It was a challenging teaching moment for me. In blitz, play is by intuition most of the time. How important is this intuition when there is plenty of time on the clock?
We backed the game up to the beginning of the endgame, when I forced queens off the board to go into an pawn ending that I considered comfortable.
Black to move
I played 37...Qg6+ after six seconds thought. For the rest of the game both players moved in one and zero seconds every move, suggesting that we were racing the clock more than playing the board. Nonetheless, during my postgame analysis, I sought to extract the truth of the position as preparation for the lesson with my student.
He wanted to know what I would have thought about if I had fifteen minutes.
Black's pawn structure is worse, I said. With queens on the board, both players are playing for three results: win, draw, or loss. Exchanging queens, I said, reduces the possibilities to two. I am playing for a win or a draw.
However, that explanation assumes correct play. In fact, with correct play this game should be drawn with or without queens. Exchanging queens eliminates the isolated pawn, as was my intent. Even so, both players have three results possible. It is a draw with correct play and a win for one player if the other blunders.
As it happens, I blundered.
38.Qxg6 hxg6 39.Kf3 Ke7 40.Ke4 Ke6 41.g4
Black to move
White's last move restrains Black from playing 41...f5?? Yet, that is exactly the move that I played.
This error gives White an outside passed pawn on the h-file. Black's king is forced to contend with the h-pawn, permitting the White king time to mop up the other pawns. See "Outside Passed Pawn" and "Fox in the Chicken Coop".
Even so, it is not as simple as racing over to the a-file thanks to Black's pawn majority in the center.
Later, my opponent erred by pushing the h-pawn prematurely. My blunder was the second to the last error, and it was the last error that decided the game.
With enough time to think, I would not have played 41...f5. Likewise, my opponent would have held back the h-pawn if not for his or her time pressure.
Today in a rapid game, I played instantly in a position that required a simple counting exercise. Counting might have taken five seconds, but I moved instantly (see "Panic Mode"). My memory or my intuition gave me the result of counting without expending the five seconds to verify. I could have invested the five seconds with ninety seconds remaining on the clock, plus a small increment that can add time.
White to move
It is helpful to be working with a skeptical adult student who wants to learn endgames. He will distrust my intuitive blitzing mode and ask how I calculate.
How do I know that White's h-pawn will promote before Black's a-pawn?
54. Rxg5?? leads to a position where both players have the same number of pawns and neither player can hope for more than a draw. That's why I played 54.Rxc7+ and went on to win the game after a successful pawn race.
If the rook exchange took place on c6, instead of c7, the Black king gets White's a-pawn one move faster. Even then, however, White's queen covers the promotion square and the Black king is too far beyond its pawn.
It seems to me that counting is not the only element of knowledge of rook endgames and pawn endgames that guided my intuition here. There are many pawn races where pawns promote in opposite corners. In the dramatic denouement of the film Searching for Bobby Fisher (1993), Josh Waitzkin's opponent promoted first. However, his king stood on the long diagonal. Josh promoted his pawn with check and skewered the king to win the queen.
My memory of the ending of this movie was not conscious during the game. Did it contribute to informing my intuition?
No comments:
Post a Comment