|The Week in Chess Magazine
Sponsored by the London Chess Center
|TWIC Home | The London Chess Center | | Shop|
More NCO, Gambits and Repertoires
Hanging Pawns; Adrain Mikhalchishin & Wit Braslawski; 203 pages; International Chess Enterprises (ICE), 1998
Four Gambits To Beat the French; Tim Harding; 264 pages; Chess Digest, 1998
A Startling Chess Opening Repertoire; Chris Baker; 240 pages; Cadogan Chess, 1998
Since the last few reviews, I have received a number of interesting and critical responses. First and foremost, my ill-considered remark that the Cochrane Gambit is 'simply unsound' and capable of refutation aroused the ire of a subculture of Cochrane-Heads (of whose existence I was previously unaware; would that I could return to those innocent days!). These loyalists sent me glowing reports on the Cochrane and in two cases, truckloads of games. At the very least, these games proved to me that, when neither side has the slightest idea how to play the resulting positions, the Cochrane is probably more dangerous for Black than White!
As a result, I spent more time than I'd like to admit investigating 1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nxf7 Kxf7. Remarkably, as if to humiliate me further, no less than world-class player Veselin Topalov just played 4.Nxf7 against Kramnik in Linares, and drew the game, which would seem to render my 'unsound' comments to the scrap heap. However, there are a few points to make about that game: (a) Topalov played this after a loss as White to Kasparov, in which the latter's superior theoretical preparation (extending past move 20 in a main line) must have depressed Topalov. So a crazy experimental response may have been just the therapy he needed in the next round; (b) More importantly, after 4...Kxf7, Topalov played not 5.d4 (the only move considered in most sources), nor even 5.Bc4, Cochrane's original idea (discredited by 5...d5!), but 5.Nc3!?, a move probably designed to avoid the known drawbacks of the other two moves. Now Kramnik responded with 5...c5 6.Bc4+ Be6 7.Bxe6+ Kxe6 8.d4 Kf7 9.dxc5 Nc6, a completely safe method which appears to me to be at least equal. We'll have to see what the players' notes say. The only theoretical comment I can find on 5.Nc3 gives it a '?!' and suggests 5...Qe8! 6.Bc4+ Be6, when Black is clearly better (Osnos and Kalinchenko in NIC Yearbook 19). I'm sure that Topalov would have played 6.d4! instead, with the idea 6...Nxe4? 7.Qh5+ g6 8.Qd5+. However, Black can play 5...Qe8 6.d4 d5 7.e5 Bb4, transposing to a normal Cochrane (if there is any such thing), and the move 5...Qe7!? also deserves strong consideration, intending 6.d4 c5.
These two lines are, by transposition, among the ones I examined in my investigation of the Cochrane with 5.d4, i.e., Black can play 5...Qe8 (as I recommended in the Big Book of Busts in 1995) 6.Nc3 d5 or 5...Qe7!? 6.Nc3 c5 (my own invention). I believe that both of these lines favor the second player, although there are all kinds of complicated byways. This is not an opening analysis column, but since Eric Schiller and I are currently updating 'The Big Book of Busts', I will address the issues in some detail in that book. In the meantime, I don't want to inhibit Cochrane players from their favorite activity (playing 4.Nxf7). Still, I should point out that generally, if White loses any potential advantage and can at best claim equality after only 7 or 8 moves of a gambit opening, then the gambit is often referred to as 'refuted'. See, for example, the well-known Danish Gambit 'refutation' 1.e4 e5 2.d4 exd4 3.c3 dxc3 4.Bc4 cxb2 5.Bxb2 d5 6.Bxd5 Nf6 7.Bxf7+ Kxf7 8.Qxd8 Bb4+ etc.. I doubt that White in the Cochrane can achieve anything this close to equality (the Danish line is variously given as '=+' or '='); but even if he somehow achieves a slightly-worse, three-pawn-versus-a-piece ending, that isn't much to brag about.
In the course of my study, I had to order Chris Baker's A Startling Chess Opening Repertoire, which features 17 pages on the Cochrane. I hadn't intended to review that book (since I didn't receive a review copy); but, having had some time to look it over, I decided that it is a book which can teach us some lessons about repertoire books. I will also look at Tim Harding's Four Gambits to Beat the French here. But before getting to those books, I want to address other issues and another recent book. (This is going to be a very lengthy column!)
