Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ... and better

Author: Rolf Tueschen

Date: 09:07:35 01/23/04

Go up one level in this thread


On January 23, 2004 at 11:57:38, Igor Gorelikov wrote:

>On January 23, 2004 at 10:12:00, Rolf Tueschen wrote:
>
>>Two progs stand at the top. One is number one in the ranking. But  is it really
>>stronger than prog number two???
>>
>>The answer is easy. If the normal variation, this famous +- value in the SSDF
>>list is say +-40 points and the difference between progs is 35 points THEN we
>>are unable to conclude anything for sure. It could be that 1 is stronger than 2
>>but also the contrary could be true. Only from values >40 on we have
>>"certainty", statistically, that a prog in that specific design is proven
>>stronger than another one.
>>
>>This is all so simply and trivial that it is satifying to be able to clarify.
>>
>
>Not so simple.
>The difference between programs depens also on pool size.
>In small pool difference between the same programs should be bigger.

Yes, correct. That was exactly what I said myself. To be "sure" that a
difference is relevant  it must be - if the "pool" is small - bigger than it is
neccessary for a bigger pool. But the other expert formerly expressed that he
could give a meaning to a difference of 1 Elo point. That however sounded like
hair-splitting to me at least. :)


>
>There are no simple things in relativity that is ELO ;-)


Certainly not. The whole topic isn't also free of contradictions. One single
hair left on your head is seen as too little, almost nothing, while 1 hair in
your soup is regarded as too much. Honestly I don't look through all this
relativity.

:)

Regards,

Rolf


>
>regards,
>Igor Gorelikov



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.