Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Statistics, chess, and computers.

Author: Erik Bergren

Date: 09:31:06 08/14/03

Go up one level in this thread


On August 14, 2003 at 07:41:32, Tim Foden wrote:

>On August 13, 2003 at 15:40:00, Erik Bergren wrote:
>
>>Here is the Theoretical and experimental data produced by
>>the programs above:
>>
>>
>>                     (After 10,000,000 matches):
>>                     (With Probability Ratio of .6237 for White to win.)
>>Number of White      Theoretical  Experimental  Experimental
>>Wins per             outcome      outcome       outcome
>>each match(="n"):    for "n":     for "100-n":  for "n"
>>==========================================================================
>>50                   821699       (823153)      823153
>>51                   804525       805157        803365
>>52                   755117       753009        756603
>>53                   679385       680519        678344
>>54                   585885       587044        585463
>>55                   484234       482899        484001
>>56                   383517       383030        383595
>>57                   291021       291419        290359
>>58                   211537       211395        211549
>>59                   147254       146946        147571
>>60                    98140        98336         98541
>>61                    62601        63001         63208
>>62                    38206        38150         38069
>>63                    22301        22647         22274
>>64                    12444        12388         12268
>>65                     6635         6534          6614
>>66                     3378         3295          3392
>>67                     1642         1662          1607
>>68                      761          757           735
>>69                      336          336           334
>>70                      142          153           129
>>71                       57           51            52
>>72                       22           16            21
>>73                        8           11             8
>>74                        3            1             1
>>75                        1            0             1
>>76                        0            0             0
>>(76 to 100 are all zero)
>
>Sorry, but this distribution looks almost exactly what you would expect for a
>50% outcome, not a 62.37% one.
>
>For 62.37% I would expect something like this:
>
>  n       tot n   tot 100-n
>---  ----------  ----------
> 50       34116       34116
> 51       54610       20120
> 52       85326       11447
> 53      128900        6196
> 54      185092        3220
> 55      258427        1686
> 56      343477         808
> 57      439408         394
> 58      539885         161
> 59      637399          68
> 60      721291          33
> 61      783080          16
> 62      816871           5
> 63      818126           1
> 64      782431           1
> 65      719842           0
> 66      631379           1
> 67      531070           0
> 68      426372           0
> 69      328604           0
> 70      240963           0
> 71      169194           0
> 72      113084           0
> 73       71670           0
> 74       43304           0
> 75       25126           0
> 76       13580           0
> 77        7066           0
> 78        3411           0
> 79        1591           0
> 80         685           0
> 81         280           0
> 82         126           0
> 83          46           0
> 84           9           0
> 85           2           0
> 86           0           0
> 87           0           0
> 88           0           0
> 89           0           0
> 90           0           0
> 91           0           0
> 92           0           0
> 93           0           0
> 94           0           0
> 95           0           0
> 96           0           0
> 97           0           0
> 98           0           0
> 99           0           0
>100           0           0
>total 10000000
>
>Cheers, Tim.
No, you >>>>>>>missed a step<<<<<< in the experiment:
each opponent gets to play with the
White pieces an equal amount ( such
as by switching side after each game,
or switching sides mid match.)
 Your data has its maximum around
.6237*10,000,000. That implies you
gave one opponent an advantage over
the other (I guess one opponent had
the white pieces for all the match)
( my experiment was perfectly
>>>>symmetric<<<<<, both opponents
had equal chances and that implies
that the most probable match result
>>>>must be 50 to 50<<<<<, thus you data is
not applicable since its maximum is at 63).
  If switching sides is not accounted for
then the equations are (relatively) easy:
r^x*(1-r)^(100-x)*100!/((x!*(100-x)!))
I assume that is the equation that
produced your data.
    These subtleties are easy to
overlook, thanks for taking a shot
at correcting me though.



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.