and perhaps also because both of those games were drawn easily after 14...Rfb8. Kurajica-Smejkal, Novi Sad 1982 is a good example between two 2550 players if you want to see how Black achieved that.
But even more importantly, Black usually doesn't play 11...0-0; he plays 11...Bb6! (at least four times as often as 11...0-0). It is also theory's choice, and considered more accurate than 11...0-0, even if the latter is satisfactory. Our authors don't even mention 11...Bb6. We can also arrive at this position via the common move order 11 Nxe4 Qxe4 12 0-0, when again Black usually plays 12...Bb6, transposing to 11 0-0 Bb6! 12 Nxe4 Qxe4. Between these two orders, the following position has been well-tested:
Black has the two bishops, the idea of ...c5, and active pieces. He has scored very well, and 13 Re1 is met by 13...Qg6, when he has done even better (about 65%, with a significantly superior performance rating). Let's assume that the authors were somehow unaware of 11...Bb6; it's still not a good idea to write a book using neither a database nor a computer engine, or for that matter your own judgment.
And on it goes. James Vigus points out that after 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Nf6 5 e5 Ne4, DPA's uncommon recommendation 6 Qe2 Nc5 7 c3 is answered in the book by only 7...dxc3(?), although 7...d3! is probably best (and even 7...Ne6 has been played with level results), citing two games including his own nice win; I've found a few more games with 7...d3 (two by grandmasters), with nary a victory for White.
It turns out that 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Nf6 5 e5 Ng4 is also punchless for White; they practically admit as much in the line 6 Qe2 Qe7 7 Bf4 d6 8 exd6 Qxe2+, claiming a small edge in what both the limited theory and I think is a completely harmless position.
I should add that throughout, the authors spend page after page analysing closely related variations in extravagant detail out to move 20 and beyond, so you needn't write off these objections on the grounds that the book is on an elementary level which therefore excuses such simple omissions.
In any case, if you're an inexperienced player, or someone looking for variety, it's all right to have a low-level set of lines like those offered by DPA. Still, it would be nice to be prepared for the moves that you will see over the board; and furthermore, little of this fits into the idea of a 'lifetime repertoire'. Speaking of which, in my database from The Week in Chess, Perelshteyn plays and has played the Ruy Lopez (1 e4 e5 2 Nf3 Nc6 3 Bb5) exclusively for many years, going all the way back to a single exception in 2001.
Actually, nearly every section of 1 e4 e5 has some kind of issue. In the Philidor Defence, we are given the same old standard 1 e4 e5 2 Nf3 d6 3 d4 solutions (3...Nd7 4 Bc4 followed by dxe5; 3...Nf6 4 dxe5 etc.). But it's hardly a secret that the top players (a significant number of them, all the way up to the 2700s) have used the Pirc Defence move order 1 e4 d6 2 d4 Nf6 3 Nc3 e5 in order to get to the Philidor (they don't fear 4 dxe5 dxe5, and 4 Nf3 Nbd7 is the transposition). The other Pirc-based move order has been 1 e4 d6 2 d4 Nf6 3 Nc3 Nbd7 (intending 4 Nf3 e5), as played by Philidor expert GM Christian Bauer, who has analysed it at length in his excellent new book 'The Philidor files'. The Philidor Defence is very much alive, then, and should have been attended to in the book before us.
Moving along, White must do something about the challenge of the Sicilian Defence. Their answer is the Grand Prix Attack (2 Nc3 and 3 f4), an old Dzindzihashvili favourite. If the authors were content with saying 'Here's one way to give White a safer position to work with', I'd be more sympathetic. They do in fact admit that the Grand Prix doesn't lead to an advantage versus perfect play (something that they are compelled to do because they advocate the Sicilian Defence in their repertoire book for Black!). But again, we will find this practice of ignoring Black's best lines, even in the authors' own games, which is something that I suspect goes way back to Dzindzihashvili's early videotapes. This is particularly unacceptable for key moves in standard positions. Let's see:
1 e4 c5 2 Nc3 Nc6 3 f4 g6 4 Nf3 Bg7 5 Bb5 Nd4 6 0–0 Nxb5 7 Nxb5 d5 8 exd5 a6 9 Nc3 Nf6
Now we get the characteristic straw-man argument from the authors: 'This is the line recommended by theory. Black wins back the d5-pawn and "the books" maintain that he stands better because of his bishop pair." However, we disagree with this assessment and believe that White obtains the advantage!'
