Author: blass uri
Date: 15:33:08 07/15/00
Go up one level in this thread
On July 15, 2000 at 18:17:35, Mogens Larsen wrote: >On July 15, 2000 at 17:43:47, blass uri wrote: > >>You can. >>You cannot say about one of them that it has more than 2500 but you can say that >>the best of them is better than 2500(You do not know which one of them is the >>best because you have not enough information) > >No, you can't do that with statistical significance. As soon as you pick a >program the uncertainty rises and the claim of GM strength vanishes. If you >consider programs as a whole, some programs will have ratings above and below >2500, but that's irrelevant if you make a claim about a specific program. And a >claim based on several programs is obviously a ludicrous thought. I did not make a claim about a specific program but only that data can prove that there is a program with rating>2500 without telling me which program. Uri
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.