Here is my third and final consecutive posting to this topic and it is about the Botvinnik-Variation.

Let us assume there is an opening variation where you reach a position that is mate in 15. You are winning! 

The problem is that you have to find 15 only moves and if you make a single mistake you are automatically lost. Even worse is that your first moves are not forced which means your opponent has a whole variety of options at his disposal.

Would you play this variation?

Here are twin games that made history. After Kasparov won the first game an improvement at move 30 was suggested in joint analysis together with other GMs. Dorfman was watching this analysis and he tried it in the very next game. Kasparov had to find a unique sequence of only moves in order to win, which equals not to lose in this position - but of course he had analyzed it at home.

http://www.chessgames.com/perl/chessgame?gid=1069914
http://www.chessgames.com/perl/chessgame?gid=1069915

While the Botvinnik may or may not be a theoretically perfect choice for white, I find it ridiculous to suggest it in such a format. I would reconsider this statement if he had gone for a sideline (early a4 for instance), but not if he wants to play the mainline.

Dink Heckler wrote on 03/08/12 at 13:22:11:


I have to agree with this assessment.

The book is a collection of games with very few comments. As annotation he usually gives computer lines that go like 10 moves deep and that's it. The repertoire is more or less a random collection of "good" computer lines.

His assessments remind me a bit of Hans Berliner and his claims that he refuted the Grünfeld and so on. Such strong statements based on one mainline with everything else ignored, seem to be a bit vague to me.

His line in the KID that he considers to be a "very nearly refutation of the Nc6-line" ends with a drawn endgame. That's as very nearly won as it gets.

The really funny thing is that he evaluates positions based on Kommodo and Houdini and if both engines disagree, he tries to play referee like "I would like to play white here". That is ridiculous! Basically the position is unclear, nothing more and nothing less.

Finally I have some issues with the repertoire supposed to be "user-friendly". He does recommend the Botvinnik-System against the Semi-Slav which is about as tough for a user as it gets. This is the equivalent of recommending Bg5 against the Najdorf and it is very far from user friendly. You are basically supposed to play a totally unclear position where every non-computer move loses on the spot. Besides that, his suggestions are also a bit dubious in this line. It goes like this 

Quote:

So basically nobody knows what is going on, but if you just put a pawn on h4 which may or may not queen 15-20 moves later, you should have the advantage.

Let's not forget that his line agains the Benko is nothing but a joke. Just accept it, play a4 and consolidate your extra pawn. If it was that easy... 
It is rather obvious that neither he nor his engines understand the foundation of the black compensation.

This book reminds me of my childhood. As a kid I memorized lots of "winning"-lines and if my opponent deviated from what he was supposed to do, I lost since I had absolutely no clue what was going on.

Basically the book can be summed up in:
As white play the Qc2-Nimzo, the Ne1/Be3-line against the KID and the Botvinnik against the Slav. With black, go with the Breyer and the Grünfeld. Don't bother with variations. The computers say, that you should be ok.

To be fair, those holes dont seem very threatening to use an understatement. With that Vienna line, black can just play for something like Nc6, d6, 0-0 and Nd4 in some order.

Ametanoitos wrote on 03/19/12 at 14:15:11:


Sure, but White could also play 4.dxc5 intending 4...Qa5 5.a3 Qc5 6.b4. Not that this should give White any edge but it would be nice to have something prepared against it.

I noticed today that 1.d4 Nf6 2.Nf3 g6 3.Nbd2 isn't covered in Kaufman's book. Do you think Black should accept a transposition to a Pirc with 3...Bg7 or go into the 3...d5 4.e3 Bg7 5.b4 variation? 3...Bg7 4.e4 d5 a la the Torre doesn't seem as strong without the moves Bg5 and ...0-0 included.

heavypieces wrote on 03/10/12 at 22:58:47:

Looking at your post of 3-8-12, which is the only post of yours that I find in the other thread, it does not seem to mention the book at all but does seem to concern itself entirely with what systems are good for lower-rated players. I am sorry if I mistook its meaning, but reading it again, Ithink it belongs there, not here.

We were on two topics in one thread, you know? And I was trying to correct it.

The thread split caught me out, so I put my review in the wrong thread...

<ctrl-c, ctrl-v>

Got the book yesterday; only looked at the White rep, but looks like rubbish, frankly.
Repertoire seems like some thought has gone into it, but that's where it ends.

Problems: too much focus on sidelines vs mainlines - for example, he could have hacked down most of the sundry Benoni coverage into two or three paragraphs, and devoted a bit more attention to the Benko (which merits only two or three paragraphs).

Relatedly, what's the point of analysing out random non-thematic endings and stinting on opening coverage.

Thirdly, he seems to get distracted by wrinkles. So, he has an idea for a repertoire, but then gets waylaid because White has an interesting idea on move 12 of some line played in a recent game, which seems to give promising play. So that gets lumped in. 
You can't organise a repertoire for amateurs on the basis of the latest wrinkle. Next month, there'll be another one and then what? Reading a Kaufman book is like taking your cat for a walk.

When it comes to the Grunfeld, I agree that it is a tough system to learn because of the multitude of pawn structures that can arise. But you can say the same about the Nimzoindian as well, even though it is has quite often been recommended even for sub 1500 players. Both openings are theory packed, but I think that a big mass of theory is actually not such a problem for club players.

Maybe it depends on your attitude to opening theory, but looking at the games played between sub 2200 players, I am quite sure it doesn't matter much what you play as long as it isn't outright bad. Because if there is a lot of theory, usually the people that play the lines won't remember much of it anyhow. If you do understand your system, in terms of typical structures, middle- and endgames, to me it seems like this matter much more (doh, sorry for that cliche). More to the point, though, perhaps lots of theory is good, because the likelyhood your opponent will follow theory is rather low?! On top of this, whatever advantage you get (often pretty small and hard to exploit at a certain level), the games often swing a lot due to other things than the opening. Additionally, maybe the reason there is a lot of theory is because there are many plausible ways to play that particular opening?

Above around 2200, I think it is more likely you get punished for not knowing theory, but still many games are decided despite not following theory. 

A an anecdote, I had a discussion with a player on my team, ranked more than 2300, and he asked me (u1800) about the theory, Marin's GM5 in this case, because he couldn't remember exactly when to sac a pawn at early moves in one of the more known line. Still, he's been playing English set-ups way longer than I have, and I think I have also seen him play a game in the same style, but not following up correctly.