THE CHESS ADVANTAGE IN BLACK AND WHITE
Author: Larry Kaufman
498 pages
$18.95
McKay (2004)

Reviewed by John Watson

How to assess THE CHESS ADVANTAGE IN BLACK AND WHITE (“CABW”)? One of the most important qualities of an opening book is whether it is useful to the average player. In this respect Kaufman’s work fits the bill; in fact, it’s hard to overstate how well it does so. I have seldom if ever seen such a sound and detailed repertoire for one color, much less for both. The literature on complete repertoires tends to be full of exotic and often dubious attacking lines; in my observation and experience these books are also full of weak analysis and holes, and most aren’t even well written. Other authors have given us bare or simplistic overviews of mostly punch-less systems. Kaufman’s book presents legitimate and often very promising ways to play. He presents a vast amount of chess material. His analysis is on a high and relatively objective level very seldom seen in books that promote particular systems. In fact, the 1...e5 recommendations for Black in this book have already provided a core for the repertoires of one of my students and for two 2300+ players that I know! I’m sure that there also are a few GMs out there who could pick up some ideas just from reading it. In particular, Kaufman saves one the time-consuming task of finding continuations to employ against the various irregular lines. The value of time saving and practical variations extends to everyone. Let me make it clear, however, that the coverage is necessarily simplified. Often the analysis is optimistic and ends at points that need clarification. I would say that CABW is useful for everyone, but primarily so for a broad audience that I see going from 1400 all the way up to 2100. This breadth is one of the book’s strengths. Kaufman also has a great many original analytic suggestions (which he attributes to the role of computer programs), and they sometimes involve modern variations that GMs use, even if not too many super-critical ones. But Kaufman’s biggest contribution in my opinion is his diligent investigation of current theory and application of his judgment to glean the most promising continuations against various lines. This is particularly true in some slow and apparently prospectless continuations. Naturally he has plenty of analysis that I would argue with on a technical level, but that’s inevitable. I am extremely impressed by this unique contribution to the literature that reflects great effort on the part of the author.

Apart from the material dealing strictly with chess, there are some problems. For example, as Carsten Hansen points out, the book is organized by games and yet the reader often doesn’t find out where the two opponents made mistakes or went wrong. Kaufman uses “!”s fairly often, but when it comes to main moves the “?” character is extremely rare. So as you are reading through a sample game (which might be played with very short time controls) you would think that the contest was extraordinarily well or even perfectly played, to the extent that Kaufman’s ultimate assessment (in favor of the repertoire-player) is based upon a presumption of mutually flawless play.

To be fair, Kaufman does give many verbal indications and explanations of why various moves are inferior. The problem is that these explanations are largely limited to alternatives in notes that are either known to be bad or fairly obviously so. Is the repertoire opponent really playing perfectly in the main games, or even in some important games in the notes? In one case I had serious questions about as many as five of the moves played in a game that was presented without annotation marks. This situation can cut into one’s confidence about the final assessment of the variation examined, and indeed some optimistic assessments seem to be based mainly upon the continuation of one flawed game. I should add in Kaufman’s defense that I agree with the majority of his analysis and feel that he makes appropriate choices of moves to prove his case. It’s just that we need more clarification of which moves were good ones and which were flawed.

Another obvious problem is that there are almost no game references outside of the very main games. This is in spite of the fact that many of the most important notes tend to come from main-line theory and often stem from a specific game. Kaufman says that “I have generally refrained from giving game or source references in side variations [jw: which make up most of the book], due to space considerations and because so many of the variations are computer-enhanced.” These strike me as feeble explanations. First of all, as someone who has written 26 books, I can tell you (and Kaufman surely knows) that game references take very little space. And oddly enough, he says, “Every recommendation in this book has substantial support from very strong grandmasters” and that “the repertoire itself [is]... in fact basically a composite repertoire of some of the world’s best players.” Since the games in the notes are unattributed, how can we know who these world-class grandmasters are, which lines they play, how recent the games are, etc.?  Most of us would naturally be interested in which top-class players use our lines, if only to emulate the refinements those masters make. Furthermore, if the variations are computer-enhanced, that’s all the more reason for us to know when and where improvements came. Here’s a brief example of how this makes a difference. Obviously it’s tough to show a White advantage, however small, in the Spanish Exchange Variation. In the following line Kaufman doesn’t mention that the world-class players are predominantly on the Black side of the board:

