[EM] RE: Bucklin
stepjak at yahoo.fr
Fri Sep 23 19:09:15 PDT 2005
--- "Simmons, Forest " <simmonfo at up.edu> a écrit :
> I like the modified ER Bucklin Whole version that Kevin and Mike have been considering.
I want to say: Although it's interesting that ERBW satisfies FBC and Majority,
I think it differentiates among adjacent ranks so little that strategically,
few would be worse off using Approval.
> I have two suggestions that might make it more viable as a public proposal:
> 1. Keep the number of possible distinct ranks down to seven or eight, for ballot simplicity.
I think a problem with this is that voters then need to be able to leave slots
empty, and we clearly have a ratings method, which makes it harder to say which
properties are satisfied.
> 2. Allow a special mark "+" to be made next to the favorite candidate in the top slot.
> Call this method ER Bucklin Whole Plus or ERBWP.
> The plus is there so that the voters who equal rank top for strategic purposes can still show
> extra moral support for their favorite, and for resolving ties.
> In case of a tie, a random ballot is drawn. The candidate with plus next to his name on that
> ballot gets to pick which of the tied candidates is elected.
> If two candidates have plus next to their names, then a coin is flipped to see which of the two
> candidates gets to pick from among the tied candidates.
It would be more clone-independent if we narrowed the field down to these two
candidates, and then drew further ballots to pick between them.
> This gives voters the expressive power that the voters want, along with a tiny bit of extra
> instrumental power for their favorite, without destroying the FBC property of Kevin's ER
> Bucklin Whole.
> Another version that would give even more meaning to the plus: if k is the first round in which
> there is a candidate with more than fifty percent, then consider all candidates with more than
> 50% in round k as tied, and use the "plus" tie breaking method described above.
> Would this version preserve the FBC?
I can't see any problem with using Random Ballot to break Bucklin ties. For
that matter, I can't see any problem with breaking ties based on the total number
of "plus" votes.
Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger
Téléchargez cette version sur http://fr.messenger.yahoo.com
More information about the Election-Methods