Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Match results - a complete(!) theory (long)

Author: Peter Fendrich

Date: 15:49:54 02/18/01

Go up one level in this thread


On February 18, 2001 at 18:25:05, mike schoonover wrote:

>this is all well and good,but what does it mean when you through
>in the theory,even a blind squirrel can find a nut sometimes!
>i have seen GNU wipe out crafty17.3 playing the black side of
>an alikine,where black lines up his white squared bishop on the h6
>b1 diagnle.
>regards
>mike

It means that you are never 100% sure of anything...
Suppose that you arrange a 10 game match with no prior knowledge whatsoever
about the players and get the result 7-3 (win=6, Draw=2 and Loss=2).
Claiming that A (who won the match) is better than B will get the probability of
91% to be true. The risk of claiming the wrong thing is 9%. That will happen now
and then...
Less games, more draws and/or closer to even scores will increase the risk of
error for such a claim.
Peter



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.