Why Do Other Chess Players Worried About High Accuracy So Much?

Hmm. Well I don't know what you mean by your achieving high accuracy in most games. What do you mean? Above 90? 80? 70?

You play bullet, so I wouldn't expect you to have the same level of accuracy as a player who plays rapid.

#2. Above 90% I mean. Facts.

#4. Right.....

I don't think that is true at all.

https://support.chess.com/en/articles/8708970-how-is-accuracy-in-analysis-determined

Because it is precious:

Accuracy is a good metric so if players are concerned with high accuracy it's simply because they want to play well & like the feedback the metric provides.

I got 95% against martin, clearly I am a genius.

In your last 20 games you have gotten exactly 0 above 90% accuracies and one 34% accuracy,

do not brag about getting 90% accuracies if you do not get them, it makes other players feel bad about themselves.

#11. Sorry about that. I didn't mean to brag about it.

Looking at the link, the way they've biased the numbers, accuracy scores are a better way of telling which are your best games rather than whether you're a good player. For example, the graph was derived from players in the rating-range 1000-1500, and peaks at about 82% accuracy. I am a 500 player, and over my last 20 games, have averaged an accuracy of 78%.

So a rating increase from 500 to 1250 corresponds to an increase in accuracy of just 4%. (admittedly this is based on me, one single point, hardly a statistically-significant analysis). On the other hand, my bad games drop to 55% over that range while my best was 95, so I've got a 40% range telling me which games are most awful!

The implication is that there's really not much difference between a 1200 player and a 500 player, compared to the difference between a 500 player on a bad day and a 500 player on a good day.

#14. That really makes sense. Thank you for your input.

I dont think Ive ever really cared about my accuracy, the only real metric im concerned with is my blunders.

I usally only look at my blunders or misstake when I analyse. I do not pay much attention to accuarcy %.

Accuarcy % can be a ego bost, but it do not really teach you anything looking at the accuarcy score. Looking at misstake can teach you something.

I suppose one big blunder in an otherwise accurate game can mean you lose immediately, so if you care about winning, looking at blunders makes sense. So does noticing big missed opportunities. I don't know how these are reflected in accuracy scores?

The other one is inaccurate-but-doesn't-matter play: if you've a bishop and knight and a couple of pawns versus a lone king, most humans will just promote a pawn or two and play the ensuing queen versus lone king easy mate ending, even if it takes another 10 or 12 moves. A computer will see a clever mate in 3 or 4 using the knight and bishop. So from a computer's perspective, the human played less accurately. But both the human and the computer were sure they'd win, and there is no prize in chess for winning quicker.

I tend to use the review function mostly to check the validity of what I did (did I get lucky by blundering and my opponent missed? Was my wonderful killer move really a killer or was there a perfect escape I never saw?), and to check for missed opportunities (how did I miss that mate in 2 that could have happened 16 moves before I finally offered a draw?)

Me like apple