A 2700+ GM Plays the French

[Dan Scoones]

Re: A 2700+ GM Plays the French

Gary,

Is the Winawer really as bad for Black as you think it is? In this age of ChessBase it's possible to get a quick answer, at least in terms of results.

First I searched Informants 1-100 for all of the C15-C19 games. White scored 69% and a +100 performance rating, while Black scored 31% and a -100 performance rating.

Then I searched the last five years of MegaBase 2007 (my latest copy) for the C15-C19 games of players rated over 2200. White scored 57% and a +20 performance rating, while Black scored 43% and a -50 performance rating.

Because Informant games are selected for strategic interest, the sample is not necessarily representative of the entire population of games. This is borne out by the differing results from MegaBase 2007. In that sample Black scored slightly less than his standard 46%, but his performance rating was quite acceptable at only -50 points. On the aggregate Black was somewhat lower rated than White, so the result is within expectations.

Inconclusive, I know, but still indicative. The French Winawer is an extremely sharp opening but it's not ready for the slag heap by any means.
If you are one of those players who has an easy time against it, count yourself fortunate and give yourself the title of Terminator (a la Moskalenko)!

Regards,

Dan

[Hugh Siddeley]

Another Winawer, this time by a 2748. Easy draw for Black in 21 moves

(3) Svidler,P (2726) - Grischuk,A (2748) [C19]
4th FIDE GP Nalchik RUS (13), 29.04.2009

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Qg4 0-0 8.Bd3 Qa5 9.Ne2 cxd4 10.Bg5 Ng6 11.f4 Nd7 12.Bxg6 fxg6 13.Qxe6+ Rf7 14.Qe8+ Rf8 15.Qe6+ Rf7 16.0-0 Nb6 17.Qe8+ Rf8 18.Qe7 dxc3 19.Bf6 Rf7 20.Qe8+ Rf8 21.Qe7 Rf7 1/2-1/2

[Hugh Siddeley]

Breyer bust?

I think someone came up with a bust for the Ruy Lopez Breyer which is 9. ... Nb8. I think the key move is somewhere around 16 or 17 where white plays a4. I'd have to find my notes on this and can't recall where I tucked them away.

Maybe you should hire yourself as Ivanchuk's second because he doesn't seem to know about the Breyer bust.

(7) Ivanchuk,V (2746) - Kamsky,G (2720) [C95]
4th FIDE GP Nalchik RUS (13), 29.04.2009

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 d6 8.c3 0-0 9.h3 Nb8 10.d4 Nbd7 11.Nbd2 Bb7 12.a4 c5 13.Bc2 c4 14.b3 cxb3 15.Bxb3 d5 16.Nxe5 Nxe5 17.dxe5 Nxe4 18.Nxe4 dxe4 19.axb5 axb5 20.Rxa8 Bxa8 21.Qg4 Kh8 22.Qh5 Bd5 23.Rd1 Bxb3 24.Rxd8 Rxd8 25.Bg5 Bxg5 26.Qxg5 Rd1+ 27.Kh2 h6 28.Qe7 Bc4 29.Qb7 Re1 30.Qc8+ Kh7 31.Qf5+ Kg8 32.Qc8+ Kh7 1/2-1/2

[Gary Ruben]

Re: Breyer bust?

Maybe you should hire yourself as Ivanchuk's second because he doesn't seem to know about the Breyer bust.

I think I prefer 25.Qe2 retaining the bishop.

Anyhow, the analysis was published and I didn't any correction or retraction.

