Re: [Bug-gnubg] Snowie error rates versus gnubg error rates

[Joern Thyssen]

On Sun, Apr 06, 2003 at 05:36:34PM -0300, Albert Silver wrote
> Just one comment: another fundamental difference between the way the two
> programs calculate ratings is in the inclusion or not of unforced moves.
> Snowie definitely includes them, which would also ease the way grades
> come out. I had a discussion on this at the forum at the Snowie site and
> was told that if they didn't, we'd all be judged as beginners all the
> time. I'm not going to enter this discussion as I think NOT including
> unforced moves far more intelligent, however, if this difference isn't
> somehow taken into account then GNU is not going to be the same as
> Snowie, it's going to be a LOT harder. 

The factor of 1.4 is an attempt to do just this. 

>From my sample of 300 matches the conversion factor between the
Snowie-type and the gnubg-type of error rates were approximately 1.4, so
this factor should take into account this difference. 

> I mention this because it just
> seems to me, from empirical experience alone, that 0.0083 as the bottom
> limit of Expert and 0.012 as the bottom limit of Advanced seems a little
> strict. I know Snowie has in practice been stricter in its grading, but
> I didn't get the impression it was THAT strict. 

Apperently it is :-) With the new threshold gnubg should be, on average,
equally strict as Snowie.

Jørn