<<44. Qb4> ( 1.68)

<First three engine preferences>: <44. Qc3> ( 0.78); <44. Rb2> ( 0.79); <44. Rc1> ( 0.89) White’s move crosses the 1.40 threshold into a lost game for White and is therefore classified as a blunder.>

Many of these evaluations should be preceded by a minus sign.

click for larger view

SF10 still gives 47...Kd7 as best move but 47...c4 is still just about winning! Black remains winning until the blunder on move 54...Qc5?? Position after 54.Kg2 (! Shereshevsky side-stepping the check on e5 so renewing the threat of 55.Qa6.)

click for larger view

But SF10 now says 54...c3! introducing the resource of ...Qe4+ is still winning e.g. 55.Qa7+ Rb7 56.Qe3 Rb6 57.Ra7+? Kb8

click for larger view

or 55.Re8 Kb6 56.Qa8 Kc5

click for larger view

click for larger view

SF10 also gives an interesting line. 48.Qa1 Rxe4 49.Ra8 (If 49.Ra7+ Ke6 50.Rxg7 Rf4 Black looks safe and winning to me.) 49...Re2 50.Qa4+ Kc7 51.Ra7+ Kb6 Unbelievable! The King escapes to d5 by running toward the White heavy pieces. 52.Qa5+ Kc6 53.Rxg7 Kd5

click for larger view

A lot of the work but not all. Computers are always seeking the objectively best moves rather than the ones that give the best practical chances.

In this game Lasker, having cocked up on move 54 and finding himself in a sorry predicament, had the chance of making a bit of a fist of it next move with <55...c3>.

click for larger view

This allows White the opportunity to win Black's Q with <56.Rc8+> but after <56...Kd7 57.Rxc5 Rxc5> the R supporting the P's advance gives excellent drawing chances.

Of course Schlechter would probably have realised that and continued <56.Ra7+> as in the game but even then having the c-pawn one square nearer promotion would have given him a bit more to think about.

However, I at least found a few really interesting. Mostly this one made me think:

<"If Black were satisfied with a draw, <42...Rb8> would have accomplished it.">

Hm, really, José? Gibing away a pawn and letting your king be driven in the corner with <43.Rxb8 Kxb8 44.Qxd6+> is a draw? So I asked Stockfish and it went on <44...Ka7 45.e5!>, and now the queen has a choice of checks and a further pawn to pick up. Although Stockfish said it's still only +0.3 instead of 0.0, i found this highly suspicious. My main line is <43.Rxb8 Kxb8 44.Qxd6+ Ka7 45.e5! fxe5 46.Qxe5 Qb7 47.Qa5+ Kb8 48.Qxg5 c5+ 49.Kg3 Qb3+! 50.Kg2 c4 51.Qd5 g6>, leading to the following position:

click for larger view

However I then realized that it's probably an elementary queen endgame where there are so few pawns on the board that both kings are naked, there's checks aplenty, and nobody will really be able to do anything except someone blunders. Which means that while I may be able to lose this with both sides, Capa and Lasker would probably be terminally bored having to play this childish pointless endgame.

Am I right with this assessment or do I miss something?

57...gxQ>

However, from then on, I must admit that I didn't really understand the game. Sure, Stockfish told me that Black was losing after <44.Qb4?>, but that Lasker then went on the quickly lose most of his advantage -- but why exactly? Also I left my laptop for a while while Stockfish computed the position after <54.Kg2>, and when I came back, I realized that its evaluation had dropped from -1.5 or -2 to close to 0.0. So I realized that even Stockfish isn't really helpful in these kind of positions.

With one or two exceptions, Capablanca's annotations are clearly written from his point of view and especially for his own point of view. Mere mortals like me just lack the understanding of knowing what he means. He definitively didn't seem to care much about other people, but anyways.

Then I saw that the game was included in Pachman's <Modern Chess Strategy> 1963 and I finally began to understand some of the shenanigans going on, why white needs to prevent a queen exchange, why the white queen does its little dance in front of the black queenside, why the black king moves over there, and I understood what went on after the pawn exchange on move 40 -- that White is simply fighting for the control of the a-file while Black is trying to prevent it. So it makes sense that after <48.Qa1>, the position is essentially a draw, and that Lasker misevaluated it by playing for a win.

So far, so good, although I must say I realized there are games where my understanding of chess is only enough to barely scratch the surface. But anyways, I think I learned a bit. Let's see if my brain manages to retain some of it. (However, my bloated ego finds some consolation in that even Lasker misevaluated this position. And Pachman says these kind of heavy piece endgames are the hardest for everyone.)

Anyways, what I did understand and find much more unusual is how Lasker completely misunderstands his predicament after <55.Qa6!>. Just like me, he thinks he's still having an attack when in reality, he should fight for survival. That usually only happened to his opponents.And three moves later, it's just a forced mate in three for him. It's not that unusual that Lasker blunders (see his Cambridge Springs disaster in Pillsbury vs Lasker, 1904), but that he blunders three times in a row must be pretty unique.

So for the first time in ages (or ever?), Lasker was behind in a match for the World Championship. The remaining games were drawn except the last one, which the old fox of course won. I'm looking forward to that one. But really, respect to Schlechter who seemed to get these positions way better than Lasker. Or maybe he was impervious to Lasker's "magic". He just knew he should get that a-file and he fought for it, not caring for anything else. Not sure.