Part Two of A Long Review of Nikita Vitiugov's The French Defence: A Complete Black Repertoire
Saturday, July 31, 2010 at 4:26PM (For part one of my review of Nikita Vitiugov's The French Defence: A Complete Repertoire, click here.)
In the first part of the review, I mentioned some of the book's pluses and addressed some less-than-compelling criticisms of the book. In this part, I'll examine some criticisms that seem to me more serious, and will compare some of Vitiugov's findings with those given in Andreas Tzermiadianos's excellent 2008 book How to beat the French Defence (he advocates 3.Nd2) and with the points of overlap with John Watson's suggestions for White in his Dangerous Weapons: The French volume.
More ChessPub objections:
(1) In the line 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Bd3 Nbc6 8.Qg4 Qa5 9.Bd2 c4 10.Be2 0-0 11.h4 f6 12.f4 Qa4 13.Bd1 Qb5 14.Bc1 Qa5 15.Bd2 Vitiugov gives, and tacitly endorses, the draw by repetition that occurred in the game Short-Shulman, Dhaka 1999. That game concluded 15...Qb5 16.Bc1? Qa5 17.Bd2 and draw agree. However, "dom" rightly notes that 15...Qb6 is a serious improvement (and a reason why 11.h4 is an error). If White plays 16.Bc1 now he gets hit by 16...fxe5 followed by 17...Nxe5! after either recapture - probably with a winning advantage.
In one sense this isn't a big deal. First of all, White doesn't have to play 16.Bc1, though he's worse even after better moves. Second, even in Vitiugov's variation, the line is still a theoretical success for Black. And of course any chess book with non-trivial content is bound to have errors. So this is far from the end of the world. It does point to something I noticed more than once, however, and it's that Vitiugov very often seems to trustingly quote games without checking them. There is independent analysis in the book, but the number of uncommented-on game references seems to me a dubious sign.
(2) In the Winawer with 7.Qg4, derdudea found serious improvements for both sides in one of Vitiugov's lines. After 7.Qg4 cxd4 8.Qxg7 Rg8 9.Qxh7 Qc7 10.Ne2 Nbc6 11.f4 Bd7 12.Qd3 dxc3 13.Qxc3 Nf5 14.Rb1 d4 15.Qd3 0-0-0 16.Rg1 Na5, 17.g4 is his main move but he also addresses 17.Rb4. From there we have 17...a6 18.g4 Nh4 (he covers 18...Ne3 parenthetically, citing Sharma-Riedel, Bad Wiessee 2009 through move 26, when White is winning) 19.a4 f6 20.exf6 e5 21.f7 Rgf8 etc., following the game Steflitsch-Poldauf, Chalkidiki 2002 which was eventually winning for Black.
As derdudea notes, however, 21.fxe5! is winning for White, and it's not hard to discover. It was played in one OTB game, admittedly not one with high-rated players, but it was also played in a pretty high-rated correspondence game. (In case some of you think this is hard to access, it shows up automatically in ChessBase's online database, which you can quickly and search in ChessBase.) And even if you miss it in your research, Rybka recommends the move instantly and gives it a big score. (+1.5 in a second or two, and it fluctuates from there to almost +2 thereafter.) To find it, it's enough to flip on the engine and give your eyes a chance to focus on the evaluation box.
As for 18...Ne3, mentioned above, it's the right move. After 19.Bxe3 Bb5 20.Qd2 dxe3 21.Qxe3 Qxc2 22.Nd4, Black should not play Riedel's 22...Rxd4? but 22...Qxh2, which is complicated but about equal after 23.Nxb5 axb5.
In sum, we have at least some initial indications of gaps in his research and a failure to use, or at least an underuse of, his engines.
Tzermiadianos vs. Vitiugov:
As they overlap in three chapters - 3.Nd2 c5, 3.Nd2 Be7, and 3.Nd2 dxe4 - there's quite a bit of material here. If I reproduce everything from my computer file it will take a couple of hours for me to finish the review, so I'll briefly summarize my findings here. Tzermiadianos (and White) get the better of the argument in the 3...c5 and 3...Be7 lines, while I found Vitiugov's judgment more reliable in the 3...dxe4 variation.
In the first two cases, it seems that Tzermiadianos has investigated more deeply and has some important original ideas. That's part of the story, but there's a second part as well. Vitiugov has the very odd habit of offering some variation for White that gives him an advantage, and instead of offering improvements for Black or at least some help via punctuation, he simply leaves the reader with some disastrous evaluation and moves on to the next thing.
The third part of the story is that Tzermiadianos engages in research. His bibliography includes all the monographs you'd expect, plus the Informant (all issues as of that time), Chess Today, New in Chess (both the magazine and the yearbooks), the TWIC and Mega databases, ChessPublishing.com, correspondence databases and Greekbase 2007. How about Vitiugov? Well, he looks at Khalifman's books (ditto Tzermiadianos for the Tarrasch volume), Psakhis' books (ditto again), Sveshnikov's Win against the French (not relevant to 3.Nd2), and Moskalenko's The Flexible French (too recent for Tzermiadianos). And that's it.
Now you might say "Big deal! Andreas Tzermiadianos is an ordinary IM. His rating is 2425. That's nothing to sneeze at, but Vitiugov is 2722! He's #23 in the world! Why should he care what all these relative fish think? Would you, Dennis, care what a low expert had to say about the openings you knew best?"