It's always noble to challenge "the books"; the problem is that I can't find a single book that assesses this position as better for Black (no fewer than five call it equal).
And do the authors back up their claim that White gets the advantage? Let's see:
10 d4
Here DPA spend almost 5 pages analysing the move 10...Nxd5 in detail, citing almost no games, but they dismiss the more important:
10...c4
Plaskett assigns this an '!' in his Sicilian Grand Prix Attack, for example (other books seem to like it as well), and it is likely one reason that this line hasn't gained in popularity for White. Against 10...c4 DPA suggest
11 Qe2'!'
And continue '11...b5 12 Re1 Bb7 13 f5!'. That position ends up with advantage to White, although I think that 12 a4! is more convincing and very bad for Black, pretty much refuting 11...b5. Be that as it may, that's all the analysis we get, and the most serious move isn't given at all:
11...Nxd5
Now let's follow a game:
12 Qxc4 Be6
13 Nxd5
I'll offer some notes on the (untried?) move 13 Re1 below.
13 Qd3 is the obvious alternative, when Rutherford- Gourlay, Aberdeen 2001 went 13...Rc8!? (A clearer route is 13...Bf5! 14 Qd1 Rc8 with excellent play, or here 14 Ne4 Nb4 15 Qb3 Bxd4+ 16 Kh1 Bxe4 17 Qxb4 Bxf3 18 Rxf3 Bc3! 19 Qc4 Rc8 20 Qe2 Bf6) 14 Nxd5 Bxd5 15 Rd1 Bxf3 16 gxf3 (Or 16 Qxf3 Bxd4+ 17 Be3 Bxe3+ 18 Qxe3 Qc7) 16...Qd5 (16...e6) 17 c3 0–0 18 Qe4 Rfd8 19 Re1 Rc7 (19...Bf6!) 20 Qxd5 (20 f5! Qxf5 21 Qxf5 gxf5 22 Bf4) 20...Rxd5 21 Re2 b5=.
13...Bxd5 14 Qe2 Rc8!?
Threatening ...Bc4. This seems good enough, but better is 14...Bxf3!, for example, 15 Qxf3 Qxd4+ 16 Kh1 0–0 with the idea 17 Qxb7!? (17 c3 Qb6) 17...Qc4! , and Black has slightly the better side of what nevertheless has to be called practical equality, e.g., 18 Re1 (18 Be3 Rab8 19 Qxe7 Rxb2) 18...Qxc2 19 Qxe7 Rfd8!
15 Rd1 Bxf3 16 Qxf3 Bxd4+ 17 Kf1
17 Be3 is the best chance for White, although 17...Bxe3+ 18 Qxe3 Qc7 isn't much to brag about either.
17...Qb6 18 Qb3
18 Qd3 Rd8.
18...0–0 19 c3 Bc5 20 Qxb6 Bxb6=
Perelshteyn-Lawson, Montreal 2005! The end position led to an eventual draw. This game by Perelshteyn was played in January of 2005, and the book before us was published in 2007. Having devoted 5 pages to 10...Nxd5, it's surely no mistake that he and that other World-Class-theorist-and-5Bb5-specialist skipped 10...c4 11 Qe2 Nxd5 altogether. Of course, lack of transparency in writing a single chess book needn't correlate to anything else. My prediction is that in his next book (hopefully not co-authored), the unassuming and deservedly popular Perelshteyn will free himself from others' bad habits and do a great job.
