Wednesday, October 31, 2007

Little Endgame Study

Taken from a recent Blitz game. White played 50. Qxf3 and eventually lost. What do you think, was his decision to take Black's queen correct? If your answer is yes, how should he have continued after 50. ... exf3?

It turns out that White's decision to trade queens was correct. In fact, after 50. ... exf3, White is winning! But White must be careful, because after 51. Se3 a4 52. Sc2?, Black wins!

Black wins.

White's knight is tied to c2 to prevent the a-pawn from promotion, and White's king cannot attack Black's g&f pawn without moving out of the g-pawn's critical squares! While White's king is moving back and forth on the 1st rank, and White's knight is guarding the a-pawn's queening square from c2, Black's king has all the time in the world to take White's pawn on f4 and then move to h2 to escort the g-pawn to g1. White's king even has to watch out for a king&pawn mate!

Carlos mentioned the winning move for White: 52. Sxe5!

White wins.

If Black takes on e5, he can do nothing to prevent the promotion of White's c-pawn, and his a-pawn is too slow (White will take with check!). If instead Black decides to advance his a-pawn right away, White moves his knight to b4, and now starts his pawn avalanche with his f-, d- and c-pawn! If Black choses not to push the a-pawn at all, moving in his king instead, the result is the same. Meanwhile, without the support of the Black king, the White king successfully holds Black's g/f pawns in check (if you forgive the pun).

So remember: do not hesitate to sacrifice your one remaining piece for the benefit of your pawn storm!


Anonymous said...

I would play exf3. I don't think black's a-file pawn can make it due to white's knight.

As in a lot of these situations it's down to the king to support his pawns. I would have moved my king up to protect and promote my pawns.

What did black actually play?

Christian said...

For White, moving his knight over to the b-file to stop the advancing a-pawn seems like the logical move. E.g. 51. Ne3 a4 52. Nc2 ...

However, as you point out, this allows Black's king to come forward and help promote his two king-side pawns. Which is what happened in the game.

So the question is: What should White play instead? Should he have avoided the queen trade altogether, or does he have something else up his sleeve after 50. ... exf3?

Anonymous said...

White has to trade Queens or he's lost, but after 50...exf3, could move his Knight to e3 i.e 51.Ne3
This way he avoids the advance of black's a pawn to promotion, and leave his King to deal alone with g and f black pawns, what he is able to. Black King can not help his pawns, cause white would promote one of his(f, d or c). And on e3, the Knight can even make a decisive sacrifice with a timely Nxe5. But in blitz games with time pressure, these positions are usually lost. I would lose it. Greetings,
Carlos, from Brazil.

Anonymous said...

Thank you so much for mentioning(did I say it right?) my name in yor post. You know, with a giant board AND TIME to think over a position, you find the answer by force. I know hundreds of players could do it in half time. I'm just an average club(class c)player, who reads your site everyday and loves chess... thank you very much!

Anonymous said...

Carlos, from Brazil...( I forgot...)

Christian said...

You're welcome, Carlos. By the way, did you know that you could post under the name "Carlos" without having to log in by clicking on "other" under "Choose an identity" and then entering your nickname?

As for endgame calculations: I believe one important aspect is to recognise some of the basic king versus pawn formations, for example, connected passed pawns vs king, two pawns separated by one file vs king, or two pawns separated by two files vs king. Also, of course, the critical squares, which jrobi explained lucidly in his last video. Such pattern recognition facilitates calculation enormously.