Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Maybe ...

Author: Dann Corbit

Date: 09:30:54 10/06/03

Go up one level in this thread


On October 06, 2003 at 11:05:41, GuyHaworth wrote:

>
>As many said previously, the absolute numbers mean nothing:  it is the
>difference between the ELO numbers for two players that indicates relative
>strength.
>
>But I assume your question is posed about the rating system as it is now, not as
>it might be with a hypothetical 200 added (in which the answer is "It has been
>done already").
>
>If a 'perfect player' starts off with a rating of 1500' and wins all its games,
>playing at each stage someone with its own rating (or the next one down), I
>wonder how long it would take before it was playing GK at the top, and then how
>long it would take to get to round numbers like 2900, 3000 etc.
>
>In fact, I do not know if there is an upper ELO limit, an asymptote for its ELO
>score.

There is no asymptote.  If (for instance) a player won 90.90909% of his games
from a pool of 2700 players, then the Elo of that player would be 2700 + 400 =
3100

If (against the same pool) the 2700 players managed only 1e-5 % of the points,
then the Elo would be 2700 + 2800 = 5500



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.