Author: Dann Corbit
Date: 11:57:36 07/29/99
Go up one level in this thread
On July 29, 1999 at 14:22:03, Herman Hesse wrote: >On July 29, 1999 at 13:39:07, Dann Corbit wrote: > >>I don't believe that you should directly link winning percentage and ce. > >Should ? > >>Whenever I add new games to my database, I see that winning percentage changes >>for each opening position. Now, has the value of the position changed any? > >No, simply the percentage is now more accurate (accepting the obvious >inexactitude). Or less accurate, depening upon how the players fared and their ability and a thousand other random variables. >>Certainly not. There clearly should be some relationship between a good move >>and winning percentage by good players and computers. But I don't think there >>is an invertible function between the two numbers. > >You don't think ? Actually, I *know* that such a function does not exist, since no mapping can be one to one and onto for a vector valued function described as above with the limnitiations I have clearly disclosed. >Please use logical constructs to illustrate rather than these bland opinions. No >purpose is being served by all opinion and no content. > >> >>When choosing a chess move, I think the computer's evaluation of the position >>should be considered together with win expectancy from your database. If both >>agree on what is the best move, then go for it. If they disagree, maybe we >>should look harder at subsequent alternatives. > >Didn't you miss the point of the Amir Ban post? That instead of counting pawns >at each node there may be a different way to represent the nodal value as >win/loss chance. Your belief in non-invertible function between the two values >doesn't seem useful or relevent. Here is the entire quotation from Amir: "A correct evaluation is one that matches the winning percentages of the position. I think white has about 54% in serious play, and if so the evaluation should be about +0.20." I disagree that winning percentage is the correct and only variable to map to centipawn evaluation.
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.