All right, let's take a break. For the sake of an ex-student who plays this variation, I tried to find an improvement. Something like 12 a4 instead of 12 Qxc4 is interesting but really nothing special. So I tried to make 12 Qxc4 Be6 13 Re1 work (instead of Perelshteyn's 13 Nxd5. Incidentally, 13 Ng5? is poor after 13...Nb6). For fun, or perhaps someone's use, here's some analysis. After 13...0–0, best seems to be 14 Qe2! and the play is complex and instructive, although I think Black has two ways to come out well enough:
a) 14...Bg4 (the simplest) 15 Rd1 (15 Nxd5 Qxd5 16 c3 Bxf3) 15...Rc8 (15...Nb4!? 16 Be3 Bxf3 17 gxf3 Rc8) 16 Nxd5 Qxd5 17 c3 Bxf3 18 Qxf3 Qxf3 19 gxf3 e6 followed by ...Rfd8. Then ...Bf6/ ...Kf8 is one plan, but ...Bf8 intending ...Kg7-f6-f5 is more ambitious, and fun!
(b) The complex route is 14...Nxc3 15 bxc3 Rc8 16 Bd2 Rc7 with the idea ...Bf5, and now best seems 17 Ne5! Bf5! 18 Qd5 19 a4 Rfc8! and apart from his doubled and isolated pawns, White may find ...Rxc3 hard to prevent: 20 Ra3!? (20 g4?! Bxd3 21 Qxd3 (21 cxd3 Rxc3! 22 Bxc3 Rxc3 23 Qe4 Qd7) 21...h5! 22 g5 Rxc3! 23 Bxc3 Rxc3 24 Qe4 Qc4! 25 Re2 Bxd4+ 26 Kh1 Bc5!; 20 Nb4 Qd7 21 a5 Bf6 (or 21...Rxc3!? ) 22 Kh1 Rxc3! 23 Bxc3 Rxc3 with equal play) 20...Rxc3! 21 Bxc3 Rxc3 22 Rxc3 Bxd4+ 23 Nf2 Bxc3 24 Rd1 Qc5 with dynamically balanced play, although I'd rather be Black. Very interesting material; naturally, in such long analysis both sides may be able to improve.
O.K., back to the book. It's hard to be as optimistic as DPA are about the lines in which Black plays ...a6 and ...e6, for example, their main, with relatively forcing moves: 1 e4 c5 2 Nc3 Nc6 3 f4 g6 4 Nf3 Bg7 5 Bb5 Nd4 6 0–0 a6 7 Bc4 e6 8 d3 Ne7 9 Nxd4 cxd4 10 Ne2 d5 11 Bb3 dxe4 12 dxe4 0–0 13 e5 Nc6 14 c3 dxc3 Here, instead of 15 Nxc3, which led to a draw in a fairly recent game between 2500+s, they give 15 bxc3!! (their notation mark), to control d4. But now they have Black fall for it by 15...Qc7?! 16 Be3 Rd8? 17 Qe1 Na5 18 Qf2 and say 'White is better. Black has trouble with his development because he has to worry about the b6 square.' That must be true, although his edge after 18...Nc4 seems rather small, e.g., 19 Bc5 b6 20 Bxc4 bxc5 21 Ng3 Bb7. I still like White by a fair margin (look at the bishop on g7!), but it's not easy after ...Qa5 or ...Bd5, trying to double on the d-file, and perhaps ...Bh6 at the right moment.
The main point is that Black doesn't have to walk his pieces into the attack on the dark squares by 15...Qc7?! and 16...Rd8?. Rather, he can play 15...b5! 16 Be3 Bb7, for example, 17 Qe1 (17 Bc5 Re8 18 Qe1 Na5 19 Rd1 Qc7 20 Qf2 Nxb3 21 axb3 Bd5=) 17...Na5 18 Qf2 (18 Rd1 Qc7) 18...Nxb3 19 axb3 Qd5, hitting b3 and intending ...Rfd8 with easy play.
