AylerKupp: The recently completed Chessgames Challenge: Team White vs Team Black, 2013 game also featured a Sveshnikov Sicilian. In that game I was a big proponent of not allowing Black's dark square bishop to be exchanged with White's knight after the Nce3, ...Bxe3 capture, and I was gratified that Yifan Hou felt similarly, although she was unable to win either. So maybe my enthusiasm to avoid that capture in the Chessgames Challenge: Team White vs Team Black, 2013 was somewhat misguided and should be curbed.In that game I was also a proponent of h2-h4 after ...Bg5 to (likely) force the Black bishop to h6 prior to exchanging knights in order to make it more difficult for the bishop to move to the q-side, an approach that Yifan Hou did not take. Since in the Chessgames Challenge: Team White vs Team Black, 2013 game use of computers was not allowed, I had several engines, some of them recently released, analyze the position after 13...0-0 14.h4 and here is what they came up with:
<Bouquet 1.8>, d=27, [+0.19]
(2) 14... Bh6 15.a4 Nxd5 16.Nxd5 Be6 17.axb5 axb5 18.Bxb5 Rxa1 19.Qxa1 f5 20.Qa4 fxe4 21.Bc6 Kh8 22.Qxe4 Qc8 23.Qc4 Bd7 24.Nb4 Bc1 25.b3 Be6 26.Bd5 Qxc4 27.bxc4 Bb2 28.Bxe6 Bxc3+ 29.Ke2 Bxb4 30.Rb1 Bc5 31.f3
(Bouquet considered 14...Nxd5 to be Black's best move, evaluating it at [+0.18]; pretty much the same for all practical purposes.)
<Critter 1.6a>, d= 26, [+0.04]:
(1) 14... Bh6 15.Nxe7+ Qxe7 16.a4 bxa4 17.Rxa4 Qb7 18.Nd5 Qxb2 19.Bc4 Bd7 20.Rb4 Qa3 21.0-0 a5 22.Rb7 Be6 23.Nf6+ gxf6 24.Bxe6 fxe6 25.Qg4+ Kh8 26.Qh5 Bg7
(Critter considered 14...Nxd5 as Black's second best move, and it evaluated it at [+0.09]; again pretty much the same for all practical purposes.
<Gull 2.2>, d=25, [0.00]
(1) 14...Bh6 15.g4 Bf4 16.Qf3 Bb7 17.Nxf4 exf4 18.h5 Nc6 19.Nxc6 Bxc6 20.0-0-0 Re8 21.Rd4 Qg5 22.Bd3 Rad8 23.Rd1 Bd7 24.Be2 Bc6 25.Bd3
(Gull also evaluated 14.Nxd5 at [0.00] but I guess it was 14...Bh6's turn to come in first when both moves were evaluated the same.)
<Houdini 1.5a>, d=28, [+0.13]:
(2) 14... Bh6 15.a4 Nxd5 16.Nxd5 Be6 17.axb5 axb5 18.Bxb5 Rxa1 19.Qxa1 f5 20.Bc6 fxe4 21.Qa6 Kh8 22.b4 Bc1 23.g3 Bb2 24.0-0 Bxd5 25.Bxd5 Bxc3 26.Qc4 Bd4 27.Bxe4 Qd7 28.b5 Qg4 29.Qd3 g6 30.Kg2 Qe6
(Houdini considered 14...Nxd5 to be Black's best move, evaluating it at [+0.08]; again pretty much the same for all practical purposes.)
<Komodo 5.1r2>, d=25, [+0.23]:
(1) 14... Bh6 15.a4 Nxd5 16.Nxd5 bxa4 17.Rxa4 Kh8 18.Ra2 f5 19.Bd3 Be6 20.Bc2 fxe4 21.Bxe4 a5 22.c4 a4 23.Rxa4 Bxd5 24.cxd5 Rc8 25.0-0 Qxh4 26.Qe2 Bg5 27.Ra6 Rf6 28.g3 Qh3 29.Bg2 Qf5 30.b4
(Komodo considered 14...Nxd5 to be Black's second best move, evaluating it at [+0.28]; again pretty much the same for all practical purposes.)
<Stockfish 4.0>, d=30, [0.00]:
(2) 14... Bh6 15.a4 Nxd5 16.Nxd5 Be6 17.b4 Rb8 18.Be2 Qd7 19.axb5 axb5 20.Ra6 Kh8 21.0-0 f5 22.exf5 Bxf5 23.Ne3 Bxe3 24.fxe3 d5 25.Ra5 Qc6 26.Qd2 Kg8 27.Bd3 Bxd3 28.Rxf8+ Rxf8 29.Qxd3 Rb8 30.g3 g6 31.Ra7 Rf8
(Stockfish considered 14...Nxd5 as Black's best move, also evaluating it at [0.00].)
So it looks like my enthusiasm for an early h2-h4 to (perhaps) force ...Bh6 was probably also misguided. It is also interesting to see how the various engines differed in their absolute evaluation of the position, although all of them thought that 14...Bh6 and 14.Nxd5 were Black's two best moves, and evaluated them the same for all practical purposes. If you averaged the evaluations as I usually do when I analyze the same position with different engines, here is what you get:
Black's Bouquet Critter Gull Houdini Komodo Stockfish
Move d=27 d=26 d=25 d=28 d=25 d=30 Avg
14...Bh6 [+0.19] [+0.04] [0.00] [+0.13] [+0.23] [0.00] [+0.10]
14...Nxd5 [+0.18] [+0.09] [0.00] [+0.09] [+0.28] [0.00] [+0.11]
Therefore 14...Bh6 "beats" 14...Nxd5 as the "best" move by a nose, and neither of them is particularly convincing for White, at least not at the search depth that I allowed these engines to run. So I guess the Sveshnikov still lives.
Kibitzing
icon.