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Bxc6 dxc6 5.0–0 Qf6 This move is dismissed by Kaufman in a six-line note without references. A look at Megabase and recent TWICs shows that Black has a small plus both statistically and in performance rating. 5...Qf6 has been used by Morozevich, I. Sokolov, and Adams, the latter cited by Kaufman as a representative of several of his favorite lines. 6.d4 exd4 7.Bg5 Qd6 8.Nxd4 Bd7 Recently Belyavsky won a game as Black after playing 8...Be7 first, but the idea is not fundamentally different. Notice that we take this variation more seriously when we see the players associated with it. From Kaufman’s note we might just assume that 5...Qf6 is an obscure and inferior line.  9.Nc3 Be7 10.Bxe7 Nxe7 11.Nb3

This is all from the game Magem Badals-Morozevich, Pamplona 1995, although it is presented without references. Mororevich played 11...0–0–0 which probably should have led to a small disadvantage. Instead, 11...Qxd1 12.Raxd1 is logical, and here Kaufman gives 12...0–0–0 13.Nc5 “with a healthy versus crippled pawn majority for ‘free’ ”  (i.e., without having to concede any bishop-versus-knight imbalance). But in the book SURVIVING AND BEATING ANNOYING CHESS OPENINGS (see the reference below) I gave simply 12...b6 (to stop Nc5) with ...0–0–0 to come, retaining the bishop in an open position and establishing a desirable pawn structure for Black. It’s hard to believe that White has the slightest advantage in that case. In fact Black might take inspiration from Kaufman’s favorite Berlin Defense when he takes up this line and pawn structure.

What else? I hate to carp on it again, but the lack of an Index of Variations is a severe drawback for a book like this which is organized by sample games and detailed notes. I simply can’t understand why author after author neglects to provide this important feature, one that makes the reader’s task so much easier. In Kaufman’s book I had to jump around from game-to-game, page-to-page and note-to-note to find the specific lines and variations that I was interested in; this is irritating and unnecessary. It only takes two or three pages to lay out the important variations and subvariations with page numbers, and if you devote five or more pages of this size you can cross-reference most reasonable alternatives from subsections and notes in considerable detail. A book of 497 pages would hardly suffer from inclusion of such an Index. To be very clear: a majority of the opening books which are organized around games don’t have any Index of Variations either, so this is hardly a problem unique to the book before us. I just want to object (again!) to this lazy practice. It indicates a lack of concern for the reader on the part of the editor and author.

For all that, you should return to the fifth paragraph above and you can guess what relative weight I attribute to these criticisms. I think that they are far from trivial but by no means change my strong overall recommendation for the book. The quality of its chess content outweighs everything else.

Okay, let me turn to a few examples in areas that I find interesting. As indicated above, I have some problems with the specific analysis in the Semi-Slav section. I have few objections to most of the first part of the book (1.e4 as White) and in fact learned a lot about openings that I don’t play, as well a few interesting things about those that I do. Before moving on, however, the reader is probably wondering what Kaufman does about the classic problem of the juncture at which a Black and White repertoire collides. In this case he has to recommend something for White against the Berlin Defense of the Spanish Game and at the same time must support the very core of his repertoire for Black. To resolve this issue Kaufman produces a minor sleight of hand: as Black, every line equalizes or better. After 1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.0–0 Nxe4 5.Re1 Nd6 6.Nxe5, for example, Kaufman says that he prefers the order 6...Be7, and if 7.Bf1, 7...0-0 8.d4 Nf5 9.c3 Nxe5 with “near equality”. Anyway, he uses the game Shirov-Hracek, Lübeck 2001 to illustrate the main line 6...Nxe5 7.Rxe5+ Be7 8.Nc3 (Kaufman says that “here or on move 9 I prefer 8.Bf1”, which leaves the reader a bit confused as to how Black might proceed. It turns out this is added to accommodate his recommendation in the White section) 8…0-0 9.Bd3