Some common Black orders are given short shrift. A suspiciously bad section is that on 1 e4 c5 2 Nc3 e6, one of the two main responses to 2 Nc3. The authors recommended 3 f4 d5 4 Nf3, not exactly the most line most feared by Black, and already they degenerate by using some odd tricks:
(a) The first piece of legerdemain is their statement that '4...d4 fails to disrupt White's development after 5 Ne2 d3?! 6 cxd3 Qxd3 7 Nc3!', [etc., with White advantage]. That's it - no other line with 4...d4. Anyone who follows the repertoire will doubtless run into that natural move, but in my main database I found only 1 game out of 51 in which the mistake 5...d3? was played after 5 Ne2. In fact, Black has a small lead in both winning percentage and performance rating after 4...d4 5 Ne2. Most grandmasters such as Bent Larsen played simply 5...Nf6, and a particularly interesting plan was played by Alburt's co-author Chernin: 5...Nc6 6 d3 f6 followed by ...Bd6 and ...Nge7, when the knight on e2 is poorly placed in comparison to standard King's Indian Attack positions. Again, even as early as move 5, the reader is given no idea how to face Black's most likely responses!
(b) The 'main line' is 4...dxe4 5 Bb5+ Bd7 6 Bxd7+ Nxd7 (here 6...Qxd7 is mysteriously skipped, intending 7 Ne5 Qc7; it is logical and has done fine in the few samples that I have found) 7 Nxe4 Ngf6 (7...Nh6 is another good, unmentioned, move) 8 d3. They stop, suggesting the follow-up 0-0/Qe2, Bd2-c3 and Rael. We are told that Black is 'somewhat cramped' and 'It's not easy for Black to come up with a plan'. However, if you look at Black's superior pawn structure, it's not hard to see why he has come up with the natural plan of ...Be7/...0-0/...Qc7, intending to enforce ...c4, in a majority of contests. For what it's worth, 5 Bb5+ scores 2.5-7.5 in my database! To be fair, that doesn't necessarily signify a theoretical advantage (never rely solely upon 'results' statististics), but it strongly suggests that the reader be given some solid guidance before being thrown to the wolves. In a book with such detail, it's remarkable that so little attention is given to this vital line.
There are several other lines in the Sicilian Defence repertoire that would at the very least make a player hesitate to use them for his or her entire playing career, or perhaps to use them at all. Indeed, it may be indicative that in 2006, using a database of TWIC issues, Perelshteyn himself played 10 games as White with 1 e4 c5 2 Nf3; I didn't find one with 2 Nc3. And in 2005 he played eight games with 2 Nf3 and only two with 2 Nc3. To be clear: I don't think that a player should use the openings he has written about if he finds them wanting, or even if not; there are no such rules in the real world of competition. But moving away from inferior openings plays a role in most talented players' success.
I don't have time or inclination to go into the similar treatment of other openings. However, I feel that I should object to the characterization of the Alekhine Defence line 1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 c4 Nb6 5 exd6 cxd6 6 Nc3 g6 7 Be3 Bg7 8 Rc1 0-0 9 b3 as 'yet another of the Dzindzi-novelties'. He does include a suggestion to improve upon his own play on move 17 of one game that he played in 1996, but the system was developed in detail in the town of Voronezh in Siberia in the 1980s, most notably by the world correspondence champion Gregory Sanakoev. The system has become extraordinarily popular and has been called the Voronezh system in every source that I've seen.
In the end, what is wrong with this book? The main point is not that so many of the lines the authors have given us above are bad, or ineffective, although that is certainly an issue. Rather, it's the lack of integrity throughout. For example, is it conceivable that grandmaster Dzindzihashvili, having played and advocated these lines for so many years, is unaware of the simple and widely-known responses that he skips over? Total objectivity in presentation may not be expected in an advocacy book - you can even argue that it's less fun for the reader. But deliberate deception on the most important points is another matter. All the more so given the outlandish claims that accompany this book.
Alburt says in the Introduction: 'Players of all strengths, from beginners to Super-GMs, will profit from this book'. It seems that a few GMs are profiting from it, but not by improving their chess.
