Comparing chess openings

Does someone know of a link, kaggle?? or another site.  
Where the 10 most common played openings are compared against each
For black and white.

Some matrix table with win/draw chance of each opening vs the other openings.

I'm just curious.

No, not really.

There is nothing that will tell you the win or draw chance.  There are statistics, but there is also standard deviation, which means the actual stats will not equal the mean.

 

For example.  I could flip a coin 50 times and get heads 28 times.  I could do it again and get heads 31 times.  I could do it a third time and get heads 28 times.  That is 87 heads out of 150 flips.

 

Does this mean that the greatest odds lie in you flipping 29 heads in 50 flips?  Simply due to statistics of past events?  No!  The result with the greatest odds is to flip heads 25 times.  That does not mean you can't flip 34 heads, or 28 heads, or only 19 heads!

 

So even a statistical chart of past results will not give you accurate win or draw chances/odds.

I think its pretty balanced

Well i dont think it is balanced, some openings have prefered black defences.
And some combinations are loosing. Although i do understand chess starts after the opening, the opening is a proper start the way you prefer to attack/defense which result in a good or bad mid game. While on top level black=draw 85% or so, on the other levels it would be interesting to see.

Check out ChessGames.com....

their openings explorer may give you what you are looking for...

Defending against the typical e4 player. Try the King’s Pawn Opening: Barnes, Fried Fox Defense, or the Reti Opening, Arctic Defence followed by Kf7. All leads to the bongcloud development.

I don’t know what you’re saying, Thriller. I actually have a minor in statistics and the mean is unrelated to the standard deviation. Furthermore, standard deviation is irrelevant with a categorization of three: it’s just win/lose/draw. It’s a categorical variable, not a discrete variable.

If the sample is large enough, the law of large numbers kicks in, and the mean reflects the “true” underlying win%, assuming that there is no systematic bias in the data. Usually there is systematic bias in that the databases only measure master games, so they probably say very little about average players.

And by large sample, again assuming no systematic bias, is just a few hundred to get a 99% confidence interval of + or - 3%. The data they have measured in the thousands.