Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CCT8 Update: New entrant

Author: Robert Hyatt

Date: 09:23:45 02/16/06

Go up one level in this thread


On February 16, 2006 at 03:48:53, Vincent Lejeune wrote:

>On February 15, 2006 at 23:26:28, Robert Hyatt wrote:
>
>>On February 15, 2006 at 17:35:06, Peter Kappler wrote:
>>
>>>On February 15, 2006 at 17:13:54, Robert Hyatt wrote:
>>>
>>>>On February 15, 2006 at 16:24:20, Uri Blass wrote:
>>>>
>>>>>On February 15, 2006 at 15:18:56, Robert Hyatt wrote:
>>>>>
>>>>>>On February 15, 2006 at 13:19:35, George Tsavdaris wrote:
>>>>>>
>>>>>>>On February 15, 2006 at 12:59:32, Peter Skinner wrote:
>>>>>>>
>>>>>>>>On February 15, 2006 at 12:55:56, Peter Skinner wrote:
>>>>>>>>
>>>>>>>>>Hello all,
>>>>>>>>>
>>>>>>>>>As promised by Vincent last week, he has entered Diep into CCT8.
>>>>>>>
>>>>>>>Nice but what hardware he will use? Any huge one or normal(fast).....?
>>>>>>>
>>>>>>> Also is this the most large number of participants for a CCT or there have been
>>>>>>>and any larger....?
>>>>>>> I remember very recently Hiarcs and an experimental Junior participating. Did
>>>>>>>that happen in the last CCT or in another different than CCT tournament.....?
>>>>>>>
>>>>>>> Also 9 rounds with so many participants seems a bit short for me. Do you
>>>>>>>consider increasing the number of rounds.....?
>>>>>>>
>>>>>>
>>>>>>9 rounds is enough to find a clear 1st place for 512 opponents.  :)
>>>>>
>>>>>This is correct if one program wins all the games but it will probably not
>>>>>happen.
>>>>>
>>>>>>
>>>>>>By the time this event ends, the top group will have all played each other.
>>>>>
>>>>>I am not sure of it.
>>>>>In theory it is possible to have 10 winners with 6 out of 9 and in this case it
>>>>>is clear that they need to beat weaker programs to get more than 4.5 out of 9 so
>>>>>not every pair of winners played.
>>>>>
>>>>>Uri
>>>>
>>>>In practice this doesn't happen.  The real problem is that by round 6, the
>>>>winner is pretty certain, although three games against weaker opponents are
>>>>left, and there the luck factor often lurks behind a pawn and jumps out to
>>>>present a surprise and knock one of the top players off.
>>>>
>>>
>>>I have to believe that you're half-kidding here.  Surely you don't believe that
>>> adding 3 more rounds increases the variance of the final result?
>>>
>>>-Peter
>>>
>>
>>Yes, it absolutely does.  Simple example:  4 strong programs, 12 weak ones.
>>after round 1, you have 8 with 1, 8 with 0, the 8 with 1 include the top 4.
>>After two rounds, the top four have two, the bottom four have 0, and the rest
>>have 1 (assuming no draws).  after 3 more rounds the top four have played.  What
>>now for those other 4 rounds?  Other programs?  With the luck factor increasing
>>variance?
>
>You assume that the strong programs always win against weak but it's not the
>case : If program1 is 200 elo stronger than program2, program2 score 3 points
>against program1 in 10 games. It's a lot of noise to find the

According to my stats, if A is 200 points better than B, A will win 3 of every 4
games.  or 6 of every 8, or about 7+ of every 10.

There will always be a few surprises along the way, which is why a _couple_ of
extra rounds is useful (log2(#players)+2 rounds, say).  but +4 is pushing it and
the last rounds are _always_ anti-climatic...



>
>
>>
>>
>>>
>>>
>>>
>>>
>>>>log2(#players) is a good number of rounds for starters.  one more gives a couple
>>>>of more top finishers a better final result.  But too many and many of the final
>>>>rounds are simply meaningless, as we've seen multiple times.  For example, at
>>>>the last WCCC they tried to fudge the round-robin pairing to put a few of the
>>>>"interesting games" on the final few rounds.  Didn't work out at all and the
>>>>event was over early...
>>>>
>>>>There can always be too many rounds in a Swiss.
>>>>
>>>>9 is on the edge.  Both in final results and in total time required to play 9
>>>>games...



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.