Or somehow mathematically a checkmate will inherently imply in better scores of "blunders", "mistakes", and inacuracies?
Interesting question. My guess in a long enough game, with the losing side making only the "mate" mistake in a totally unforced scenario, this would happen. A backrank with a rook, down 4 queens down cannot possibly be the result of a better game management, I think
A self-mate game could also be a good candidate. Only last move forced, a computer wouldn't consider it bad till the very end, if a move does not lose advantage
A self-mate game could also be a good candidate. Only last move forced, a computer wouldn't consider it bad till the very end, if a move does not lose advantage
Personally I like Kolja te most... bat tats just me
Pretty sure it's impossible.
@repetitaiuvant
But in your scenario, no matter how much of a centipawn advantage the computer assigns to the "better" side, that advantage will be undone and then some by the move that results in checkmate.
If I make a billion dollars, then buy something that costs a billion and one dollars, the original billion dollars is gone. If I have a centipawn advantage of whatever number, but then get myself checkmated, then I have none of that advantage left and have instead given it all back, plus enough extra for me to lose.
@repetitaiuvant
But in your scenario, no matter how much of a centipawn advantage the computer assigns to the "better" side, that advantage will be undone and then some by the move that results in checkmate.
If I make a billion dollars, then buy something that costs a billion and one dollars, the original billion dollars is gone. If I have a centipawn advantage of whatever number, but then get myself checkmated, then I have none of that advantage left and have instead given it all back, plus enough extra for me to lose.
Chess is brutal. You can play a 'perfect' game against a guy just pushing wood, and then lose it all to a single absurd mistake. Like the saying games, the winner is the person who makes the second to last mistake.
I imagine there's probably some correlation, but no it's certainly not inherently true that the winning side is the one that made fewer mistakes.
I imagine there's probably some correlation, but no it's certainly not inherently true that the winning side is the one that made fewer mistakes.
Do you refer to centipawn loss or to inaccuracy/mistake/blunder? Relating to i/m/b it is pretty obvious that you can lose a game with only one blunder while your opponent can make an almost infinite number of blunders on the way.
So I just played myself in two different browsers, check this:
Basically I let "Anonymous" play the scholar's mate but blunder away the a1 rook on the way. "I" had to make an inaccuracy playing 3. ... a5 to allow "Anon" to blunder the a1 rook. And then obviously "I" overlooked the scholar's mate to lose the game.
white: 2/2/1
black: 1/0/1
1-0
So I just played myself in two different browsers, check this:
Basically I let "Anonymous" play the scholar's mate but blunder away the a1 rook on the way. "I" had to make an inaccuracy playing 3. ... a5 to allow "Anon" to blunder the a1 rook. And then obviously "I" overlooked the scholar's mate to lose the game.
white: 2/2/1
black: 1/0/1
1-0
@Chessty_McBiggins that would be true if checkmate has an "infinite" penalty. But in reality it is a finite number. I expect a big enough number of blunders on the winning side will eventually offset it.
Also, I think the centipawn loss is with respect to the "best move". It is not a conserved quantity that gets undone, if I'm not mistaken. There is an upper limit to how much centipawns a single move can lose
Also, I think the centipawn loss is with respect to the "best move". It is not a conserved quantity that gets undone, if I'm not mistaken. There is an upper limit to how much centipawns a single move can lose
I did some BS quickly. The cap for a blunder is +-10 pu, that means it approaches 2000 ACPL ((10+10)*100).
White has a lower ACPL.
White has a lower ACPL.
wonderful @Sarg0n. Counterexamples are always more powerful!
This topic has been archived and can no longer be replied to.