Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty 18.15 vs Comet B45 Final score 22.5 to 13.5

Author: Peter McKenzie

Date: 15:16:58 06/12/02

Go up one level in this thread


On June 12, 2002 at 12:36:09, Dieter Buerssner wrote:

>On June 11, 2002 at 07:37:52, Peter McKenzie wrote:
>
>>Ask yourself this question:
>>
>>"If I ran this very same match 100 times, but with Crafty 18.15 vs Crafty 18.15,
>>then how many times would I see this score (22.5 - 13.5)?"
>>
>>What answer do you come up with?
>
>I am not Jorge, but I got an answer:
>
>E:\dcrand>cmatch 36 35 25 40 1000000
>Result of chess matches between equal opponents
>White wins 35.0%, black 25.0% and 40.0% draws
>A match of 36 games was simulated by 1000000 Monte Carlo tries
>
>               result       probability         <= this          > this
> 18.0 - 18.0  ( 50.0%):          8.582%          8.582%         91.418%
> 18.5 - 17.5  ( 51.4%):         16.799%         25.381%         74.619%
> 19.0 - 17.0  ( 52.8%):         15.748%         41.129%         58.871%
> 19.5 - 16.5  ( 54.2%):         14.058%         55.187%         44.813%
> 20.0 - 16.0  ( 55.6%):         11.950%         67.136%         32.864%
> 20.5 - 15.5  ( 56.9%):          9.670%         76.807%         23.193%
> 21.0 - 15.0  ( 58.3%):          7.436%         84.243%         15.757%
> 21.5 - 14.5  ( 59.7%):          5.508%         89.751%         10.249%
> 22.0 - 14.0  ( 61.1%):          3.878%         93.629%          6.371%
> 22.5 - 13.5  ( 62.5%):          2.555%         96.184%          3.816%
> 23.0 - 13.0  ( 63.9%):          1.646%         97.830%          2.170%
> 23.5 - 12.5  ( 65.3%):          0.992%         98.822%          1.178%
> 24.0 - 12.0  ( 66.7%):          0.557%         99.379%          0.621%
> 24.5 - 11.5  ( 68.1%):          0.309%         99.688%          0.312%
> 25.0 - 11.0  ( 69.4%):          0.164%         99.852%          0.148%
> 25.5 - 10.5  ( 70.8%):          0.084%         99.936%          0.064%
> 26.0 - 10.0  ( 72.2%):          0.036%         99.972%          0.029%
> 26.5 - 9.5   ( 73.6%):          0.018%         99.990%          0.010%
> 27.0 - 9.0   ( 75.0%):          0.007%         99.997%          0.003%
> 27.5 - 8.5   ( 76.4%):          0.002%         99.999%          0.001%
> 28.0 - 8.0   ( 77.8%):          0.001%        100.000%          0.000%
> 28.5 - 7.5   ( 79.2%):          0.000%        100.000%          0.000%
>
>So, assuming that on average white will win 35% of the games, black will win 25%
>of the games, and 40% of the games are draw, one can expect in 6.4% of such 36
>game matches a result of 22.5 - 13.5 and more extreme.

Very nice Dieter.
I think Jorge thought such a result was impossible with equal opponents :-)



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.