I think this objection is deeply confused. If there's a Vitiugov-Tzermiadianos match in the cards and I have to bet, my money's going on the 2700. But that's not what we're evaluating here. First of all, Tzermiadianos and his sources are running engines, and if I have to bet on a Rybka-Vitiugov match, my money is NOT going on the 2700. Second, it's pretty clear that most of the time Vitiugov isn't using an engine at all (or if he is, he's not telling us). The vast majority of the time, he's happy to supply a game citation with no further commentary or punctuation. That's something that doesn't require 2700-level talent. Third, let's assume Vitiugov's results are all the product of at least cursory analysis and computer checking on his part. While I'm sure that he would be more reliable, minute for minute, than Tzermiadianos in his findings, the latter is using other sources as well. Even if Vitiugov is better than each of them singly, together they are going to find things that he doesn't. Whether it's because they're spending more time, or leading each other to investigate more deeply, or because each of them as a special little spark for each line taken individually, the accumulated product of all these good players working together will often exceed the solo product of a great player.
Maybe it's difficult for Vitiugov to access many of those sources. It's possible, though I'd prefer a more innocent explanation like the first one, that he doesn't bother with them out of a sort of snobbery or laziness. Whatever the case, I think it's a mistake for ChessStars not to give him access to those other works and strongly insist that he use them. It's not just that others could have something valuable to say; it's also good business. Let's say that some poor French player has been taking a beating from a player at his local club using the Tzermiadianos book. Eager for help, he rushes to buy the Vitiugov book, only to discover that the older book has the fuller, more up-to-date analysis in several places where they overlap. If I'm that club player, I'm pretty annoyed, and maybe I get on ChessPublishing.com's forum and write about what a lousy book this is.
Now, it's not a lousy book, but it's nowhere near as good as it could have been. And I will say this: on occasions where it's clear that Vitiugov has done his own thinking and isn't just supplying some game reference, his judgments hold up. Unfortunately, he doesn't always supply verbal or analytical defenses of those judgments, but they still proved to be reliable where I checked them. One such example (see the game file for full details) comes in the 3...Be7 Tarrasch. Both Vitiugov and Tzermiadianos cite a 2007 game Rozentalis - N. Pert, where the latter offers an interesting gambit that gave him good play. As usual, Vitiugov just gave the game score without any punctuation, while Tzermiadianos gave the game continuation, along with a couple of improvements for both sides. He's quite enthusiastic about the line, but Vitiugov tells us he doesn't trust it. He's right: 14.d6! (instead of Rozentalis's 14.Ne4) is very strong, e.g. 14...b6 15.Ne5 bxc5 16.Ndc4 Ba6 17.Rd1! with a frankly miserable position for Black. Vitiugov's judgment is right, but why not just give the line and not bother with the full game score?
Watson vs. Vitiugov:
I only spotted two areas of possible overlap here between the two books. One of John Watson's "dangerous weapons" against the French is 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.exd5 exd5 5.Qf3. Whatever its merits, we can't compare what he has to say with Vitiugov's analysis, because the latter doesn't so much as consider 5.Qf3.
Round 1: Watson, by default.
Second, Watson tries to rehabilitate the variation 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.Qg4. The key position comes after 5...Ne7 6.dxc5 Nbc6! (my exclam) 7.Qxg7 Rg8 8.Qxh7 d4 9.a3 Qa5 10.Rb1 dxc3 11.Be3 Bd7. Vitiugov's acerbic comment: "it is sufficient to calculate the number of developed pieces of both sides in order to evaluate the position correctly." No further analysis is given; again, for better or for worse, he is trusting the reader. Watson's analysis continues, and starts with 12.Nf3 cxb2+. Once again, Vitiugov's judgment is correct, and with 12...0-0-0 instead Black has a very nice advantage.
Round 2: Vitiugov.
The Bottom Line:
(Yes, I know you've had to wait a while for this. Don't feel bad, it has taken me a lot longer to prepare and write this post than it did for you to read it.) By now you should have a pretty good idea about this book. It offers a wide-ranging repertoire, but it's sometimes superficial and often rather clipped in its presentation. His judgments look pretty reliable, but they're often in need of further elaboration from him (or computer work from us). His game references are generally sensible, but he misses important games sometimes and rarely consults with what contemporary analysts are doing. In sum, it's a good guidebook, but not so good on the details. If you can deal with the need to fill in the gaps and check his work, then it's worth your time; if not, then don't get it. (And if you play the French or the Tarrasch against it, you should certainly get Tzermiadianos' book!)
Finally, you can replay all the lines I've discussed in these two posts and much more, here.
Reader Comments (6)
Many thanks for this very helpful review.
awesome reveiw dennis. I have the book already but still learned a couple of things about it from reading your observations.
Thanks for this excellent review. While I was a little bit harsher especially on the author´s attitude, I have to admit this is a very balanced review with great substance, doing him more justice than my own words did.
I've mentioned this on the Chess Publishing forum, but unless I am being exceptionally stupid, the Contents pages don't seem to list 3.Nc3 Nf6 4.Bg5. Which is a bit odd. I understand that the recommendation is 4...dxe4, but I wouldn't know this from the Contents.
Yes, you are correct that it's not mentioned in the Table of Contents. On the first page of the Steinitz Variation section (page 204) he writes this: "I will repeat that after 4.Bg5, I recommend to Black to enter a favourable version of the Rubinstein variation with 4...dxe4 5.Nxe4 Nbd7." So when you have the book there's no real mystery, though it's a little confusing when you look at the ToC. But then again, what's he to do there, when it has already been covered under a different move order?
Yes, I agree, it's tricky. Although trickier still is the situation of the person contemplating buying the book but not having it their hands - a normal situation these days, of course - and not having a clue where 4.Bg5 has gone. Perhaps a short bracketed note in the ToC - (4.Bg5 - see 3...dxe4 4.Nxe4 Nd7 5.Nf3 Ngf6 6.Bg5) - might have helped.
One of the hardest problems when writing of producing any book, I think, it's working out what other people might not see, or might not understand - precisely because you can see it, and understand it, yourself.