Author: Dann Corbit
Date: 11:59:55 01/28/99
Go up one level in this thread
This list should be formatted in the same way as the SSDF. Much critical information is missing. Especially number of games and +/-, opponent strength, etc. But if you examine the SSDF list and see the range of probability for the strenght, it is believable. Since the SSDF list is so much better, I don't see a lot of value of this list unless it gets reformatted to contain more information. Also, I would like to see both one and two standard deviations posted for the strength figures. That would give a much more realistic estimate of how strong the programs are next to each other. If the curve is skewed, kurtosis might also be interesting.
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.