9...c6 10.b3 Ne8 11.Qf3 d5 12.Bb2 Bf6 13.Re2 Nc7 14.Na4 Bxb2 15.Nxb2 Qd6 16.Rae1 Bd7 17.Bf5 Rae8 18.Bxd7, 1/2-1/2.

But needing to show something for White in his corresponding repertoire, Kaufman cites Borisek - Mitkov, Terme Zrece 2003, which continued 5.Re1 Nd6 6.Nxe5 Be7 7.Bf1 Nxe5 (here we have the confusing transposition 7...0-0 8.d4 Nf5 9.c3 Nxe5 10.Rxe5 d6 11.Re1 d5 12.Bf4 Bd6 13.Bxd6 Nxd6 14.Nd2 Bf5 when “White must be preferred”, but he goes on to say that “it’s not much”) 8.Rxe5 0–0

9.Nc3! (Kaufman: “I think that this is better that 9.d4 as analyzed in the Black portion of this book, which was written before this game was played.”) 9...Bf6 (Kaufman analyses alternatives on each move, with White remaining better) 10.Re1 b6 11.d4 Bb7 and here instead of 12.Bf4 which led to a small advantage, he likes 12.Qg4 Re8 13.Bd2 g6 14.Bd3 a6 15.Rxe8+ Nxe8 16.Re1 with a “much superior rook” that gives White the edge. Shirov went on to win. All this is a bit confusing. Since he goes back to partially correct the record in Black’s section, he really should have just admitted in that section there that according to his own analysis White keeps a small edge in all 5.Re1 lines and that Black just has to live with it. As a side note I personally suspect that Black is fully equal after 5.Re1 and that White gets nothing at all versus correct play! But that’s not the goal of Kaufman’s repertoire for White and in any case there are plenty of points at which Black can equalize in other defenses to his 1.e4 lines, so this is no big deal.

As indicated above Kaufman presents a set of unambitious but still worthwhile lines against the Sicilian, about which I feel that he is overoptimistic. This is understandable given the context (getting an advantage versus the Sicilian with simple lines!), but nevertheless worth mentioning. I no longer know the theory related to what parts of his Moscow Variation hold up or don’t, but I checked 1.e4 c5 2.Nf3 d6 3.Bb5+ Bd7 4.Bxd7+ Qxd7, which used to frustrate me when trying to teach 3.Bb5+ as an anti-Sicilian weapon. Kaufman gives both 5.0–0 (without c4) and 5.c4 for White. I was interested in the former and was a little surprised at Kaufman’s enthusiasm for 5.0–0 Nc6 6.Qe2 Nf6 7.Rd1 (“!”).

Now he uses 7...Rc8 as his main line. Against the logical 7...g6 Kaufman suggests the line 8.c3 Bg7 9.d4 etc., with a nice advantage for White. But with the c3 square unavailable for a knight, Black could have played the forcing line 8...Qg4! which hits e4 and certainly should have been mentioned, e.g., here’s a few plausible lines: 9.d3 (9.Re1 Ne5 – even 9...Bg7 improves upon Kaufman’s main line – 10.Nxe5 Qxe2 11.Rxe2 dxe5 trades Black’s odd pawn structure for White’s and gets a weak backward pawn on the d-file. There might follow something like 12.Na3 Bg7 13.Nc4 Nh5!? 14.d3 – 14.g3 0-0-0 – 14...Nf4!? 15.Bxf4 exf4 16.e5 0–0–0 17.Rd1 Rd5) 9...Ne5 10.Nbd2 (10.Be3 Nxf3+ 11.Qxf3 Qxf3 12.gxf3 Bg7 13.d4 cxd4 14.cxd4 e6 15.Nc3 Ke7 16.Rac1 Rhc8 is also comfortable for Black) 10...Nh5 11.Nf1 (11.h3 Nf4) 11...Nxf3+ (or 11...0–0–0; or 11...Bg7 12.Be3 Nxf3+) 12.Qxf3 Qxf3 13.gxf3 Bg7 14.d4?! (but White’s structure reduces the value of his center anyway, e.g., 14.Be3 0–0–0 15.d4 cxd4 16.cxd4 Kb8 17.Ng3 Nf6; or 14.Ng3 Nf6 15.Be3 Kd7 16.d4 cxd4 17.cxd4 Rhc8) 14...cxd4 15.cxd4 Rc8 16.Be3 Rc2 17.Rab1 Kd7 with ...Rhc8 to follow.

