Hello everyone,
I would like to know if this position is a draw.
8/4k3/4p3/2p1Pp2/2B2P1p/4b3/7B/5K2 w - - 26 61
Stockfish seems to see an advantage but I cannot find a winning idea for white.
Any suggestions for the best try for white to win?
Thanks in advance for any comprehensive evaluation of this position.
Yes, this is a draw.
Stockfish just counts material to come to an illusory advantage.
It would be won for white if white could exchange dark ksquare bishops, but there is no way to acomplish that.
@tpr I was surprised by Stockfish's +2.3 even though materially it is just a Bishop for two pawns, but appareantly it weighes the Bishop pair as well. But considering that white's dark square bishop is useless and boxed in there is nothing to do for white, I guess.
A high number which doesn't change means nothing. It's no ounce better than a constant 0.00. Fortresses have high constant numbers.
The derivation has to be positive or in other words: a good position which cannot be improved is not a good position.
It's a "theoretical draw" in practice the side with 2 bishops has winning chances. Black must defend accurately.
The evaluation will remain positive for one side until Stockfish spots a move or sequence of moves that forces a draw via one of the various methods, at which point it will go to 0.0 I think. And this makes sense because, even if it is theoretically a draw, if black were to make a small mistake opening the position up, white would be in a position to capitalize. So until there is literally a draw on the board, the evaluation makes sense.
A position which is -seems to be- good (+x) and can‘t be improved is not good (0.00).
Regardless of your theoretical definition, it does not explain what the engine analysis shows in practice, which is the very topic of this thread. In this particular case, it basically amounts to engine depth, we all know it is a drawn position but the analysis here is not deep enough that Stockfish can prove it.
But if we are going to get theoretical, it can be improved - by a mistake by your opponent. And this is true of any position at all, ever. You can never make a move that improves your position - you either make the best move, or one of multiple basically equally strong moves; or you make the position worse. You can try to set traps, but it is still dependent on your opponent making a mistake to fall into them. The mistake by the opponent in the endgame above would be a pretty bad mistake, far worse than some pawn move in the opening that comes back to bite them on move 35 or whatever, but nevertheless your opponent has to make a bad move at some point for your position to improve.
Stockfish's evaluation merely exists for:
1. candidate move selection (and debugging thereof)
2. UCI protocol compliance
Any evaluation other than a mate score is subjective.
"I don't believe in fortresses"
