Mike, --- MIKE OSSIPOFF <[EMAIL PROTECTED]> wrote: > You continue: > > Can ATLO make it so that given the following sincere preferences, the > B>C faction has no incentive to bullet vote? > > I reply: > > Yes. If the C voters appy ATLO below C, and rank C 1st and B 2nd, with their > ATLO line between C and B, then the B voters will have no incentive to > bullet vote. > > Of course if the C voters and B voters aren't sure which of {B,C} is the CW, > and neither wants to elect A, and neither wants to be had by the other, then > both the B voters and the C voters should rank B and C, in sincere order of > preference, applying ATLO immediately below their favorite.
That's an interesting approach. I wonder if it's monotonic? > You continued: > > My conclusion is that there is no way to satisfy SDSC and address the > defection problem without electing C at least as often as B on the following > ballots: > > I reply: > > I have no idea what that means. But SD, SSD, BeatpathWinner/CSSD, MAM, and > RP meet SDSC. And, with ATLO, those ballots that you list below wouldn't > happen, given the preferences that you listed above. 49 A 24 B 27 C>B (ATLO line between C and B) Also interesting, this is a case where a method can satisfy SDSC while failing votes-only Minimal Defense, according to which A mustn't be elected on the above ballots. (So, in my claim above, I should have said that I don't believe it is possible for a method to satisfy votes-only Minimal Defense and also address the defection problem, without electing C with at least the probability that B is elected.) It could be unpleasant for the C voters if, as it turns out, the B voters don't prefer C to A. The voters have to be a little clever to use the ATLO. Kevin Venzke __________________________________________________________________ Découvrez le nouveau Yahoo! Mail : 250 Mo d'espace de stockage pour vos mails ! Créez votre Yahoo! Mail sur http://fr.mail.yahoo.com/ ---- Election-methods mailing list - see http://electorama.com/em for list info