> >Sorry, I meant PR (Proportionnal represention) > Oh, okay. Yeah, I was just talking about single-winner. PR is a totally different problem. > > >>I really don't think so. Please note that I'm only talking about >>situations where there is a sincere Condorcet winner. If there is no >>sincere CW, then I think that no equilibrium Votes will exist, as I >>defined them in my last post. But when there is a sincere CW, then there >>will be equilibrium Votes, and all of them will result in the election of >>the CW. > >So we need another method for those cases ? or at least, we must agree >that there is no unique solution :-/ > Perhaps it's fair to say that there is no unique solution. For any outcome, a relative majority will prefer a different outcome. That's not to say that some outcomes cannot be deemed better than others, but it may mean that a strategic equilibrium cannot be found. The method I have proposed for resolving majority rule cycles is weighted pairwise. http://fc.antioch.edu/[EMAIL PROTECTED]/voting_methods/weighted_pairwise.htm But I have not claimed that it can produce a full strategic equilibrium in the case of a sincere cycle. I don't know if any method can, without it being unfair in some way.
my best, James ---- Election-methods mailing list - see http://electorama.com/em for list info