Author: Roger
Date: 12:04:01 01/29/00
Go up one level in this thread
You could just throw out the draws. You still get the significance that program A is better than B. Roger On January 29, 2000 at 13:43:37, Christophe Theron wrote: >On January 29, 2000 at 11:58:36, Steve Maughan wrote: > >>Michael >> >>>Thanks for the offer, although I'm one of those guys who's on a Mac, and so >>>cannot use it. >> >>:-( >> >>>Just a question, does your program take into account draws as well, or just >>>wins and losses? >> >>The binomial distribution works on integers and not fraction so only win / loses >>are accounted for, but one can interpolate the likely scores. >> >>Steve Maughan > > >I'm not sure I understand. But if you don't take draws into account, then your >results are not OK. > >For example, when there is a very high probability of draws, then the actual >error margin is much smaller than normal. > >If the probability of draw is 0 (as it seems to be in your program), then your >error margins are going to be overevaluated. > > > Christophe
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.