Winning starts with what you know
The new version 18 offers completely new possibilities for chess training and analysis: playing style analysis, search for strategic themes, access to 6 billion Lichess games, player preparation by matching Lichess games, download Chess.com games with built-in API, built-in cloud engine and much more.
Last night I presented a pair of games on my weekly Playchess lecture purporting to have been played by Anatoly Karpov and Bobby Fischer. Indeed, they were played by those two men – just not against each other (although they could have been – see below).
Of course, it was an (early) April Fools’ gag. Last year I paid homage to the day as well, though without a gag, so this year I decided to take things up a notch. Unfortunately, my weekly show didn’t fall on April 1, but I did offer a hint in my blurb, referring to the program as the show for the week of March 28 to April 3. Naturally, I did the best I could to present an incredibly implausible story (Karpov and Fischer played some serious though informal games nearly 30 years ago, and no one even knew about it?) in a plausible guise, with reasonable success.
Nevertheless, I hasten to assure you, that while the background story was essentially true – Torre did come to Las Vegas to give a simul, stayed with Filipino friends, I was told later that Fischer had been in the van, and Fischer and Karpov did meet in the Philippines in the mid-to-late 70s to discuss an unofficial world championship match. But the two never played – or at least if they did, I have no knowledge of this.
The games I showed did bear a very interesting relationship to each other, though! In 1968, Fischer introduced an interesting gambit – essentially the Grand Prix Attack against the Sicilian, but with colors reversed (and thus a tempo less) – against IM Anthony Saidy, and won a fine game. In 1973, Juan Bellon played that same gambit against then-Candidate Anatoly Karpov, and after a see-saw battle the game concluded in a draw. What made the games especially interesting, and in a way making the show a bit less like a hoax and more like historically informed speculation, is that they reached the exact same position in each case through White’s 23rd move!
While few in my audience were likely to know about the Bellon-Karpov game, Saidy-Fischer is much better-known, and both games could have been found with a database search. So, to help delay detection, I reversed colors and the queenside-kingside orientation of the pieces – successfully, it seems, as none of the comments I noticed during the show made reference to either of the real games. What my viewers saw was the following:
Understandably, there was much head-scratching about what the opening might have been! I started the show from this point to avoid worries about plausibly constructing this position.
As for the games: they deviated from this point: Bellon played 23…Nb4 (Ng5 in the reversed and rotated diagram), after which Karpov was fine, while Fischer played 23…Kh7 and went on to win a nice game. Had this really been a Karpov-Fischer game, would Karpov have fared better than Saidy? It’s likely that he would have put up more resistance, but whether it would have sufficed to save the game is something we’ll never know.
In last night’s hoax, Karpov saved it the first time and lost the second time, but those were really just the Bellon-Karpov and Saidy-Fischer games; presented here, with all the pieces where they ought to be, with some light notes:
Karpov,Anatoly (2660) - Bellon Lopez,Juan Manuel (2400) [A25]
Madrid (5), 01.12.1973
1.c4 e5 2.Nc3 Nc6 3.g3 f5 4.Bg2 Nf6 5.d3 Bc5 6.e3 f4 7.exf4 0-0 8.Nge2 d6 9.0-0 Qe8 10.Na4 Bd4 11.Nxd4 exd4 12.a3 a5 13.b3 Bf5 14.Nb2 Qg6 15.Qc2 Nd7 16.Re1 Nc5 17.Bf1 Ra6 18.Bd2 Rb6 19.Bxa5 Rxb3 20.Bd2 Ra8 21.a4 h5 22.h3 Ra6 23.a5
Nb4 [23...Kh7 is Saidy-Fischer] 24.Bxb4 Rxb4 25.Ra3 b6 26.Rea1 Qe6 27.axb6 Raxb6 28.Ra8+ Kh7 29.Qd1 g6 30.Na4± Nxa4 31.R8xa4 Rxa4 32.Rxa4 Bxh3 33.Ra7+- Bxf1 34.Rxc7+ Kh6 35.Qxf1 h4 36.Kg2 Rb2 37.Kf3 [37.f5 threatening 38.Qc1+ 37...Qe3 38.Kg1 Qd2 39.fxg6 Rc2 40.g7 Kh7 41.g8Q+ Kxg8 42.Qg2 Qe1+ 43.Qf1 Qd2=] 37...d5 38.gxh4 [38.c5!?] 38...Rb3 39.cxd5 Qxd5+ 40.Kg3 Qf5 41.f3 Rxd3 42.Rc6 Rc3 43.Rd6 Kh5 44.Kg2 Rc2+ [44...Qxf4 45.Rd5+ Kh6 46.Qf2 Qc1 47.Qg1 Rc2+ 48.Kh1 Qf4 49.Qg5+ Qxg5 50.hxg5+ Kh5 51.Rxd4 Kxg5=] 45.Kg3 Rc3 46.Kg2 Qxf4 47.Rd5+ Kh6 48.Qe2 [48.Qf2 see the note to move 44] 48...Qc1 [48...Re3 49.Qf2 d3 50.Qg3 Re2+ 51.Kh3 Qf6 52.Qg5+ Qxg5 53.hxg5+ Kh5 54.Rxd3 Kxg5 55.Kg3=] 49.Rh5+ ½-½
49...Kxh5 50.Qe5+ Kxh4 a) 50...g5 51.Qe8+ Kh6 (51...Kxh4?? 52.Qh8#) 52.Qe6+ Kg7 53.Qe7+=; b) 50...Kh6?? 51.Qh8#; 51.Qg3+ Kh5 52.Qg4+ Kh6 53.Qh4+ Kg7 54.Qe7+=; or 49...gxh5 50.Qe6+ Kg7 51.Qe7+=.
