Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: True Ratings: there is no such thing

Author: Mike Hood

Date: 03:03:01 01/06/04

Go up one level in this thread


On January 05, 2004 at 14:15:33, George Sobala wrote:

>Please get away from the misapprehension that there is such a thing as a "pure"
>and "absolute" true rating. It does not exist. One's rating is _entirely_
>defined by the opposition one plays. Thus Fritz's anti-computer rating may be x
>and its anti-human rating may be y, and they are both equally true and valid
>numbers, against that particular type and mixture of opposition.
>
>Thus there is _no_ way to translate an SSDF rating to a FIDE rating, because
>strengths against one type of opponent may be weaknesses against another.

There IS a way to correlate ("translate" is the wrong word) the SSDF ratings
into FIDE ratings. A significant number of verifiable games has to be played
between computer programs and humand of differing strengths in order to unite
the pool.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.