I take it you like white's play here. If you see Chucky, remind him a queen alone can not make a child. :)

[Gary Ruben]

Re: Another Winawer, this time by a 2748. Easy draw for Black in 21 moves

(3) Svidler,P (2726) - Grischuk,A (2748) [C19]
4th FIDE GP Nalchik RUS (13), 29.04.2009

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Qg4 0-0 8.Bd3 Qa5 9.Ne2 cxd4 10.Bg5 Ng6 11.f4 Nd7 12.Bxg6 fxg6 13.Qxe6+ Rf7 14.Qe8+ Rf8 15.Qe6+ Rf7 16.0-0 Nb6 17.Qe8+ Rf8 18.Qe7 dxc3 19.Bf6 Rf7 20.Qe8+ Rf8 21.Qe7 Rf7 1/2-1/2

I don't see the big deal here. Svidler played a draw line against a STRONGER player.

It looks like nice safe chess.

[Hugh Siddeley]

Re: Another Winawer, this time by a 2748. Easy draw for Black in 21 moves

22 points makes Grischuk the "STRONGER" player? I bet you wouldn't tell Svidler that to his face, the many times Russian champion from St. Petersburgh.

[Gary Ruben]

Re: Another Winawer, this time by a 2748. Easy draw for Black in 21 moves

22 points makes Grischuk the "STRONGER" player? I bet you wouldn't tell Svidler that to his face, the many times Russian champion from St. Petersburgh.

Do you think Svidler doesn't know he played "sister kissing" chess?

That game is 21 moves of nothing.

[Alan Baljeu]

Re: A 2700+ GM Plays the French

I used to play the classical French (3. Nf6) as black. Then I noticed my results were about 35% when other openings gave me 50%. I switched to the Rat, and am much happier for it.

[Egidijus Zeromskis]

Re: A 2700+ GM Plays the French

A 2700+ GM Plays the French in the Canadian Open
D. Kagramanov - A. Shirov :)

[Kevin Pacey]

Re: A 2700+ GM Plays the French

...Winning is, quite frankly, everything.

Maybe, but naturally a player losing a won position is much less relevant to an opening's reputation, particularly so in cases where the win is easy and a gross blunder is made (not that that happened to Kamsky).

I've been checking out the Najdorf lately. It's winning percentages are excellent for Black, as is it's overall win percentage, even at the highest levels. As always, it seems, there is a need for qualifications. Most of the time people play for a win, especially in the Najdorf (with either colour). If White wishes to draw, 6.Bg5 against the Najdorf may pose theoretical problems if Black wants to win, based on what I've seen. There always seems to be a draw or drawish option available to White if he wants it, unless Black goes into an inferior line (though I've seen some ideas for Black in certain sub-variations that need testing).

Kasparov once went all the way down a theoretical line that lead to a perpetual check as Black vs. Vallejo Pons in a Najdorf Poisoned Pawn (Moscow 2004). Maybe he was saving a winning-attempt novelty for another game, a draw would do, or he was just testing his lower-rated opponent's memory. Who knows? Personally, I've looked for games where I know a player must win (e.g. a Candidates match) and note what defence they chose if they were Black. Gurevich once tried to draw Short in the last game of a such a vital match using an Exchange French. It didn't work.

Also, if you compare the Winawer to the Najdorf, Black scores a moderately lower win percentage with the Winawer in my databases. However, also in my databases, at the highest levels the 2.c3 Anti-Sicilian sees a comparable moderate win percentage for Black to any variation of the French where Black is out for the win. Granted, this needs to be taken with a grain of salt, since the number of modern high level rated games with either opening is statistically small, but it gave me pause for thought.

[Gary Ruben]

Re: A 2700+ GM Plays the French

Maybe, but naturally a player losing a won position is much less relevant to an opening's reputation, particularly so in cases where the win is easy and a gross blunder is made (not that that happened to Kamsky).

I've been checking out the Najdorf lately. It's winning percentages are excellent for Black, as is it's overall win percentage, even at the highest levels. As always, it seems, there is a need for qualifications. Most of the time people play for a win, especially in the Najdorf (with either colour). If White wishes to draw, 6.Bg5 against the Najdorf may pose theoretical problems if Black wants to win, based on what I've seen. There always seems to be a draw or drawish option available to White if he wants it (though I've seen some ideas for Black in certain sub-variations that need testing).

