Author: Gian-Carlo Pascutto
Date: 10:55:58 05/08/01
Go up one level in this thread
On May 08, 2001 at 11:09:19, Larry Proffer wrote: >If the objective of the match is to 'find the best program to play Kramnik', >then what matters is that *all* the possible best programs are selected to >compete, no? > >If a program that could be 'best' doesn't compete, then .... ? As Marty has already answered, the match has no statistical meaning for determining 'the best program to play Kramnik' in that case. I really hope there are no people left on this forum who actually believe it has. >Number of games is irrelevant. Depends on what you want to establish. For the goal mentioned above, no, as it has no statistical meaning anyway. As for determining who is the stronger of Fritz/Junior, yes. We can now tell (without doing the acutal math 'cos I'm lazy :) with a high degree of confidence that Junior and Fritz are of nearly equal strength. >The 'validity' of the winner being 'best' is affected only by the pool of >programs used. This is false. If a sufficiently large number of programs participate (in this case, all that could possibly beat Kramnik), then the number of games gets significant (among other things). >Exclude a program, or exclude three programs, and the 'validity >of best' crashes in inverse proportion to the 'Chessbase win chances'. This is proven false by example. Assume a pool of Crafty, Fritz and Junior. Now exclude Fritz from the pool. The validity of best has decreased as have the Chessbase win chances. -- GCP
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.