Among the responses to my 'preliminary' NCO review, a few were quite interesting. For one thing, I got notes from two of the authors, John Nunn and Graham Burgess, addressing issues raised in that review. Terrific! I welcome and encourage email from authors of the books that I review, and hope to receive more in the future. In particular, John and Graham made a couple of points that I feel are worth mentioning. On the issue of the hundreds of 'novelties' in NCO, Graham pointed out that their use of that term had been quite restrained, and that by 'novelty', they had meant serious and important contributions to theory, i.e., genuine improvements and not just one-move suggestions, for example. If suggestions and 'significant refinements' are added to novelties, Graham estimates that the total of such contributions in NCO rises into the thousands, rather than hundreds. With respect to the novelties themselves, Burgess and Nunn sent me a number of intriguing new examples. Botvinnik Slav fans will find several remarkable contributions in Burgess' section on that opening, for example. I don't have room to show too much analysis in this review, but I can't resist giving one novelty which Nunn mentioned, because it should delight the heart of any chessplayer. In the Budapest Gambit line with 1.d4 Nf6 2.c4 e5 3.dxe5 Ng4 4.e4 Nxe5 5.f4, an interesting idea is 5...Nbc6!?, after which Lalic's new Budapest book, for example, condemns 6.fxe5?! on account of 6...Qh4+ 7.Kd2 Qf4+ 8.Ke2 (he also analyses 8.Kc2 Qxe4+ 9.Kb3 Nd4+ etc.) 8...Qxe4+ 9.Be3 Bc5 10.Qd3 Qxe5! with pressure. Instead, Lalic recommends 6.Nc3! Bb4 7.Be3, although 7...Ng6 doesn't seem bad in that case. What Nunn found was an exercise in geometry: 6.fxe5 Qh4+ 7.Kd2 Qf4+ 8.Kc3! Qxe5+ 9.Kd2!, intending 9...Qf4+ 10.Ke1! Qh4+ 11.g3 Qxe4+ 12.Qe2, thus refuting 5...Nbc6. Marvelous!
Another example illustrates the sort of issues that arise when you're trying to save space. On page 98, footnote 6, is a line which goes 1.d4 Nf6 2.c4 c5 3.d5 e6 4.Nc3 exd5 5.cxd5 d6 6.e4 g6 7.f3 Bg7 8.Be3 0-0 9.Nge2 Nbd7 10.Ng3 a6 11.a4 Ne5 12.Be2 Bd7 13.f4 Nfg4 (a controversial alternative to 13...Neg4) 14.Bg1 Qh4 15. fxe5 Bxe5 16.Qd3 c4 17.Qf3 f5 (pretty much forced for White since move 13) 18.0-0-0 fxe4 19.Qxe4 Qg5+ (thus far Levitt in Informant 66), and here Nunn finds 20.Rd2! Bf4 21.Qd4 with a clear advantage to White. This refutes existing theory, and Nunn says that, since the move 13...Nfg4 was thus apparently unsound, he originally omitted it entirely from NCO! But co-author Joe Gallagher (who wrote a book about the Saemisch King's Indian including this variation, since it transposes exactly) argued that the refutation should be included, and ultimately they did so. I fully agree with Gallagher, since otherwise the reader would be left in the dark. Well, the interesting thing about this story is that, as I was preparing this review, I put the position into ChessBase and started to wonder if there weren't alternatives. To make a long story short, I now think that Black has major improvements here which make his game fully playable, beginning with either 19...Bf4+ 20.Kb1 Rae8 21.Qxc4 Ne5 or, more accurately, 19...Rae8! 20.Qxc4 (20.Bxg4 Bxg4 is messy, but works out nicely for Black) 20...Bf4+ 21. Kb1 Ne5 (this avoids 21.Qd4 in the above line). This is a book review, not an opening tome, but I think that this line looks fine for Black; and if that's true, it means that it's a good thing the line was included! Overall, this is the kind of attention to detail which I talked about in my review, and which does great credit to NCO's authors. After all, it would have been simple to just insert current theory straight out of the Informant and move on.
I'd better get on to some reviews before I forget! I haven't read the first book in the above list very thoroughly, and so will limit myself to some general comments. Hanging Pawns is part of a new series of instructional middlegame books by ICE, including forthcoming titles such as Rook versus Minor Pieces, Compensation for the Queen, Exchange Sacrifice, and the like. Each book will be co-authored by writers including GMs Adrian Mikhalchishin and Josef Dorfman (one of the world's most respected trainers). Hanging Pawns consists of a series of more than 180 hanging-pawn positions, taken mostly from top-flight (and often famous) games. Karpov, who has been the modern genius of these kinds of positions, is appropriately represented by 13 positions, easily more than any other player. There isn't a great deal of general theory, which is mostly confined to a short Introduction with various typical structures and a list of principles for playing hanging pawn positions. I very much like the choice of games and the analysis I have played through so far; but the claim that the book 'systematizes' these positions is not supported, as far as I can tell. In fact, there are no chapters or thematic divisions at all, just one example after another. So the student who really wants to learn is going to have to get out his board and commit himself to working through and absorbing the lessons of each example. Someone looking for easy generalizations won't get much help here; but honestly, generalizations are not the way to get to the delicate truth of this incredibly subtle issue anyway.
