What defines a perfect move?
Sort:
This has been something that has really been puzzling me lately. Okay, it's really for us to intuitively know that aimlessly sacking your queen on move five is never a good idea. So we do know what bad moves are. And we know that if only one checkmate is presented to someone, that checkmate is the perfect move. But how do we compare moves in a position in the drawing range for both sides? Maybe this sounds kind of stupid, but really, how do we which one is better?
May 31, 2017
0
#3
Good point. I just finished a game where Stockfish 8, 20 ply, rated both sides moves as shown below (I had Black). For the vast majority of moves, Stockfish considered several moves to be essentially as good as the "best" move. So there are often several moves that are as "perfect" as each other. They equally develop, threaten, protect, etc.
The blue moves are not all great moves, they just agree with Stockfish's choice of best move. NONE of the moves are actual blunders except White's move 34, the ? and ?? indicated Stockfish's like another move more than just a little better. Note how often Stockfish lists the difference between White and Black as 0.00! This was a very drawish game until White's move 34. The game itself is annotated here:
https://www.chess.com/forum/view/game-showcase/outside-passed-pawn-wins-great-endgame
MickinMD wrote:
Is there even a such thing as a perfect move? Maybe the perfect move is the move that challenges the opponent the most. But how does a perfect chess player define challenging the opponent, if he sees the game to the end no matter what the opponent plays on the next move and every move after. Maybe a move that will barely be able to draw if the opponent plays perfect, but almost all the other variations are winning, is the best move. Because from the perfect player's point of view, it's going to be a draw anyways.
Good point. I just finished a game where Stockfish 8, 20 ply, rated both sides moves as shown below (I had Black). For the vast majority of moves, Stockfish considered several moves to be essentially as good as the "best" move. So there are often several moves that are as "perfect" as each other. They equally develop, threaten, protect, etc.
The blue moves are not all great moves, they just agree with Stockfish's choice of best move. NONE of the moves are actual blunders except White's move 34, the ? and ?? indicated Stockfish's like another move more than just a little better. Note how often Stockfish lists the difference between White and Black as 0.00! This was a very drawish game until White's move 34. The game itself is annotated here:
https://www.chess.com/forum/view/game-showcase/outside-passed-pawn-wins-great-endgame
Strangemover wrote:
Hindsight usually.
Exactly. In this thread, for instance, I analyze in hindsight how Carlsen could have won his first WCC game against Karjakin:
https://www.chess.com/forum/view/game-analysis/how-carlsen-could-have-won-game-1-of-the-2016-wcc