A Master Knows When to Break 'The Rules'

Continuing with Lasker's Moves that Matter,
the third game between Lasker and Capablanca at St. Petersburg 1914 was one of Lasker's most famous clutch games. Trailing the Cuban by one point with three games to be played, he needed to win.

In the diagrammed position, Lasker played 12.f5.

Kasparov (KAS): !; '"It has been wrongly claimed that this wins the game, but I would like nothing better than to have such a position again" (Capablanca). According to the classical laws of the Steinitz theory, this is indeed a dubious venture. White devalues his Pawn majority on the Kingside, giving himself a weak, backward Pawn on e4, whereas Black is presented with a powerful outpost at e5. Aren't these rather too many defects for one move?! But Lasker's eagle eye is trained on the e6-square.'

Soltis (SOL): '"One of the most famous, paradoxical and deep moves in chess" (Damsky). Why paradoxical? Because it "immediately destroys three rules of Steinitz's theory" (Vainshtein). White creates a backward Pawn on an open file, he grants Black a wonderful outpost at e5, and he cripples his Kingside majority. Tarrasch had ridiculed f4-f5 when it occurred in a similar position (e.g. Fahrni-Janowsky, Nuremberg 1906). But Tarrasch's knee-jerk classicism had been proven wrong when Lasker won [Lasker-Salwe, St.Petersburg 1909] and the f4-f5 idea had worked reasonably well in other previous games -- including Alekhine-Lasker from the previous round.'

Capablanca went astray two moves later when his superb positional judgement failed him.

St. Petersburg (final) 1914
Capablanca, Jose Raul

Lasker, Emanuel
(After 11...f7-f6)
[FEN "r1b1r1k1/1pp1n1pp/p1pb1p2/8/4PP2/1NN5/PPP3PP/R1B2RK1 w - - 0 12"]

Black played 12...b6

KAS: 'Tarrasch and Reti recommended 12...Bd7 13.Bf4 Bxf4 14.Rxf4 Rad8 with the idea of ...Nc8 and ...Nd6, but the Bishop is more active on b7, from where it attacks the e4-Pawn.'

SOL: [After giving the same line as KAS] 'Reti also suggested 12...g5 with the idea of 13.fxg6 Nxg6 14.Rxf6 Be5.'

13.Bf4 Bb7

KAS: ?!; 'A fundamental mistake: in general it is advantageous for Black to undouble his Pawns, but in the given instance the d6-Pawn will become a chronic weakness. Therefore more logical is 13...Bxf4! 14.Rxf4 c5', with analysis leading to equality.

SOL: ?; '"Here it would doubtless seem better to play 13...Bxf4 14.Rxf4 c5", Capa wrote'.

The game continued 14.Bxd6 cxd6, and Capablanca resigned on his 42nd move. Lasker played several excellent moves which are well documented in the literature. After his 13th move, Capablanca tried to break out of his passive position, but failed. To play through the complete game see...

Emanuel Lasker vs Jose Raul Capablanca, St Petersburg f 1914
http://www.chessgames.com/perl/chessgame?gid=1258181

...on Chessgames.com. Shaken by the defeat, Capablanca lost his next game to Tarrasch and finished a half point behind Lasker.