GM Mikhalchishin is an experienced and highly-respected GM who can be counted on to analyze such positions accurately and in depth. Unfortunately, the book does not indicate how the authors divided the work, something that I hope will be rectified in the other books of this series. Hanging Pawns is a nicely-bound paper-back with an attractive cover, and an improvement in physical quality over some previous ICE publications. I can recommend this book for anyone who is willing to put some work into tackling a difficult and fascinating aspect of chess. A certain base strength is necessary to follow the annotations, so my guess is that anyone from about 1600 and up could learn a lot from this book. I know that I could.
Four Gambits To Beat the French by Tim Harding, is the latest Chess Digest publication and, like the Silman book I discussed in review #3, it is physically of a much higher standard than works from this publisher were a few years ago. The typesetting and editing are also well done. It seems that editor Roy DeVault is learning quickly; one need only compare the lack of editing in the Gufeld and Stetsko Classical French book I mentioned in an earlier review. Of course, the author also makes a big difference; Tim Harding has been writing for many years (well before the Age of Databases), and knows how to put together a coherent book. He has always been willing to explore unusual variations and ones which are difficult to write about, and Four Gambits is no exception. Harding's particular strength has been to identify areas of theory which other books have neglected (I plead guilty), and that alone has made him a valuable author over the years.
There seems to me, however, to be a problem with the message of this book. At the outset, I'll grant that the book's intended audience is not a sophisticated one, as Harding makes clear in his introduction. The lines of the book are meant to be fun, as he says, even to the extent of sacrificing results for pleasure, or for playing in the 'more relaxed' settings of email and online blitz play. Okay, that sets up a very modest goal for the book, but still, I wonder if readers will be satisfied with what they get. The title does suggest that you can 'beat the French' with the chosen gambit lines, and two of them are rather positively described: the 'murderous' Milner-Barry and the 'IM-killer' Winkelmann-Reimer. Incidentally, as far as I can make out, Harding never describes why it kills IMs or which IMs have died in the process of facing it. Perhaps this is buried in a note somewhere, but all the introduction to the line says is that "not for nothing is it called the 'IM-killer'." Odd.
Anyway, to return to my point, it is not always clear how ultimately effective these four gambits seem to be, especially as they are gambits for White, who at the very least doesn't want to emerge from the opening with a disadvantage. But that's exactly the problem; White can't even seem to equalize against good play! And Harding, a very experienced French player himself, is honest enough about the theoretical value of the lines. Regarding the Winkelmann-Reimer (probably the best of these gambits), he says: "I think the WRG is perhaps objectively unsound...", but that it's a "better practical chance than most of White's sharp alternatives to 4.e5", which is not saying much. In the end, he doesn't offer anything equal for White after 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.a3 Bxc3+ 5.bxc3 dxe4 6.f3 e5 (Huebner's remedy), and in lines such as 6...Nd7 7.Nh3, he doesn't consider 7...Ngf6 8.fxe4 Nxe4 (intending 9.Qg4 Ndf6), which I have played and believe favors Black. As for the Milner-Barry (the second-best of the gambits), Harding seems to concede that Black has a good game after just about every possible move order for White, once you look closely. And, as with the other gambits, he has the odd habit of giving apparently strong moves by Black without suggesting how White should improve, e.g., in the line 1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Nc6 5.Nf3 Qb6 6.Bd3 cxd4 7.cxd4 Bd7 8.0-0 Nxd4 9.Nxd4 Qxd4, he gives 10.Qe2 an '!' (to avoid certain ...Qxe5 lines, which he likes for Black), and then gives a note with the standard remedy 10...f6! leading to a large Black advantage, but without any suggestions for White. He also fails to mention the effective 13...Bc5 in this same line after 10...Ne7 11.Nc3 a6 12.Kh1 Nc6. Finally, he lists but fails to appreciate the strength of some irregular (he says 'inferior') options for Black, such as, in the above line, 10.Nc3 Ne7, which leads to 11.Nb5 Qxe5 12.Re1Qb8 13.Nb5 Bxb5 14.Bxb5+ Nc6 15.Qxd5 Qd6 and Black's position is much better than is indicated by the only game ever cited.
