Hi Chris, --- Chris Benham <[EMAIL PROTECTED]> a écrit : > Kevin, > Your "Improved Condorcet//Approval" (ICA) method I take attempts > to minimally modify Condorcet//Approval(ranking) so that it meets > Sincere Favourite (your version of FBC).
Yes, that's right. > http://nodesiege.tripod.com/elections/#methica > > http://nodesiege.tripod.com/elections/#critsf > > > 48: A>B > 02: B > 49: B=A > 01: C > > A>B 48-2, A>C 97-1. > > In this virtual 2-candidate election, ICA elects B. > > To fix this, I suggest: > > "3-slot ballots, default rating Bottom, Top and Middle interpreted as > approval. > > If the Top ratings winner T has a TR score higher than T's > maximum pairwise opposition score then elect T. > > Otherwise elect the regular ICA winner." > > This seems to be a pure improvement. What do you think? I would say that by taking TRs away from your favorite, so that he is no longer the TRW, you could cause another of your favorites to win according to the first measure, as opposed to the first favorite failing to win according to that measure, causing a different and inferior candidate to win according to ICA. This seems entirely possible to me, because I can't see any obvious relationship between which candidates can possibly win by the first method as opposed to the second. I'd say I'm not so worried about the 2-candidate Condorcet failure possibility of ICA, due to the puzzlingly indifferent voter behavior it requires. Kevin Venzke __________________________________________________ Do You Yahoo!? En finir avec le spam? Yahoo! Mail vous offre la meilleure protection possible contre les messages non sollicités http://mail.yahoo.fr Yahoo! Mail ---- Election-Methods mailing list - see http://electorama.com/em for list info
