Author: Daniel Jackson
Date: 12:42:30 10/02/04
Go up one level in this thread
On October 02, 2004 at 04:59:01, Chris Welty wrote: >On October 02, 2004 at 00:13:09, Daniel Jackson wrote: > >>Bet you wouldn't believe that a stronger program could lose 10 straight games >>yet it has happened. > >Of course it will happen if you play enough games. Yes, and it has happened right from the start! I remember Bruce Moreland being surprised by this nearly 4 years ago. > >> Simply put there isn't enough games to be certain of anything, >>but it appears that these settings might be the reason...and I say might...we >>don't know yet.. >> >>I don't know and you don't know....could be book wins and losses...we don't >>know, we have no games to look at...see to see why this result occurred. > >We can never be 100% certain, however enough games were played that the result >is statistically meaningful. It is not within the margin of error. > >More games always help of course. Yes, a lot more games...I don't think we disagree as much as you think, and the point is you need to be over 95% certain or better IMO. The basic results do look good, so we'll see...
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.