Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Who is better? Some statistics...

Author: Peter Fendrich

Date: 03:16:58 06/12/01

Go up one level in this thread


On June 11, 2001 at 12:37:37, Gian-Carlo Pascutto wrote:

>On June 11, 2001 at 12:34:54, Gian-Carlo Pascutto wrote:
>
>>On June 11, 2001 at 11:27:21, Peter Fendrich wrote:
>>
>>>Your 100-game example is OK of course but the table covered also as few as 10
>>>games.
>>
>>Actually, the 55-45 example doesn't match up with the table.
>>
>>I verified it with your program and in a worse-case-scenario
>>(all decisive results) a 56-46 result gives a probability of
>>88%, which is well below the 95% that Bas suggests.
>
>Bas wrote:
>
>Therefore the noise window is [45, 55] with 95% confidence.
>
>Since sigma is 5 and the window is with 2*sigma, this should
>be:
>
>Therefore the noise window is [40, 60] with 95% confidence.
>
>Which is very close to what the table suggests.
>
>--
>GCP

Yes, right. I'm too lazy to verify figures!
What I meant was that 100 games is enough if you want to approximate with the
Normal distr.

//Peter



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.