Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ELO Formula/Table

Author: Jouni Uski

Date: 22:07:24 10/13/99

Go up one level in this thread


On October 13, 1999 at 13:33:39, Ratko V Tomic wrote:

>Thanks for the formula.
>
>> ELO difference = -400 * log (1/res-1),  where  res = result%/100%
>
>Is this base 10 or 2 or natural log?
>
>(If res=W/(W+L) then dE=400*log(W/L), and comparing your formula to the other
>formula for res=75% giving dE=200 points, in this thread, that would suggest
>base 10 logarithm is assumed. Is that right?)
>
>Also, what's the range of error? (Is error on W and L assumed to be sqrt(W) and
>sqrt(L) for large enough W and L values?)

Of course log is base 10 as ln is base e. There are no error range - calculation
is correct :-)



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.