Unbalanced Human Openings V2.00 released

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

User avatar
Werner
Posts: 2864
Joined: Wed Mar 08, 2006 10:09 pm
Location: Germany
Full name: Werner Schüle

Re: Unbalanced Human Openings V2.00 released

Post by Werner »

Hi Larry,
why do you think so?
Does this happen too allthough each opening is played with black and white ?
My ideas is there are less draw games when using unbalanced openings.
Werner
lkaufman
Posts: 5942
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA

Re: Unbalanced Human Openings V2.00 released

Post by lkaufman »

Werner wrote: Wed Feb 17, 2021 6:19 pm Hi Larry,
why do you think so?
Does this happen too allthough each opening is played with black and white ?
My ideas is there are less draw games when using unbalanced openings.
Yes, it reduces draws a lot. This is fine if all testers in a group use it, but if some testers use it and others use normal (drawish) openings, the elo differences will pretty much just depend on the relative number of games played by the two book types. Just for example, if you are testing a new top level engine like SF or Dragon, it will get a much higher rating if a high percentage of games are played with UHO than otherwise. Of course if all engines were tested with say half UHO and half normal openings that would be fine, but that is unrealistic given that each tester makes his own choice.
Komodo rules!