|Domdaniel: This game is about correctly evaluating bishops -- bishops vs knights, the bishop pair, bishops of opposite colours, etc.|
Black made a telling error on move 21, with 21...bxc6. This left him with a weak pawn at c6, and a locked-in 'bad' light-square bishop on e8. It also leaves White with a strong knight on e5 -- so strong that Black later exchanges his dark-square g7 bishop for it, and the resulting dark square weaknesses cost him the game in the end.
Back at move 21, Black is not exactly comfortable, but neither is he lost. In this position he had two reasonable alternative moves, each giving up bishop for knight:
click for larger view
He can play 21...Bxc6, putting his LSB on a good square and solving its problems. Of course White can play 22.Nxc6 (exchanging his good knight for a previously bad bishop) and after 22...bxc6 Blacks has a weak point on c6 anyway - but not a bad bishop.
Alternately, Black can play 21...Bxe5, exchanging his other bishop for the knight. He still gets the dark square weaknesses on the kingside, but not so bad as those in the game. After 22.dxe5 Bxc6 his LSB comes to life.
Why did Spielmann reject these moves? At the time, it was widely felt that bishops were *absolutely* better than knights. Capablanca called the win of bishop for knight 'the minor Exchange'. Tarrasch wrote of the power of the bishop pair. Players did not lightly concede this advantage, even in positions like this where one bishop is bad and the pair's future is remote. Masters like Chigorin and Nimzowitsch, who voluntarily gave up bishops for knights, were deemed eccentric.
Now, of course, we know that the value of bishops and knights depends on the position. Yet players still make this kind of mistake, trying to keep the bishop pair when an exchange would be better.
Note that later in the game Rubinstein had no qualms about giving up a bishop with 26.Bxd5! -- some players might hesitate, thinking bishops of opposite colours drawish. But they can be deadly in combination with heavy pieces, as here. 26.Bxd5 removes Black's best-placed defender: never mind dogma about the bish pair.
This kind of mistaken thinking about bishops of opposite colour is also still prevalent. As Alburt and Palatnik wrote of this game "Bishops of opposite color can be drawish in an endgame. But for the attacker, an opposite-color bishop can be like an extra piece!"
In this game, Rubinstein showed a better understanding of how bishops may change in value. Despite being a world-class GM, Spielmann was a little too reliant on traditional dogma.
He understood this himself. Three years later, at Karlsbad, he was singled out for praise by Nimzowitsch for having made adjustments to his style of play.