And my rating is 700
It'll soon be 850 or more, but I failed 100 games because I didn't pay any attention
Accuracy Score in the Analyzer
Sort:
Accuracy is not as high in blitz games as suggested above. Yet, all many of my 1500 opponents seem to be in the 90s and play like clockwork. I don't know if its notes or electronic assistance or just playing lots of masters on way up but something is amiss. Just not worth aggravation to play here. But lots of good resources and streamers. Ty for good content. 2 bad about community ethics.
Sometimes I'm sitting, analyzing a game, stroking my beard. My accuracy rate is always in the 90's, and less frequently in the 80's at times. And it maintains that pattern even when a game is riddled with so many mistakes and inaccuracies. Does this mean that the accuracy score is measured against the opponent and derived in such a way?
May 31, 2020
0
#26
SchaakVoorAlles wrote:
So far as I can tell the accuracy scores are unhelpful, or plain wrong, and useless.
I just won a game in which I made:
- 2 excellent moves to my opponent's 1
- 7 good moves to my opponent's 2
- 2 inaccuracies to my opponent's 4
- no mistakes (to my opponent's one)
- 15 best moves to my opponent's 17 (the only category where he did better)
His accuracy score was 93.1 to my 92.5, yet anyone that saw the game would agree that I played much better chess than my opponent.
This makes no sense at all. "Best" moves seem to be very overvalued compared to everything else.
I have played games where post mortems with Fritz and Stockfish at home tell me I played terribly, with almost every move substandard (if not an actual mistake) yet chess.com gives me accuracy scores in the high 90's.
What is going on here?
The accuracies given could just be medians and exclude outliers that could screw up an advantage for a player.
It also might help to indicate what the time controls were for these games.
May 31, 2020
0
#27
SchaakVoorAlles wrote:
WilliamJohnB wrote:
The accuracies given could just be medians and exclude outliers that could screw up an advantage for a player.
It also might help to indicate what the time controls were for these games.
I don't understand what you are saying about outliers.
Time control was 3 days per move, but I usually took only a couple of hours.
I really do not see how 2 additional best moves can outweigh two inaccuracies and a mistake as well as one additional excellent move and five additional good moves by me. Nor do I see how the scores can be so high when at home the engines are telling me that I did not play very well at all. In fact some moves that chess.com gives as "best" are considerd slight errors by the engines I use at home. I understand why Fritz might have a different opinion, but not why Stockfish disagrees. I thought chess.com used Stockfish as its analysis engine.
I have scored many accuracies of 97+ and 98+ for games that are pretty poor.
What is the point of rewarding poor chess with scores that make it look as if we are all IMs? It might flatter weak players, but it is of no help to anyone that is serious about improving.
By outliers, I am referring to games where one really bad move was made and cost the player the game or cost them a win and the other 25-30 moves in their game were either excellent/good.
In reference to the time control, it would make sense that spending more time on moves would likely result in better play (given that playing ability is constant).
In regards to your other comments, I think it could be because the engine that Chess.com uses to analyze games and compute move accuracy is different from the ones that you are using. In addition, I think it depends on how long you let the engine you are using to analyze a position think about it. For example, a move that an engine would initially consider bad would be considered great after thinking about it for a few minutes.
I just won this game https://www.chess.com/analysis/game/live/5089990146 and am about 1500-1600 on blitz and rapid. apparently I get all best moves but still am .3 off. answers?
Duolingo_is_under_u wrote:
I just won this game https://www.chess.com/analysis/game/live/5089990146 and am about 1500-1600 on blitz and rapid. apparently I get all best moves but still am .3 off. answers?
Are one of those 'best moves' accompanied by an alternative? If the answer is yes, then you have your answer. It's possible one of those alternatives scored a little higher, but the difference must be quite negligible for the two to be presented side by side.
Jul 3, 2020
0
#30
dcb1970 wrote:
First let me say how terrific the analyzer is. The "retry" mistakes feature and ease of use of all the functions is really first rate. I do look forward to an API offering where I can pass a pgn with N games in it, and get back a nice .pdf or .xls file with stats and reports for all N games. Anyhow, the question:
What is the range of reasonable accuracy scores to expect in a slow chess game? I ask because I ran my 3rd graders quads through it today, and 2 of the games he had 95 and 97.5 accuracy scores. His opponent in one had 55, another 92. All the ELO ratings are 1400 - 1700. Curious what one expects for this metric.
I'm glad someone likes the chess.com analysis.
Jul 3, 2020
0
#31
Duolingo_is_under_u wrote:
I just won this game https://www.chess.com/analysis/game/live/5089990146 and am about 1500-1600 on blitz and rapid. apparently I get all best moves but still am .3 off. answers?
100% accuracy is impossible. a game with all book moves gets 99.9
ok cool!
Lol it all look like a joke to me.... I player a few player today itself and I am a 800 level player (no shame) and my opponents who are also 800 level some of them play with more than 90% accuracy. I don't know if the fair play policy is working fine or not. I am not even sure if they are cheating or not. But if they are cheating it is so disheartening to lose because of the cheating. I fell Lichess is doing fine in that respect. They have some bot who detect cheaters instantly if they make stockfish recommended moves more often then expected based on rating. I have been slowly migrating to lichess and I can see the rating jump there. Because chess.com promotes leaders board people tend to flaunt their rating and because of that they cheat. so it becomes more rewarding for people to cheat on chess.com. I am not sure if people can relate to it or not, but for beginners it is really important that they really play the same level a 800 level player cant beat engine recommended moves. If anyone from chess.com is listening, please do something about it. Humble request. thanks
Brilliant is just when the computer didnt see a move. Ive had the most obvious move be brilliant oon the website in the past. It was obviously nothing special its just the comp didnt find it. They are right it is only a matter of time you will get a brilliant move.
Jan 4, 2021
0
#36
blueemu wrote:
The CAPS accuracy metric is statistical, so the score for a single game doesn't mean much. It might not mean anything at all.
The average nuclear family in the USA has two-and-a-half kids, but if you went into a neighbor's house and saw two-and-a-half kids on the couch, you would probably find it disturbing, not normal.
There are far too many factors that affect the CAPS accuracy... whether or not the two players follow a known book line, how far out they follow it, whether the play is relatively forced (with few options for either side), whether one players makes several blunders (it doesn't take Bobby Fischer to see that capturing a free piece is a good move), etc.
The CAPS metric is only meaningful if it is applied to a large sample of games by the same player.
I wouldn't have said it any better, fantastic comment. You have to be a writer by day, chess player by night. Loved this explanation. 
I just got about 2000 OTB