How wrong is SF15.1's WDL model ?

General discussion about computer chess...
Post Reply
User avatar
deeds
Posts: 754
Joined: Wed Oct 20, 2021 9:24 pm
Location: France
Contact:

How wrong is SF15.1's WDL model ?

Post by deeds » Wed May 31, 2023 8:06 am

After 1.g4,

The SF15.1's WDL model evaluates the black's winrate at 59.5% (at D40) :
Image

SCID announces the black's winrate at 93.8% (after 500 games at TC 1m+1s) :
Image

Post Reply