The question could be made more interesting by rephrasing it to read: What percentage are wins as compared to draws?

As @77and33is100 notes, the winning percentage for black and white is the same because of symmetry. Without having checked I would strongly suspect that only one version is stored in the data.

Of course mate in 5 is not worse than mate in 3 in most contexts, particularly in most games. I was specifically thinking of the more theoretical discussion of say endgame tablebases, where you could see statements like: "this position is mate in 13 with perfect play".

Apart from losing a piece (which is of course mroe than enough reason), bxc3 also has the advantage to support the d4 pawn and thereby strengthen the center and limit the scope of the bishop on g7. Furthermore leaving the queen on a3 you still keep an eye on the pawn on e7 which would be double attacked if the Nf6 moved.

@David That is not the usual definition of perfect play. For instance If you had forced mate in 3 and afterwards you have only a forced mate in 5, this would (usually) not be considered perfect play.

@user21820 Do you have an example game in mind, where it is only sharp/tactical for one side?

@stevec True, and while often humans can exclude a large number of variations easily, in this position, except for queen sacrifices, there are not a whole lot of moves to be excluded at first glance. That's what makes this problem particularly difficult.

Python-chess is a library that can find all legal moves, make moves on the board, etc. It should be easy to write a loop to count all combinations.

How do you count transpositions? E.g do you count 1. Ke2 2. Qd2 and 1. Qd2 2. Ke2 as two or as one?

If this is so, why can't we say that ratings also measure the absolute strength?