Likewise, after his main line 7...Rc8 8.c3 e6 9.d4 cxd4 10.Bg5 Be7, Kaufman says: “Effectively conceding that White has won the strategic battle in the opening.” He continues 11.cxd4 h6 12.Bxf6 Bxf6 13.Nc3 0–0

In Ovetchkin-Kobalija, Togliatti 2003 White played 14.d5?! Ne5 which “activated Black’s army”, but Kaufman suggests 14.Rac1 d5 15.e5 Be7 16.Qd3 to meet …Bb4 with Ne2. “The blocked position favors knights over bishops.” But this is a standard French Defense position in which Black has traditionally done fine by advancing his queenside pawns and creating weaknesses. More ambitiously, because of his good bishop he can open up the f-file by 16...f6 17.exf6 Rxf6 followed by ...Bd6 and ...Raf8 with a standard French position that is at least no worse for Black. If White protects f3, moves like ...Qf7-h5 come. Perhaps this is a position that White can defend with equality, but not one that I would like to play for him.

Kaufman indicates several interesting ideas using the Tarrasch Variation versus the French Defense. This is especially so in the 3...Be7 variation where he goes for one of the most aggressive continuations 4.e5 c5 5.Qg4 and makes a good case for it, definitely improving upon some subvariations that I had thought harmless in my own French book. Those are likely playable for Black but are irritating positionally and more difficult to handle than I (or conventional theory) suggest. On the other hand, Kaufman apparently doesn’t know about my suggestion 4...Nh6, but you can’t have everything.

I do think that Kaufman has some problems trying to avoid the lines with 1.e4 e6 2.d4 d5 3.Nd2 c5 4.exd5 Qxd5 that are doing well for Black. His solution is 4.Ngf3, long thought to be harmless, and I think that it is. For example, after 4...cxd4 5.Nxd4 Nc6 6.Bb5 Bd7, Kaufman strains to gives White an advantage but I think falls short. In one dull variation (7.Bxc6 Bxc6) he would need to show an advantage when Psakhis correctly says that there is none whatsoever. And there is a more promising line for Black after 7...bxc6 that I like because it gives him positive chances; Kaufman has Black playing poorly and passively, thus neglecting fairly obvious improvements.

I also found the main 3.Nd2 Nf6 line to be quite interesting inasmuch as I haven’t kept up with the theory. After 4.e5 Nfd7 5.Bd3 c5 6.c3 Nc6 7.Ne2 cxd4 8.cxd4 f6 9.exf6 Nxf6 10.0–0 Bd6 11.Nf3 0–0 12.Bf4 Bxf4 13.Nxf4 Ne4 Kaufman suggests 14.Qc1 (avoiding the lengthy theory on 14.Ne2 and 14.g3) 14...Ng5 15.Nxg5 Qxg5 and says simply that the position after 16.Ne2 favors White:

This is typically safe but the assessment is probably optimistic. The theory and annotated games that I saw didn’t bear it out, not even after the unambitious 16...Qxc1 17.Raxc1 a6!. The weakness on d4 tends to balance that on e6, and importantly the rooks will likely be exchanged on the c-file, as happened in the only game I found where the opponents didn’t agree to a draw earlier! Nor has the main line with 16...Qf6 17.Qe3 e5 18.dxe5 Nxe5 been shown to give White anything, with Black’s active piece play balancing out White’s potential target on d5. This is open to argument of course. White gets play, which is the most important thing. However, I’m not sure how useful such lengthy lines will be to the average player. Usually Kaufman neatly avoids such messy theoretical choices in his book without sacrificing real chances for White; it’s a bit harder to do so here.

