Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Uri, your prediction will be close or exact !

Author: Osorio Meirelles

Date: 08:45:25 11/25/02

Go up one level in this thread


On November 25, 2002 at 07:08:19, Jorge Pichard wrote:

>On November 25, 2002 at 07:06:05, Jorge Pichard wrote:
>
>>
>>I predict 21-19 or 21.5-18.5 for tiger15.
>>
>>18-16 suggests that tiger15 is slightly better so this is the reason that I
>>predict 3-3 or 3.5-2.5 for tiger15 in the remaining games.
>>
>>Uri
>>
>>
>>A 2 points or even 3 points difference doesn't indicate that Chess Tiger 15 is
>>better than Deep Fritz, we need at least a 5 points difference in a short match
>>of 40 games ,or a 10 points difference in a 80 games match. After all the
>>programs have played each other, only then we can estimate the real difference
>>between Chess Tiger 15 and Deep Fritz.
>>
>>Pichard.

 Given that the standard deviation is proportional to the square root of the
number of games, then a 10 points diference for 80 games, would be the
equivalent to a 5 point difference in 20 games.  I also assume that this depends
somewhat in the distribution of wins and draws ( if there are a great number of
draws, a 3 point difference might be significant ).  Do you agree ?









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.