The Hou-Koneru Bishop Endgame

In the ongoing Women's World Chess Championship, a key encounter came in the semifinal, as the two highest-rated players, Humpy Koneru and Hou Yifan, met.  In the first game of the two-game match, Hou had white, and Koneru played the Berlin Defense to the Ruy Lopez, clearly aiming to achieve a draw so that she could play for a win of the game, and the match, with white in the second game.

At move 28, they reached the following position:

Hou Yifan - Humpy Koneru,

Women's World Chess Championship, Round 5.1 (2010)

Position after 28.Kxd3

Hou has a large advantage here.  The pawn formation is the classic one from the Ruy Lopez Exchange Variation (although, as mentioned above, this game actually started as a Ruy Lopez Berlin Defense), where White has a mobile pawn majority on the kingside while Black's queenside pawn majority is useless because of the doubled c-pawns.  Here Black might appear to have some compensation because White's kingside pawns are on the same color as her bishop, but White can (and, indeed must) advance those pawns so her bishop is by no means "bad."

More important is the difference in king activity: White's king can take up a powerful position in the middle of the board, while Black's is stuck on the 7th and 8th ranks, in part because of her earlier mistake 25...g6?, blocking the king's path from f7-g6-f5.  Black may already be lost but at the very least she is in for a long and painful defense.

Ulf Andersson's Dynamic Endgame Play (Part 1)

For a certain type of chess player, I imagine, there exists the following hierarchy: First, there are boring chess games; second, there are incredibly tedious chess games; and last, somewhere between somnolence and apathy, there are the games of Ulf Andersson.

This reputation for boring play is not entirely undeserved as Andersson, especially in recent years, has had a tendency to agree to many short draws.  Witness Andersson-Ruck, Banja Luka 2007:

1.Nf3 g6 2.c4 Bg7 3.Nc3 c5 4.g3 Nc6 5.Bg2 e6 6.e3 Nge7 7.O-O O-O 8.d4 d6 9.b3 d5 10.cxd5 cxd4 11.Nxd4 Nxd4 12.exd4 Nxd5 13.Nxd5 exd5 14.Be3 1/2-1/2

In that tournament, Andersson scored 4.5/9, a perfect record of nine draws with no wins and no losses.  His games with White, in particular, consisted of draws of 14, 25, 14, and 11 moves.

I don't know Andersson's reasons for agreeing to so many short draws, but I can speculate, based in part on his recent tendency to drop out of open tournaments before they are over, that perhaps he feels with particular acuteness the phenomenon, known to so many chess players, that the pain of a loss far outweighs the pleasure of a victory.

But Andersson's games were not always like the one quoted above and, indeed, in the past a simplified position like the one reached at the end of Andersson-Ruck would have been the starting point for a long attempt to grind out a win, rather than the time to shake hands.

In this post and its sequels, I will discuss some of Andersson's fascinating wins in the endgame.  In addition to showing that Andersson's games are not all boring, I want to make another point about the nature of good endgame play.  For some players--probably including many of the ones who find Andersson boring--endgame play is a matter of slowly and patiently grinding away with a static advantage (like a knight on a strong outpost) or against a static weakness (like a weak pawn).  The exploitation of static advantages is of course a key component of good endgame play, but very often, to win in a simplified position, one must play dynamically.  As there come to be fewer pieces on the board, the differences in their activity can assume decisive significance.  Andersson is a master of such dynamic endgame play.

The Elusive Win in Kramnik-Carlsen (Part 2)

As a commenter pointed out, I have not addressed Black's moves other than 69...hxg5 in response to the proposed 69.g5.  I attempt to fill that gap here.  As a reminder, here is the position we are looking at:

There are two sorts of alternatives to 69...hxg5 to consider here: 69...h5 and various waiting moves with the Black king.  Before I begin, though, I should note that I focused on 69...hxg5 because it seemed the most direct way for Black to achieve his goal: eliminating the remaining White pawns.  As we will see, however, this is not the only way that Black can achieve a draw in this ending.

a) 69...h5

It seems foolish at first sight to leave White with the advanced pawn on g5, but Black has a very specific goal in mind.  If Black can capture the pawn on g2, force White to advance the other pawn to g6, and get his King back in time, then he achieve the following fortress position:

White cannot force Black's king away from e7 and f8: if White's king gets too close, then it is stalemate.

The Elusive Win in Kramnik-Carlsen

Today at the London Chess Classic, Vladimir Kramnik and Magnus Carlsen played a fantastic game. At move 62, Kramnik exchanged down into what he must have thought would be an easily-winning piece-up endgame. At move 65 the players reached the following position:

Carlsen saw a clever drawing idea, however: he would send his king deep into White's position. If White tried to push his own king forward, Black would attack the White pawns from behind. There thus followed:

65...Kd4 66.Kf3 Kd3 67.g4 Kd2 68.Be6 Kd3

Here Kramnik played 69.Kg3 and Carlsen defended marvelously to achieve the draw.

In this post, I want to examine the alternative 69.g5, which I believe was winning.

Stalemated Knight

I happened to be looking at the otherwise unremarkable game Rodriguez Crespo (1904)-Kolar (2196), August Open Cambre ESP 2006, when I arrived at the following position:

White has been trying to make something of his good-bishop versus bad-bishop advantage in a typical French ending, but Black seems to be holding. (A couple of moves earlier, White made the mistake of allowing the Black bishop to get out to h5.)

Now things start to get interesting, however.

44...Nf7? (44...Kf7 45.Ng2 Ng6 and Black should be fine.)

45.Ng2! Aiming for f4, forking the h5-bishop and the e6-pawn. Now Black is in trouble.

45...Bg6? (45...Ng5 46.Nf4 Bf7 47.h4 Nh7 48.Kg4 Nf8 seems to hold for Black. White has no way of breaking through.)

46.Bxg6 Kxg6 47.Kg4? (47.h4! wins, not letting the Black knight out via g5. Black is running out of moves and the White knight will start to eat up his pawns. For example, 47...Kf5 48.Kf3 Nh8 49.Ne3+ Kg6 50.Ng4 Kg7 51.Nf6 Ng6 52.Kg4 Nf8 53.Kh5. Now it is a draw again.)

47...h5+ 48.Kf4 Ng5 49.h4 Ne4 Wait a minute, you must be asking, how can this be a draw? Isn't White going to lose his pawn on a4?