Saidy,Anthony - Fischer,Robert [A25]
USA, 1968
1.c4 e5 2.Nc3 Nc6 3.g3 f5 4.Bg2 Nf6 5.d3 Bc5 6.e3 f4!? 7.exf4 0-0. It's very strange that Fischer's 6th move is universally condemned as bad (though very interesting), but in my research this move isn't addressed. But 7...d6 leaves Black with reasonable compensation and a possible transposition to the game, without allowing White the chance missed on move 8.
8.Nge2. 8.fxe5 looks risky, but Black can't exploit the f2 square and White seems to get a large advantage after 8...Re8 (8...Nxe5? 9.d4+-; 8...Qe8 9.Bxc6 dxc6 10.d4 Bb4 11.Nge2+/-) 9.Nge2 is Shredder 9's suggestion, when it thinks Black is doing rather poorly (9.f4 d6 10.Bxc6 bxc6 11.d4 is the standard "refutation" – see Soltis's Bobby Fischer Rediscovered (2003) and Wade & O'Connell's 1972 Bobby Fischer's Chess Games, but it's not so clear after 11...Bb4 12.Nf3 Bh3 13.Qe2 c5 14.a3 Ba5, when the edifice of White pawns is in serious danger of collapse.) 9...Rxe5 10.0-0 Nd4 11.Nxd4 Bxd4 12.Bf4 Re8 13.Nb5 Be5 14.Bxe5 Rxe5 15.d4+–.
8...Qe8 9.0-0 d6 10.Na4. 10.Be3 Bxe3 11.fxe3 exf4 12.exf4 Qe3+ 13.Kh1 Ng4 14.Bd5+ Kh8 15.Kg2 Qe8!-+ Maric; 10.Ne4 Nxe4 11.dxe4 Qh5 12.Be3 Bg4 13.Qd5+ Kh8 14.Nc3 Nd4 gives Black some play (Soltis). In fact, it wins the queen, though White has reasonable material compensation after 15.Bxd4 c6 (15...Bxd4 is perhaps better for Black, giving him some advantage) 16.Bxc5 cxd5 17.Bxd6 d4 18.Bxf8 Rxf8 19.f3 Be6 20.Nd5 is unclear!]
10...Bd4 11.Nxd4 exd4 12.h3 h5! 13.a3 a5 14.b3 Qg6 15.Nb2 Bf5. Soltis refers to this as the "most Nimzovichian" of Fischer's games, reminiscent of the very famous Johner-Nimzovich game from Dresden 1926 (look it up, audience!). Black's play here is all about blockade: he's going to seal White in, then destroy him. Ironically, though, the computer is quite happy here with the White pieces!
16.Qc2. 16.Re1 Nd7 17.Kh2 Nc5 18.Bf3 is slightly better for White, according to the oracle Shredder 9, though I'm not sure even it believes what it's saying.
16...Nd7! 17.Re1 Nc5 18.Bf1 Ra6!=/+ 19.Bd2 Rb6 20.Bxa5. 20.b4 would be a nice pawn sac if it worked, liberating his queenside, but it doesn't, according to Soltis: 20...axb4 21.axb4 Nxb4 22.Bxb4 Rxb4-/+ 23.Re7 and now 23...Qf6! 24.Rxc7 (24.Rae1 Ne6!-+; the ugly 24.Ree1 is forced. 24...Qd8–+.
20...Rxb3 21.Bd2 Ra8 22.a4 Ra6!-/+. Incredibly, and I mean that literally, the computer finds this position equal. 23.a5 Kh7 (23...Nb4 is Karpov-Bellon) 24.Red1 b6. But now Shredder 9 agrees: Black is clearly better. 25.Be1. 25.axb6 Raxb6 26.Ra2 Nxd3–+ Maric 27.Bxd3 Bxd3 28.Nxd3 Rxd3-/+ 29.Re1 Rbb3–+.
25...bxa5–+ 26.Na4 Rxd3! 27.Bxd3 Bxd3 28.Qa2 Nb4 29.Qa3. 29.Qb2 Nxa4 30.Rxa4 Bc2 31.Rxa5 Rxa5 32.Bxb4 Bxd1 33.Bxa5 c5 might not look so bad for White, but between the passed d-pawn and White's porous light squares on the kingside, Black is almost guaranteed a win here.
29...Nc2 30.Qb2 Nxa1 31.Rxa1 Nxa4 32.Rxa4 Qe4 33.Bxa5? 33.Qd2 Rb6 34.Ra1 Bxc4–+ Maric. 33...Rxa5 34.Rxa5 Qe1+ 35.Kh2 Qxa5 0-1.