As I say, Harding deserves credit for his honesty. In the French Wing Gambit, 1.e4 e6 2.Nf3 d5 3.e5 c5 4.b4 cxb4, he gives 5.d4 Bd7 6.a3 Qa5! 7.Bd3 Bb5 and simply says "This is a hard line for White to play"; whereas after 5.a3 Nc6 6.axb4 Bxb4+ 7.c3 Be7 8.d4 f6"!", for example, he just says that "White badly needs to come up with something new in this line". There are several other examples throughout the French Wing Gambit, which, incidentally, is just not a very good opening. Finally, in the 'Alapin-Diemer' gambit, beginning with 1.e4 e6 2.d4 d5 3.Be3 dxe4, Harding seems to concede that Black doing well in the line 4.Nd2 Nf6, and he tries to defend positions like 4...exf3 5.Nf3 Nf6 for White; but even on the face of it, this is just a normal Blackmar-Diemer Gambit with the undesirable and passive move Be3 for White.
Okay, I don't want to go on forever, but thought that the reader might like to hear about a few remedies to some of these gambits. Let me also mention some of the good qualities of Four Gambits. Harding includes a number of his own games with both Black and White to introduce various lines, with fun and instructive notes. His annotated games (a lot of them from correspondence play) include a lot of new moves and new theory in these lines (confirming that there's a lot of investigative space in such lines). From Black's point of view, a 1...e6 player will be interested in the refutations and effective answers to a wide variety of main- and side-lines, and to alternative gambits. As I mentioned, Harding plays the French and understands it very well. So this book is a definite contribution to theory, and will contain a lot of games and analysis you've never seen before.
The bottom line is that Harding thinks that the average player can have some fun and fool some opponents with these lines. He also thinks that there's room for innovations and creativity, room which is lacking in most heavily-analyzed theoretical lines. He's probably right on both points, and I think that his recommendation of such lines for Internet, correspondence, and email players makes sense. Just be aware that you won't find a lifelong anti-French system here, and that if you use these systems, you'll be trading the pleasure of some enjoyable attacking wins for the pain of some depressing losses in which you never had anything for the pawn.
Chris Baker's A Startling Opening Repertoire is one of those everything-in-one-book opening repertoires for White. The repertoire begins with 1.e4, and the lines are for the most part meant to 'startle' the opponent, although Baker tosses in a fairly standard (and ultimately harmless) remedy to the Caro-Kann with 3.Nc3, a normal anti-2...Qxd5 Scandinavian solution, and main-line f4 systems against the Pirc and some Modern systems. The rest of the repertoire is an odd mix of extremely safe, mostly passive lines which offer no advantage to White, and a few wild attacking lines (which may offer even less!). Baker describes the systems (inaccurately, in my opinion) as "in the main potentially very aggressive", a wording which doesn't inspire confidence.
Now this book is already very popular, and I won't deny its appeal to the average player. Also, I know how difficult such a book is to write, and Baker has worked hard to provide the reader with ways of playing against even the most trivial and inferior continuations by the opponent. Importantly, he has played many of these systems himself, and offers a great deal of his own original analysis in their support. But I do think that the book's approach and choice of variations share the drawbacks of so many other repertoire books. Essentially, the idea is that you won't have to work very hard to learn these systems, and supposedly, your opponent will tend to be surprised by your choice of lines (although I think that this is less likely than the author would have us believe, at least against any opposition for which the choice of opening is relevant). By avoiding mainstream theory, one can 'throw the opponent on his own resources' and gain the upper hand.
In general, I disagree with this approach. I feel that students should be learning lines which they'll want to keep for a long time, if not for life. Ideally, they will take up other systems as they go along, and gain flexibility thereby, but not abandon their old ones. How does that apply to the lines in Baker's book? Well, I don't doubt that many of the people reading this have already tried the first system in his book, the Max Lange Attack and related lines. This approach has practically disappeared at the professional level, but it has been a favorite of students of mine for many years. Without exception, however, they have abandoned it as time went on. Of course, it's also not going to be very 'startling', since the Max Lange is one of the first things a 1...e5 player studies. And apart from the fact that I don't like the move order which Baker recommends after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5, namely 4.0-0 Nf6 5.d4 Bxd4 6.Nxd4 Nxd4 7.f4 d6 8.c3 Nc6 9.f5, he neglects to say how to respond to the simple 4...d6. Well, I suppose that, with some patching up, one could use this as a surprise weapon, but the book is not off to an auspicious start. The next chapter is the aforementioned Cochrane Gambit against the Petroff Defense, about which you know my opinion, and again, it is a system which is unlikely to stay in one's repertoire very long. There follows a very well-written chapter on the Philidor, including a section on 1.e4 e5 2.Nf3 d6 3.d4 Nf6 4.dxe5. That's fine. But when I went to check the currently-popular move order 1.e4 d6 2.d4 Nf6 3.Nc3 e5 (which can also arise after 1.d4 d6), Baker gives 4.dxe5 dxe5 5.Qxd8+, and concludes that 'it is very hard for White to prove a substantial advantage' (in fact, Black gets equality in his lines), and that 'White will have to consider going into the main line of the Philidor', which, he admits, is outside the scope of the book. Hmm.
