Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Crafty, too arrogant to play dozens of GM's?

Author: Robert Hyatt

Date: 08:47:14 10/27/99

Go up one level in this thread


On October 27, 1999 at 00:42:09, James B. Shearer wrote:

>On October 26, 1999 at 20:16:37, Dann Corbit wrote:
>
>>On October 26, 1999 at 19:38:08, Jeff Anderson wrote:
>
>                       <deletions>
>
>>>When you are very picky about who you play, good things can happen to your
>>>rating.
>>
>>In the long run, it does not matter who you choose at all for computation of an
>>ELO rating.
>>
>>The values will normalize unless some kind of cheating is going on (and they
>>only cheat against you).
>>
>>Playing ELO 2700 players and playing ELO 2000 players will not change your
>>rating any differently.  That's why GM's would not fear lower ranked opponents
>>in a real tournament.
>>
>>The only connection between ELO and who you play is the generic pool connection.
>
>      This is not accurate.  The ELO system is based on a model of the real
>world.  Like all models this model is just an approximation of the real world.
>This means that some potential opponents will appear stronger when playing you
>than their ratings would indicate while others will appear weaker to you than
>their ratings would indicate.  By selectively playing those opponents which do
>not play up to their rating against you, you can definitely raise your rating
>above what it would be playing all comers.
>                              James B. Shearer


Sure, but there is one important difference. When playing Brian the other night,
someone noticed that the "assess" command (for crafty) said : win: +0, draw:
-16, loss:-32.

If you play someone that far under you, you had better win every game, or else
your rating _only_ can go down.

This entire subject is a bit of a red herring, IMHO, anyway.  The "automatic"
programs have this problem.  The "manual" programs don't.  Because when they see
a human that gives them trouble (cptnbluebear is a good example) they won't play
him after they lose a game.  Yet Crafty (or ferret, or other automatic programs)
will go head-to-head with him for 8 straight hours.  If the program can't
compete, its rating is going _down_.  So it is kind of funny to see the manual
programs complaining, when they are _far_ more selective than crafty is.  :)



This page took 0.01 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.