First preference Copeland: Difference between revisions
(Initial page) |
(Add monotonicity failure.) |
||
| (One intermediate revision by the same user not shown) | |||
| Line 1: | Line 1: | ||
First preference Copeland is a variant on the [[Copeland method]]. It was devised by Forest Simmons in 2006. <ref>{{cite web|url=http://lists.electorama.com/pipermail/election-methods-electorama.com/2006-December/117180.html|title=Clone proofing Copeland|website=Election-methods mailing list archives|date=2006-12-30|last=Simmons|first=F.}}</ref> |
First preference Copeland is a variant on the [[Copeland method]]. It was devised by [[Forest Simmons]] in 2006. <ref>{{cite web|url=http://lists.electorama.com/pipermail/election-methods-electorama.com/2006-December/117180.html|title=Clone proofing Copeland|website=Election-methods mailing list archives|date=2006-12-30|last=Simmons|first=F.}}</ref> The name is due to [[Kristofer Munsterhjelm]].<ref>{{cite web|url=http://lists.electorama.com/pipermail/election-methods-electorama.com/2008-July/120010.html|title=[Election-Methods] A Better Version of IRV?|website=Election-methods mailing list archives|date=2008-07-13|last=Munsterhjelm|first=K.}}</ref> |
||
In first preference Copeland, each candidate is assigned a penalty: the sum of first preferences of the candidates that beat the candidate pairwise. The candidate with the least penalty wins. |
In first preference Copeland, each candidate is assigned a penalty: the sum of first preferences of the candidates that beat the candidate pairwise. The candidate with the least penalty wins. |
||
| Line 5: | Line 5: | ||
First preference Copeland passes [[Smith criterion|Smith]] because supposing otherwise, that some candidate Y not in the Smith set wins; then for a candidate X in the Smith set, Y is beaten by every candidate X is beaten by, and some more. Thus Y's penalty must be higher than X's, so Y couldn't have been the winner. |
First preference Copeland passes [[Smith criterion|Smith]] because supposing otherwise, that some candidate Y not in the Smith set wins; then for a candidate X in the Smith set, Y is beaten by every candidate X is beaten by, and some more. Thus Y's penalty must be higher than X's, so Y couldn't have been the winner. |
||
However, FPC fails both [[independence of clone alternatives]]<ref>{{cite web|url=https://munsterhjelm.no/km/yahoo_lists_archive/RangeVoting/web/2006-December/msg00040.html|title=Wait a minute - Simmons' "cloneproof" method is not really cloneproof!|date=2006-12-31|last=Smith|first=W.|website=RangeVoting Yahoo list mirror}}</ref> and the [[monotonicity criterion]]. One example of monotonicity failure is as follows: |
|||
However, FPC fails both the [[monotonicity criterion]] and [[independence of clone alternatives]]. |
|||
{{ballots| |
|||
8: A>B>C |
|||
2: A>C>B |
|||
9: B>C>A |
|||
12: C>A>B}} and C wins. Then raise C on the two A>C>B ballots: |
|||
{{ballots| |
|||
8: A>B>C |
|||
9: B>C>A |
|||
14: C>A>B}} and A wins. |
|||
{{stub}} |
{{stub}} |
||
Latest revision as of 18:59, 16 December 2022
First preference Copeland is a variant on the Copeland method. It was devised by Forest Simmons in 2006. [1] The name is due to Kristofer Munsterhjelm.[2]
In first preference Copeland, each candidate is assigned a penalty: the sum of first preferences of the candidates that beat the candidate pairwise. The candidate with the least penalty wins.
First preference Copeland passes Smith because supposing otherwise, that some candidate Y not in the Smith set wins; then for a candidate X in the Smith set, Y is beaten by every candidate X is beaten by, and some more. Thus Y's penalty must be higher than X's, so Y couldn't have been the winner.
However, FPC fails both independence of clone alternatives[3] and the monotonicity criterion. One example of monotonicity failure is as follows:
8: A>B>C 2: A>C>B 9: B>C>A 12: C>A>B
and C wins. Then raise C on the two A>C>B ballots:
8: A>B>C 9: B>C>A 14: C>A>B
and A wins.
This page is a stub - please add to it.
References
- ↑ Simmons, F. (2006-12-30). "Clone proofing Copeland". Election-methods mailing list archives.
- ↑ Munsterhjelm, K. (2008-07-13). "[Election-Methods] A Better Version of IRV?". Election-methods mailing list archives.
- ↑ Smith, W. (2006-12-31). "Wait a minute - Simmons' "cloneproof" method is not really cloneproof!". RangeVoting Yahoo list mirror.