Author: Dann Corbit
Date: 12:26:58 11/29/99
Go up one level in this thread
On November 24, 1999 at 22:10:52, Dave Gomboc wrote: >On November 24, 1999 at 20:34:49, Dann Corbit wrote: >>On November 24, 1999 at 20:28:13, Dave Gomboc wrote: >>>On November 24, 1999 at 19:04:52, Dann Corbit wrote: >>[snip] >>>>It might be interesting to do a round robin, then a swiss of the top ten. >>> >>>Is that your English going south? :) >>> >>>Usually, it's the swiss that's done with the big field, and the round-robin with >>>the small one (e.g. the contestants that succeeded in the swiss.) >>Maybe that's how they do it, but it assumes that you know a seeding to start >>with. With chess games, I think it is a bad assumption since the programs get >>changed all the time. How can you seed a swiss tournament when you don't have >>any idea what the relative strengths are? >> >>So play a round-robin tournament (two games at each board) to calculate the >>seeding, then play a swiss tournament. >> >>IOW, I think (if they play swiss then round-robin) they are doing it exactly >>backwards from the way that they should be. > >You don't need to have a seeding to play a swiss at all. Just draw lots. In that format, the swiss is unfair. In order for it to work properly, you need to know the relative strength of the opponents. It answers the question "Who appears to be strongest" very quickly (and is analogous to any algorithm to form a heap) but it does not do so fairly. That's why tennis tournaments don't start out with Agassi verses Sampras in the first round. One of them will get clobbered just like Schmedlap Verses Schmoo. Are the two winners really of about the same strength and the two losers also? To use the Swiss in this way (without proper seeding) is incompetent.
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.