Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: What ELO is perfect chess?

Author: Dann Corbit

Date: 14:02:13 02/22/02

Go up one level in this thread


On February 22, 2002 at 16:52:13, ALI MIRAFZALI wrote:

[snip]
>Your problem stems from the fact that you (erroneously) believe that the
>perfect player would beat Kasparov 100 percent of the time and therefore his
>rating would be so so high.

We'll remove Kasparov from the pool, and have him play against me all the time
instead.

>Also we are talking about 1.Computers 2.Humans
>3.Relative measurements.I never said the faster you play the higher your elo
>goes.It is a Comparison.If a Computer is 2200 elo at 40/2 and a human player is
>also 2200 at 40/2 in aBlitz match the 2200 elo 40/2 computer would beat the 2200
>elo 40/2 human.This became apparent very early when chess 4.5 (late seventies)
>began beating masters at 5 min games but had less  succsees in slow chess.

Only if the distribution is gaussian.  But what if we are talking about a gamma
distribution, or even a poisson[1]?

[1] Not to be confused with a "fish" distribution[2]
[2] Or even the baskets of remants that were picked up afterwards[3]
[3] It was a dark and stormy night, and the ship was at sea, so the captain
said, "Jack me' boy -- tell the men a story."  So Jack began his story {goto 3}



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.