[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-gnubg] cube decision ratings
From: |
Joern Thyssen |
Subject: |
Re: [Bug-gnubg] cube decision ratings |
Date: |
Sat, 14 Sep 2002 17:04:08 +0000 |
User-agent: |
Mutt/1.4i |
On Tue, Sep 10, 2002 at 04:04:06AM +0200, Robert-Jan Veldhuizen wrote
> Hi,
>
> GNUBG now counts only "actual or close" decisions for cube skill. This
> usually reduces the number of decisions by a lot, and as such tends to
> increase the error per cube decision also by a lot.
>
> What is the current threshold for "close" cube decisions?
equity ( no double ) > 1
or
equity ( no double ) - equity ( double, take ) < thr
or
equity ( no double ) - equity ( double, drop ) < thr
where thr is 0.25
> Perhaps the text ratings for cube decisions should get different ranges
> now, cube errors per decision are typically much bigger now?
Yes, that was why I introduced it. Trivial cube decisions shouldn't be
included in the ratings. For example, cube decisions in money game where
you have 4 chequers on the bar.
> My ratings dropped a lot :-)
:-)
Jørn
--
Joern Thyssen, PhD
Vendsysselgade 3, 3., DK-9000 Aalborg, Denmark
+45 9813 2791 (private) / +45 2077 2689 (mobile) / +45 9633 7036 (work)