Author: Dann Corbit
Date: 15:35:44 05/11/04
Go up one level in this thread
On May 11, 2004 at 18:13:16, Uri Blass wrote: >On May 11, 2004 at 18:09:19, Dann Corbit wrote: > >>On May 11, 2004 at 02:40:57, Uri Blass wrote: >> >>>On May 10, 2004 at 20:48:10, Dann Corbit wrote: >>> >>>> Program Elo + - Games Score Av.Op. Draws >>>>1 Ruffian_210: 2829 320 378 3 83.3 % 2550 33.3 % >>>>2 Ruffian_202: 2741 409 382 2 75.0 % 2550 50.0 % >>>>3 Crafty-1912: 2557 201 278 9 38.9 % 2636 33.3 % >>>>4 Rebel12_CB : 2550 600 600 2 50.0 % 2550 0.0 % >>>>5 List512 : 2359 382 409 2 25.0 % 2550 50.0 % >>> >>>I do not understand the reason that Crafty's rating is not equal to Rebel >>>when Rebel got exactly 50% against Crafty and did not play more games. >> >>They are equal. In fact, surprisingly so given the volume of data. > > >2557!=2550 so I do not see that the rating are equal. 2550 +/- 600 2557 +201/-278 Think of it as interval arithmetic and you will see that all is fine.
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.