What accuracy % do you consider a “good” game?

For me, good accurace is 90-95.My game with most accuracy is 98.2 whis is very good even for super grandmasters.

I want to lower the bar. Anything greater than 66.6 is a good rating.

imo, 85% and above is good enough. (I had 95% before)

My highest accuracy was 94.6, I got it today, I'm a 775 rated player, am I a prodigy or just lucky lol

I agree. You ruined it somewhat by your second sentence, but still.

Accuracy is meaningless trash. It can be used for nothing, especially not for something complicated like determining a "good game".

Probably not a prodigy, neither pure luck. Just a relatively short solid game where you could just take what your oponent gave you.

It's remarkable how slowly accuracy improve going up in elo (at list in fast time control) . Just because stronger opponents give you tougher positions to play.

My idea of a good game is one in which BOTH players play pretty well. It's hard to link it to accuracy, but let's say no more than ten or fifteen points spread between the accuracy scores of the two opponents.

Winning because of an opponent's one-move blunder does score the full point, but it's more fun to win because he had Kings and you had Aces.

It depends completely on the type of game which was played. If it is a completely positional game then its stupidly easy to get high accuracy scores since many moves will work... It doesn't really matter if u play Nc3 first or play b3 Bb2 first.

i‘ve played 2 games with 100% accuracy. Is that good for being rated 900?

it says 1200 on the Account, but i‘m 900 rapid

I played %94.3  but I wasn't logged in so now I am sad.

Time control also mater's a lot

75

I think around 80%

I sometimes play a computer opponent that is consistently hard for me to beat, and then just use hints the whole game;  I try to guess the move before a click on the hint icon. I hope to learn this way. I just did it, and what I find odd is that I got a 93.3 accuracy. You'd think that it would be 100%.

A win is a good game.

Accuray doesn't seem to be very relevant because the artificial intelligence validates the moves that are probably valid in the sense that there is no rebuttal, something detected as not good enough compared to the data put by the programmers.

If for example I play with White after 1) e4 d5 2) e5 this move is a strategic opening error but the position is not losing or does not present anything concrete to make the AI says that it is not terrible 2. e5, so 2.e5 is validated because AI didn't see anything wrong. Now with Black after 1) e4 d5 2) e5 e6 this move is also a strategic opening error but the AI sees good opening of the French Defense and thus AI validates but in this precise case it was badly played.

In my opinion there are many moves that the AI validates but which are from a human point of view bad because they have no strategic idea or an idea that is not adapted to the position which for example transforms an advantageous or dangerous position [for the opponent and for us] into a more or less equal position but which by calculation is more certain that our position will not lose. 

I got 100 befor

I favor whatever degree of accuracy the game I'm winning has.

I use centipawn loss average from Stockfish analysis to determine how accurately I played. 

I've stopped caring about chess.com accuracy after I read somewhere the chess.com accuracy gets tweaked depending on the elo bracket you are in. I dont think it's very relevant.

The chess.com accuracy figure is just a chess.com thing - in analysis, the standard metric is centipawn loss - to either evaluate moves individually, or a sequence of moves.