Perplexing Sidelined Knight!
This highly perplexing ending just surfaced on ChessToday.Net (Mikhail Golubev commenting). Thanks to chess enthusiast Kurt Stein for bringing this intricate ending, and the problems computer engines have with it, to my attention.
GM Viktor Laznicka (CZE) – GM Viktor Bologan (Moldavia)
World Cup Khantiy-Mansiysk
As a preamble, I enjoyed GM Josh Friedel’s Chess Life Online narrative of the trials and tribulations just to get to this Siberian way-station. And then, to be eliminated practically as soon as one arrives is truly agonizing!
I thought it was bad enough to venture up to Toronto for a David Lavin tournament from New York City (taking People’s Express to Buffalo, then transferring to a bus across Niagara Falls and being faced with hostile customs questions) – this is worse. 🙂
Here’s the action after Laznicka launched a clever combination to put Bologan in quasi-zugzwang.
It’s a good position to test chess engines, because most scenarios are well beyond the engine horizon, even for the big names such as Rybka.
Position after 55. Kg2
Black to play. Is this real zugzwang or quasi-zugzwang or pseudo-zugzwang?
Golubev indicates white is playing for a win, and that he surely is, but what’s the correct result? A great computer test!
Black played 55…d4 here and lost slowly. White’s king *carefully* approached the pawn and never allowed black knight fork tricks.
It is my contention that 55…Nb7!, shuttling between b7 and d6, draws if black leaves the pawn on d5 for the time being. The point is, when white tries to approach the p/d5, THEN black gets fork tricks.
Example:
55…Nb7! 56. Kf2 Nd6 57. Ke1 Nb7 58. Kd2 Nd6 59. Kc3 Kxh4! – only now! and black is saved due to a fork.
Or, 59. Kd3 Kg4! 60. Kd4! is the only way for white to draw. If 60. f6? Kf5! 61. Kd4 Ke6! wins for black.
In fact, there are some pure fantasy variations here with DOUBLE fork tricks!
Here is a really nice line from the start position:
55…Nb7 56. Kf1 Nd6 57. Ke1 Nb7 58. Kd1 Nd6 59. Kc1 Nb7 60. Kb2 Nd6 61. Kc3 Kxh4! (well-timed!) 62. Kd4! (not 62. f6? Ne4+ 63. Kd4 Nxf6 64. b7 Nd7) and look at this position:
Black to play. What's the best move?
Unbelievable analysis position. It turns out black actually has two moves.
But NOT 62…Nxf5+?? 63. Kxd5 Ne7 64. Kc5! and the pawn queens – a common beginner’s error to snatch a poison pawn like this. The mundane line is 62…Kg5 63. Kxd5 Nb7 and draws. Hidden, though, is something much prettier.
62…Nc4!! Wow!!! 63. b7 Na5!! (Fork Trick #1) 64. f6? (64. b8=N! draws!) 64…Nxb7 65. f7 Nd8!! (Fork Trick #2!!) and now forced is 66. f8=N (another under-promotion on a different square) 66…Kg5 and black has an edge (but not a forced win) in the resulting ending after 67. Kxd5 Kf5! Wow!! The multiple fork tricks and the multiple under-promotion defenses are really something special.
Conclusion: I don’t see any win for white if black just hangs tight with Nd6-b7-d6 shuttle, waiting for WK to approach. Readers?
Thinking Your Way To Chess Mastery – 2nd Installment
The second installment of my live Internet-TV show has been postponsed to Monday, December 14th, at 2 PM PDT (5 PM EST). Register for free at Chess.Com and tune in (under the “Fun” tab on the right, you see the “TV” link). This is different from most chess videos online because here you get people and chessboards, imagine that. And live Q&A throughout.
Chess Today
I just started subscribing to GM Alex Baburin’s excellent, regular, chess periodical (emailed to the readers with PGN and CBV attached). Good stuff!
In “CT”, I noticed GM Nakamura missed many chances to put long-suffering Ni Hua away, starting with the crunching 35. Rxh6 winning – for example 35…Ke4 36. e6! and fini.
My Laznicka-Bologan analysis (above) made it into CT Issue #3319.
And Readers Deserve to See
Maria Yurenok (photo from John Saunders London Press Release)
London Calling
HOW IT WORKS:
IM Mark Ginsburg Presents A Personal Chess History 