Forum for General Aspects of Tournament Systems and Tiebreakers

Impossibility of "Fair" Tie-breaker? [#2261]

Back to forum     Back to page

New reply

 
reply
71.234.188.145: Impossibility of "Fair" Tie-breaker? (2010-04-15 18:52) [#7639]

Bob McGuigan: The work of Kenneth Arrow in social choice theory says that under certain reasonable conditions there is no fair way to make a societal ranking of alternatives from individual rankings. See [ext] this Wikipedia article for more information. Breaking ties isn't the same sort of thing but Condorcet's paradox, the problem that leads to Arrow's result, also is seen in tournament results. For example, in a round robin tournament with eight players suppose players A, B, and C are all tied for first place with the same won/lost scores of 6 - 1. Suppose further that A defeated B, B defeated C, and C defeated A and A, B, and C won all their games except the losses already specified. This creates the Condorcet condition and it seems that none of the formal tiebreaking methods based only on performance in the tournament will break this tie. In this sort of situation it seems fairest to me to have co-champions and to split the prizes.

X
132.246.27.86: Re: Impossibility of "Fair" Tie-breaker? (2010-04-15 22:03) [#7642]

pwaldron: This paradox occurred in practice during the 2009 USYGC playoffs ([ext] here). The top three players beat each other in a circle, and as it turned out were tied according to SOS and SDOS even when taken to infinite order. In the end we had to run a playoff.

 
Back to forum     Back to page

New reply


Forum for General Aspects of Tournament Systems and Tiebreakers
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library