Kasparov once went all the way down a theoretical line that lead to a perpetual check as Black vs. Vallejo Pons in a Najdorf Poisoned Pawn (Moscow 2004). Maybe he was saving a winning-attempt novelty for another game, a draw would do, or he was just testing his lower-rated opponent's memory. Who knows? Personally, I've looked for games where I know a player must win (e.g. a Candidates match) and note what defence they chose if they were Black. Gurevich once tried to draw Short in the last game of a such a vital match using an Exchange French. It didn't work.

Also, if you compare the Winawer to the Najdorf, Black scores a moderately lower win percentage with the Winawer in my databases. However, also in my databases, at the highest levels the 2.c3 Anti-Sicilian sees a comparable moderate win percentage for Black to any variation of the French where Black is out for the win. Granted, this needs to be taken with a grain of salt, since the number of modern high level rated games with either opening is statistically small, but it gave me pause for thought.

I think if they ever get tablebases which solve chess, the Winawer will be completely lost. The main thing is many players don't know how to beat it and todays databases help players avoid it with those who do know to beat it. It's important to play material your opponent has had problems with, to see if he's learned it.

The Najdorf is a bit different. I think 6. Bg5 is almost a bust for the Najdorf. The fly in the ointment is the poisoned pawn variation.

Usually, though, a strong player is safe playing almost anything against the weaker players.

I noticed in Correspondence that the Sicilian Dragon is getting hard to defeat.

[Kevin Pacey]

Re: A 2700+ GM Plays the French

I think if they ever get tablebases which solve chess, the Winawer will be completely lost. The main thing is many players don't know how to beat it and todays databases help players avoid it with those who do know to beat it. It's important to play material your opponent has had problems with, to see if he's learned it...The Najdorf is a bit different. I think 6. Bg5 is almost a bust for the Najdorf. The fly in the ointment is the poisoned pawn variation...I noticed in Correspondence that the Sicilian Dragon is getting hard to defeat.

I imagine the Winawer is solid enough in many cases that Black would have more than one sub-variation that might to lead to a draw in a tablebase... however the Winawer Poisoned Pawn might not be one of them. Either side might be losing, though I'd suspect White might have a winning defence in a tablebase. It may be the same way for the Najdorf as a whole, with the English Attack and 6.Bg5 possibly being busts/losses. You and I are both speculating, but the Winawer strikes me as more solid than the Najdorf just based on its typical closed static pawn structures, where Black perhaps usually has more time to brace for any onslaught than in a Najdorf. That's if White can even establish a sustained initiative.

Databases are less useful in many cases at club level, or during events where many players have few of their games as database entries. Even for those players with many entries, it depends how recent their games are whether a player may have learned from his losses yet.

The Dragon 9.0-0-0 Yugoslav Attack has a fearsome reputation in theory and (in the past) in high level otb practise. I've seen some ideas for Black, but the high level stats aren't encouraging. There haven't been many 2600+ vs. 2600+ Dragon games in a number of years according to my databases. The Dragon strikes me at least at a glance as more logical than the Najdorf (5...a6 doesn't help develop a piece). However I wouldn't be totally shocked if a monster tablebase established that certain key lines in either case may lead to a violent or barren draw fairly early in the middlegame.

[Alan Baljeu]

Re: A 2700+ GM Plays the French

I think if they ever get tablebases which solve chess

Technicality: A tablebase is a database containing all possible positions relative to some ending. 7 piece tablebases contain gigabytes of data. The entire universe couldn't hold 32 piece tablebases (unless said universe is much bigger than scientists currently suppose). :)

the Winawer will be completely lost. The main thing is many players don't know how to beat it and todays databases help players avoid it with those who do know to beat it. It's important to play material your opponent has had problems with, to see if he's learned it.

Can you give a hint? Is the winning move 3. e5? a3? Qh5? :)