Baker gives a series of 'safe' lines against the Sicilian. For some reason, he devotes an amazing 35 pages just to the Rossolimo Variation (1.e4 c5 2.Nf3 Nc6 3.Bb5), much more than the space given to his anti-2...d6 line (2.Nf3 d6 3.d4 cxd4 4.Qxd4) and anti-2...e6 line (2.Nf3 e6 3.Nc3 Nc6 4.Bb5) combined. Although the Rossolimo has some bite and has been widely employed at the top levels, the other two systems are harmless for a variety of reasons, and 4.Qxd4 in particular has suffered setbacks in the past few years, after a brief period of popularity. With the lack of ambition inherent in these systems (such that their surprise value is not often of much use), one has to wonder "Why play 1.e4 at all?", or "Wouldn't 1.e4 c5 2.d3 be easier, also equalizing?". Similarly, against the French Defense, White plays the innocuous 1.e4 e6 2.Nf3 d5 3.Nc3, when Baker's comment that "I have yet to be convinced that Black can demonstrate a clear route to equality" could just as well be replaced by a statement that none of Black's established answers fails to achieve complete equality! Just for example, in one equalizing line given by most books, 3...d4 4.Ne2 c5 5.c3 Nf6, Baker suggests the unusual 6.Ng3 Nc6 7.Bc4, giving no examples and one short analytical line with extremely cooperative play by Black. White has only played this way once in my databases (and lost), perhaps because Black has several effective answers, for example, 7...a6 and if 8.d3, 8... dxc3 9.bxc3 Na5 (or 9...b5 10.Bb3, and, among other moves, 10...Ra7 intending ...Rd7) 10.Bb3 Nxb3 11.axb3 b6. The position after 5...Nf6 occurs 20 times in Megabase 99, by the way, with White scoring a miserable 3 wins, 11 losses, and 6 draws.
This general pattern of ambitionless play with White is pursued in the Alekhine Defense, 1.e4 Nf6. Now there are many promising paths for White in this opening, but Baker gives 2.Nc3 (is this startling?), and after 2...d5, 3.e5. He then, incidentally, neglects some lines Alekhine-expert Graham Burgess has suggested for Black. But there's a more important problem: What if Black plays 2...e5, transposing to openings like the Vienna, which are out of Baker's repertoire? To this, he simply says that 2...e5 is 'unlikely', and fails to suggest anything! Again, it seems that the desire to suggest a risk-free line predominates, even in a case where promising alternatives were so easy to find. Such an attitude seems to me more appropriate for Black systems than for White.
Remember: I'm just skimming through the book and noticing these problems, which is not a good sign. On the other hand, there's a reasonable answer to these sorts of criticism: A Startling Chess Opening Repertoire is a book for the club player, and simply tries to give him or her some systems, mostly not too difficult to learn, to use in selected contests on a popular level, or to incorporate into one's repertoire as a flexible second line. The details, one could argue, aren't that critical in such a context. Baker's work is also valuable as a theoretical contribution; to his credit, he openly shares his analysis and a number of strong, original ideas in these openings. As long as the potential reader views the book in this light, it might be useful for the right audience. And, as mentioned earlier, some of Baker's suggested lines could be legitimate long-term weapons as well (e.g., versus the Pirc, Modern, Scandinavian, and some irregular defenses). The danger I see, however, is that the reader will believe that he is getting more, i.e., the 'surprising' and 'aggressive' systems the book seems to promise. As a whole, I find this book a bit disappointing, for the same reasons that most 'complete' repertoire books disappoint. After all, it isn't hard for an author to suggest lines which equalize for White; the trick would be to describe ones which offer lasting advantages, however small, and interesting prospects. Unfortunately, that probably can't be done in a single volume, at least with sufficient theoretical underpinning, since it would require suggesting lines which strong players regularly employ. In conclusion, I find this an honest and original effort, but I wouldn't want my students to use most of the recommended lines.