Yay - I Got A 97.6 Accuracy - For Me That's Rare 😊

I get what you mean. I wouldn't call it "meaningless" because it is still a useful statistic. It's just not to be taken literally as a percentage of how often one finds the best move. It's more about how closely your choice aligns with the engine depth used for analysis.

"Accuracy" is still probably one of the best predictors of rating, but this is over large sample sizes and factoring in other analytics such as game shape for each accuracy. The chess.com "insights" tool shares this and more, but it's only available for diamond members.

But this is not even true. Chess.com's accuracy score is weighted by a ton of unrelated things, like players' ratings, to and beyond the point where they lose the meaning that you attribute to them.

The game above is a perfect example. The guy played a beginner's mistake in the opening: Qf3, which blocks the natural development of the knight and puts the queen to an uncomfortable square where it can be kicked. But the script takes into account that this is a beginner, and it gives a completely unrealistic 97 percent accuracy score.

one time i got 98.9 accuraccy

Understandable @magipi and maybe opponent rating should be weighted less than it does, but chess.com wanted this to be a statistic more for beginners to understand if their moves were good for their level or not. IM Danny Rensch talks about that well in this video here:

That's rare for almost anybody, by the way. Top players are maybe 98% per solid game -- 99% at best. Everybody else is 80% or 90%, etc.

However, it's more impressive to look at an entire match or peak period.

For example, Fischer came back in 1992 after 20 years of not really playing, and had about 98% in the 1992 rematch. Elo roughly 2700.

On the other hand, Fischer's peak was almost 98% in 1971, putting him in #3 spot for best accuracy in history by World Champion, and year. Kasparov was #4 in 2001, for context. That's how great Fischer truly was.

On the other hand, Carlsen is known for playing almost perfect Chess time and time again. Fabi as well. And Kramnik is not far behind.

I suspect you may be thinking of a CAPS1 accuracy. CAPS2 is more demanding.

Here is the review of game one of the 1992 rematch. 94.4% accuracy stat.

Here is game two. 92.5%.

And so on.

Hello

magipi - 

I've been a member here for almost 8 years, and from the beginning of that 8-year timeframe have been playing the blitz games (with a rating of over 1,000). I had regularly played blitz games up until two years ago. And it was just a month ago when I started back on the blitz games. (However, I do not consider my current rating a reflection of a beginner's skill level because after having been absent from the blitz games for two years along with other personal traumatic reasons, I've lost my ability to focus. Hence, I came down from an 800 rating to 300+ just within the last month.)

Regardless, one important factor that should be made known to you -- and that is:

THE CHECKMATE I PLAYED HERE (REFERENCED IN MY POST #1) IS THE KNOWN: SCHOLAR'S MATE. APPARENTLY, YOU ARE NOT FAMILIAR WITH THIS 4-MOVE CHECKMATE BECAUSE YOU SAID:

"The guy played a beginner's mistake in the opening . . . " 

THIS EXACT 4-MOVE CHECKMATE IS TAUGHT BY GRANDMASTERS AND COACHES. LOOK IT UP ON THIS SITE -- Scholar's Mate. AND I LEARNED THIS SCHOLAR'S MATE HERE ALMOST 8 YEARS AGO. HENCE, EVERY 4 OF THE MOVES IS NOT A MISTAKE!

By the way, because of your assumption that I'm a "guy" (without having to ascertain what my gender is by just making a couple of clicks as it may offend some people), I AM FEMALE.

Recently I played the best blitz game of my life

98.1 percent accuracy

Only ten moves thoughh

I think my best thirty move game accuracy is like 94.5 or something