Finally, in the Russian Defense (a.k.a. Petroff’s) I’d like to look at the Kaufmann Attack (that’s with 2 “n”s and no relation to the author), which I recommended in SURVIVING AND BEATING ANNOYING CHESS OPENINGS (“SBACO”, published in 2003 by Eric Schiller and I). This goes 1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.c4. Here is great example of how Kaufman sidesteps the massive body of Petroff theory to present a simple but non-trivial system. Ever since writing SBACO I’ve been teaching it to my students and honestly believe that it holds some promise at any level. Kaufman generously credits our book in his introductory remarks, adding that it “predated the novelty which motivated my recommendation.” Here’s the line he’s talking about, a truly critical one which is probably less convincing for White than Kaufman makes it seem:

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.c4 Nc6 [One of Black’s best moves] 6.Nc3 Nxc3 7.dxc3 Be7 8.Bd3 Bg4

Now comes the novelty Kaufman refers to, citing a game Iordachescu-Pavasovic, Bled 2002: 9.Be3! In SBACO we gave 9.Be4 0–0 10.Qc2 and followed Stefansson-Sokolov, Reykjavik 2000: 10...h6 11.Be3 Be6 12.b3 Bf6 13.0–0–0 a5 14.a4 Ne7 15.h4 Qc8 (In light of Kaufman’s next comment, note that in this position White has a 20–7 ratio of attacked squares on the enemy side of the board) 16.Bg5! with a killing attack that led to a quick win. Obviously this can be improved upon, but it’s not clear either that Black can fully equalize. Nor that 9.Be4 isn’t as good as or better than 9.Be3. 9...Bxf3 Better than 9...Ne5 10.Be2 Nxf3+ (10...0–0!? 11.0–0 Be6) 11.Bxf3 Bxf3 12.Qxf3. Kaufman notes: “If you compare the number of squares attacked on the enemy half of the board (as advocated in some manuals), it’s 14 to 1. It’s no wonder the computers like White here”. That is an interesting (and often relevant) way of looking at things, but is a harsh judgment upon those who play Hedgehogs and King’s Indians! 10.gxf3 Pavasovic played 10...Ne5 here and Kaufman follows that game as his main line. He also analyses 10...Qd7 with a continuation that I find unconvincing for White:  11.Rg1

11...g6 Probably the best move, but 11...0–0 is also possible, when the ambitious 12.Be4 (12.Qc2 f5 13.f4 may be a better try) 12...f5 13.Bd5+ Kh8 14.Qd2 runs into 14...Na5! with ...c6 to follow. 12.f4 0–0–0 13.Be4. Kaufman stops here, favoring White’s bishop pair. That may not be the case after 13...Rhe8 intending ...Bh4. Then I’m not sure how White should proceed. Not 14.Qa4?! due to 14...Bh4! 15.Bxc6 Qxc6 16.Qxc6 bxc6. Probably the whole line after 9.Be3 is dynamically balanced, with 9.Be4 needing more tests.

You can see how thought provoking this book can be. I do think the repertoire would serve a large middle range of players better than it would strong masters, if only because some of the lines presented seem to peter out against informed opposition. But everyone, and especially those intending to form a repertoire for Black using 1.e4 e5, will find much of value in it. THE CHESS ADVANTAGE IN BLACK AND WHITE deserves to be a bestseller and will likely achieve that status. It is simply the best comprehensive repertoire book that I have ever read.

Click if you would like to purchase THE CHESS ADVANTAGE IN BLACK AND WHITE.