Author: Dann Corbit
Date: 10:56:44 11/18/04
Go up one level in this thread
On November 18, 2004 at 10:55:13, Rémi Coulom wrote: >Hi, > >I would like to understand the theory of the Elo rating systems used for >computer chess. I have searched for a while on google, and found references to >the name of Ken Thompson, but not his algorithm/theory. The docs of EloStat make >a reference to the CSS German magazine, but I could not find an electronic >version. I also found Mark Glicksman's papers, but they do not seem relevant, >since they seem to mainly deal with time-varying strength. > >So, any description of the algorithms used by the SSDF and EloStat would be very >welcome (I can read German). http://www.ratingtheory.com/ http://www.chessmetrics.com/ http://www.schackportalen.nu/English/erating.htm Here is the rating program of Royal Clifford Jones, Jr (first link) ftp://cap.connx.com/pub/tournament_software/prog10.cpp And permission to use it: ftp://cap.connx.com/pub/tournament_software/Re%20Your%20chess%20rating%20systems.txt You will find that the FICS server source code has the Glicko Elo rating calculation in it. Here is USCF Elo calculation: ftp://cap.connx.com/pub/tournament_software/USCF.C ftp://cap.connx.com/pub/tournament_software/USCF.H I wrote it from first principles and anyone is free to use it for any purpose they see fit.
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.