Author: Robert Hyatt
Date: 20:11:46 08/30/00
Go up one level in this thread
On August 30, 2000 at 19:31:07, Pete Galati wrote: >On August 30, 2000 at 11:13:07, Robert Hyatt wrote: > >>On August 30, 2000 at 05:29:35, Uri Blass wrote: >> >>>On August 30, 2000 at 05:20:51, Vincent Diepeveen wrote: >>> >>>>On August 29, 2000 at 19:03:52, Amir Ban wrote: >>>> >>>>>On August 29, 2000 at 13:26:36, Vincent Diepeveen wrote: >>>>> >>>>> >>>>>>The whole pairing was changed because of AMIRs fear for SOP. Now the last >>>>>>round SOS gets the world title for free, the rest simply doesn't >>>>>>count. >>>>>> >>>>> >>>>>Why don't you take your lunatic ramblings elsewhere ? >>>>> >>>>>Amir >>>> >>>>If you don't show them in world champs anymore sure. >>>> >>>>Note i'm not lunatic you are. You even protested against a free point >>>>against Pacque expert because it *might* influence your SOP. >>> >>> >>> >>>If you are right about it then the protest did not help so >>>I do not think that the pairing was changed because of Amir. >>> >>>Junior did not enjoy from playing better players. >>>The opposite Junior suffered from it because it scored less points(it could only >>>draw against tiger and by playing against xinix in the last round Junior could >>>probably get a better place. >>> >>>Uri >> >> >>I don't understand the pairing issue at all. This is a mathematical process >>that is finite and has rules to cover _every_ possible case. It is just like >>teaching someone to do long division. The steps are cast in stone, there is >>no room for varying at all. >> >>Why can't the pairings at WMCCC events be done the same way? There is no room >>for "A really ought to play B". It is either "A must play B here" or "A must >>play somebody else and B must play somebody else here." >> >>I learned to follow a simple recipe of instructions 45 years ago. It isn't >>hard. Or is it? Based on some pairings over the past few years, you would >>assume there is great flexibility in who plays who in each round. I have >>run many tournaments (human) and there is zero flexibility. _if_ the pairing >>rules are followed to the letter. >> >>The only flexibility is in which 'system' you will use. IE if you have three >>players with a score of 4/5, what do you do? (a) bring up the highest rated >>player in the next-lowest pool to fill this group out to 4? (b) pair the top >>two players in this group and pair the third with the highest seed in the next >>lower group? (c) drop the lowest player in this 4.0 group down into the next >>group? etc.. >> >>There are options, but they are to be chosen _before_ the first round is paired, >>and then the same rules are to be used for the entire tournament. Somehow this >>doesn't get done at WMCCC events. At least not since Mike Valvo stopped >>directing them. > >If by some odd miracle the next WMCCC _does_ end up getting held in North >America (I'm not holding my breath), then maybe the program's pairings will be >taken care of better. Maybe. I never tried to understand how it's done myself. > >Pete It does tend to get harder if you have a small number of players and a large number of rounds. This is rare (except for WMCCC events) and can easily break programs not tested under such extreme conditions. But it is doable by hand... It is doable by program although I have not personally looked at what is available to see how they do with 8-9 rounds and 12-14 players, which is really an oddball tournament.
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.