Chess.com's algorithm deducts accuracy points for having a god awful avatar.
ong, get a W pfp like me @tigger
Chess.com's algorithm deducts accuracy points for having a god awful avatar.
ong, get a W pfp like me @tigger
you know you can completely outplay your opponent and get a 95% accuracy but then blunder one move and resign the next and the stats look all "messed up"...?
accuracy and rating estimator are just algorithms. they aren't human calculations and they depend on the closeness of your play to the engine's top lines. so theoretically you can make 30 best moves but lose to someone who made 5 blunders. the accuracy will be higher but you still lost. likewise your estimated rating will be higher but again, you still lost. hopefully that helps
They dont show the real mistake blunder evaluation elo. Thats shows the real rating if you want to trust in yourself!!
They dont show the real mistake blunder evaluation elo. Thats shows the real rating if you want to trust in yourself!!
Does this mean anything? Did you miss out some words or what?
Well imagine two drivers racing around the track the first driver is smooth all the way round to the last corner then loses control. The second driver is ragged all the way round but is close enough on the final corner to pass and win the race. The accuracy is how well they used the racing line. The blunder is the spin on the last corner and has a major detriment to the accuracy as the car no longer follows the racing line. So you both played a similar number of categorised moves, but the moves may have had a greater accuracy value for that one side cumulatively for the options available to them and then a big drop off at the end. The 1200 and the 2200 are a different scoring system. Clearly the blunder was so bad it dropped one rating and inflated the other.
I once had 83% accuracy
I won
The guy had higher rating (given by only 200 points, my rating is sh*t)
And I suck
How the hell did I have a 83% accurate game bruh 😭
Chess.com's algorithm deducts accuracy points for having a god awful avatar.
true though
Bro had one blunder, you had none, yet the algorithm thinks "huh this guy played 0.2 points better"
Why does no one understand accuracy? Explain to me what "0.2 points" better is.
Side point: We are people-the best move is not always the best.
you know you can completely outplay your opponent and get a 95% accuracy but then blunder one move and resign the next and the stats look all "messed up"...?
accuracy and rating estimator are just algorithms. they aren't human calculations and they depend on the closeness of your play to the engine's top lines. so theoretically you can make 30 best moves but lose to someone who made 5 blunders. the accuracy will be higher but you still lost. likewise your estimated rating will be higher but again, you still lost. hopefully that helps
i wasn't focused that he lost and had a higher accuracy, I'm saying that our moves were the same skill for basically the whole game (same best moves, same book moves, same good moves, etc) except my excellent to his blunder and somehow the blunder is better than excellent
Bro had one blunder, you had none, yet the algorithm thinks "huh this guy played 0.2 points better"
Why does no one understand accuracy? Explain to me what "0.2 points" better is.
Side point: We are people-the best move is not always the best.
you have to be serious
He had 83, the opponent had 83.2
0.2 points more
Cheaters play stupid moves at the start and waste time, then they become super strong, I think this chess engine can't detect. How many times do they become with 10 sec on clock moving 100% engine moves?
So the engine says they are stupid, but they actually destroy at the end game.
Its cheaters strategy.
At the end of the game they are better than Magnus but at the start of the game they make blunders
That's how they cheat
how in the world does he... have... a... higher accuracy...
by the way on this game, the estimated ratings thing was just bugging out like crazy, it didn't show the game phase rating for some of the phases, and it said i was 2200 and my opponent was 1200: I made sure that it wasn't analyzing from move 1 too (cause that's what i thought at the beginning cuz one of his first moves was a blunder) but i specifically even analyzed the whole game over again and it's still broken
Ive thought about this sometimes and I've concluded personally that it's because not all excellent moves are created equal(or any category). some of his excellent moves may have been stronger than some of yours
We don't know the accuracy algorithm. It hasn't been disclosed but we do know that it's not simply a weighted average of those categorised move qualities.
I would guess that it's total evaluation loss from the engine best move, modulated by move count and player rating, to give an aesthetically pleasing percentage score.
how in the world does he... have... a... higher accuracy...
by the way on this game, the estimated ratings thing was just bugging out like crazy, it didn't show the game phase rating for some of the phases, and it said i was 2200 and my opponent was 1200: I made sure that it wasn't analyzing from move 1 too (cause that's what i thought at the beginning cuz one of his first moves was a blunder) but i specifically even analyzed the whole game over again and it's still broken
Ive thought about this sometimes and I've concluded personally that it's because not all excellent moves are created equal(or any category). some of his excellent moves may have been stronger than some of yours
that was my original guess yeah
how in the world does he... have... a... higher accuracy...
by the way on this game, the estimated ratings thing was just bugging out like crazy, it didn't show the game phase rating for some of the phases, and it said i was 2200 and my opponent was 1200: I made sure that it wasn't analyzing from move 1 too (cause that's what i thought at the beginning cuz one of his first moves was a blunder) but i specifically even analyzed the whole game over again and it's still broken