Saturday, September 22, 2007

I'm hiding in the bushes because I'm afraid of the c file

So I lost. I was sad to semi-ruin it for New York, but I enjoyed the game quite a lot and felt that I played reasonably, at least most of the time.

In fact, my sadness over losing didn't last long because Pascal's endgame was so incredible. He appeared absolutely calm and relaxed the entire time, and the combination of that and the fact that I had no idea how I would even start trying to win the position, made me feel like I was watching something amazing, something really far above anything I would ever be capable of.

A few days before the match I looked up my previous Sveshnikov games in Chessbase to try to imagine what Chris would see when he sat down to prepare. Out of six games, I only won one, and that's when I was white. OK, I played up mostly, but still, a record of +0 = 1 -4 isn't easy to rationalize. It occurred to me that I should switch to the accelerated dragon or something like that --something where if you make one mistake you aren't hopelessly, depressingly, strategically lost -- but the truth is that I should just learn to play better.

(173) Williams,Chris - Vicary,Elizabeth [B33]
USCL 17.09.2007

1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 e5 6.Ndb5 d6 7.Bg5 a6 8.Na3 b5 9.Bxf6 gxf6 10.Nd5 f5 11.Bd3 Be6 12.0–0 Bxd5 13.exd5 Ne7 14.c3 Qd7

This move order is an idea of Rogozenko's, described in his excellent book The Sveshnikov Reloaded (page 294). More normal is [14...Bg7 15.Qh5 e4 16.Bc2 0–0 17.Rae1 (or 17.Rad1) ]. Chris played 15.Qh5 -- this is not supposed to be the most accurate response, because black can gain some time attacking the queen, either with Rg8-g6-h6 or sometimes Ng6-f4. I should mention that while this line with 14...Qd7 15. Qh5 looks better than normal for black, (and I think was for me in this game) the actual results at the top level are heavily skewed in white's favor. The basic plan for black here is Rg8-g6, Bg7, Ke8-f8-g8-h8 Rag8.

According to Rogozenko, instead of 15. Qh5, white should play on the queenside with [15.Nc2 Bg7 16.a4 Rogo gives 16...e4 here as the best move for black, but I think he's just following the one high level game played in the line. When I was preparing this I started looking at ways to delay ...e4, since I recently learned (after playing the Sveshnikov in blissful ignorance for maybe 10 years) that black actually doesn't want to play ths move. Generally she has to in the mainlines after Qh5 in order to blunt the light-squared white bishop, and ...e4 does have the advantage of freeing e5 for the black bishop or knight. But it also fixes the black pawn structure in an unpleasant way, facilitating white's undermining moves f3 and g4 and creating squares for the white knight on d4 and f4. Additionally, getting the bishop to move off the d file means the white queen defends the d5 pawn.

So I decided I wanted to castle here instead. (16...0–0 Now 17.Qh5 is no longer possible, because after 17...e4 18.Be2 the d pawn hangs: 18...Nxd5

This is happening because the white knight is already on c2. If it had been on a3 still, the bishop could have retreated to c2, and on Nxd5 recaptured Bxe4, owing to the pin on the fifth rank. Instead of 17. Qh5, white can play 17. Nb4 or maybe 17. axb5 first. Black will try to play a quick ...e4 and ...f4 (It's not the same at all to play ...e4 if you can follow with ....f4, because you can cover the light square weaknesses with .... f5), maybe ....f3, maybe ... Qf5, maybe ...Be5, Ng6-e5 or-h4. Alternately, if white takes on b5, sometimes black can play on the queenside with ...b4 and against the d5 pawn with Qb7. ) ]

15...Rg8 16.Rad1 Rg6 17.c4 e4 I hesitated a lot here. ... e4 is structurally commital, but I was too curious which way he would retreat the bishop. During the game I expected Be2, the consistent choice that keeps the pressure on b5 and forces me to defend or push it. I was reluctant to play ... b4, even though I guess it's the normal response to c4, because it pushes the a3 knight towards the center. ....Rb8 is tempting, but I had the idea that I would want this rook on the kingside and so didn't want to commit it unnecessarily. 18.Bb1 Bg7 19.Qe2

19...Rb8 [19...b4 20.Nc2 Bxb2 21.Nxb4 Be5 22.f4 I spent some time staring at this position in my head, but couldn't figure out how to evaluate it or what to do. Rybka's suggesting 22...Bc3 23.Nc2 Rb8 and I guess that is both obviously good and not very hard to see. ] 20.b3 I was surprised to see this; it's not a terrible move, but I can't see why white could possibly want to play it, other than to avoid losing a pawn. 20...Kf8 21.f3 Qa7+ 22.Kh1 Qc5 23.Nc2

23...Be5 My original intention, back on move 21, was to take on c4 here, but at the last moment I became really worried about 24. b4 [23...bxc4 24.b4 and now if I move the queen, my center collapses: 24...Qc7 (24...exf3 25.Qxf3 Qc7 Here Rybka is giving black –0.45; I guess I don't understand anything. If you get why black isn't worse, please leave me a comment and explain. I have nice pieces, sure I do, and I'm even up a pawn(!), but what am I going to do, win an endgame with my beautiful tight pawn structure?)

25.fxe4 fxe4 26.Qxe4 Rybka is claiming equality for black, but during the game I was scared of the open e and f files, plus I bet Fritz will say white is winning. Fritz always hates the Sveshnikov, so I avoid consulting it. ] 24.fxe4 Rh6 [24...Bxh2 I checked this-- would have been funny if it worked. 25.Kxh2 Rh6+ 26.Kg3 Rg6+ 27.Kf3 fxe4+ 28.Qxe4 but there's nothing] 25.h3 f4

[25...bxc4 I again considered taking on c4, which seems to be the best move, but I expected b4 and thought I'd lose c4 later and be down two pawns. 26.b4 Qc7 27.exf5 maybe that's just paranoia.] 26.cxb5 axb5 But now I get killed on the open c file. 27.Ne1 Ng6 28.Nd3 Qc7 29.Qg4 [29.Nxe5 Nxe5 30.Rxf4 This is the most obvious continuation, and I while I am down two pawns, my knight is just lovely and his bishop sucks a lot, so I figured maybe I could hold this... ] 29...Re8 30.Rc1 Qa7

I wanted to dream about Qe3. 31.Qg5 Rh4 32.Nxe5 Rxe5 33.Qf6 Qd7 [33...Qe3 I should have tried this. I rejected it for some reason, a reason I've forgotten the exact nature of, but not the right one. The truth is that 34.Rc8+ Re8 35.Rc3 is the only good line for white and I should have made him find it. After (35.Qxd6+ Kg7 the black king is totally safe. ) ] 34.Rf3 Reh5

Just a last try, threatening to sac on h3. 35.Kg1 [35.Rfc3? Rxh3+ 36.Rxh3 Rxh3+ 37.gxh3 Qxh3+ 38.Kg1 Qe3+ 39.Kg2 Qxc1] 35...Rh6 36.Rfc3 Kg8 37.Rc7 Nf8 38.Qg5+ 1–0
A nice game by Chris, who is awesome.


Anonymous said...

welcome to the ghetto

Anonymous said...

oh sheet, i think williams uses fritz 10 yo when he prepares, like fritz 10 is sick, and have u heard of chris williams? hes fckin awesome...

Anonymous said...

Thanks for the annotations, but sorry about the result. You'll win next time!

Anonymous said...

At least you never got up six pieces...

Anonymous said...

yo comment number 3, dont delude her yo its mean, lol

Anonymous said...

yo comment number 5, you are a stupidface