On Dec 14, 2007, at 6:57 AM, Dave Ketchum wrote: > BECAUSE we are debating different elections. Lets think on what we > make of Diego's scenario: > I saw it as intelligent voting, but did not go back to create > utility numbers consistent with those decisions. > You came up with proposed utility numbers and ask what different > voting might be consistent with those numbers. > > So I will try at utility for what I saw in Diego: > voters' utility for {A or C} is {100, 10, -200} > voters' utility for {B} is {100, 50, 50} > > Thus: > If A or C are solid winners, utility score for A or C winning is > positive. > If A and C are near ties better, as with Diego, for B to win.
I agree that there are possible utility distributions that favor a vote for B. I don't agree that {100, 10, -200} is a reasonable interpretation of A >> B > C, which, if it means anything, means that the utility gap between A & B is larger than that between B & C. Aside from that, I think we're more or less on the same page here. ---- Election-Methods mailing list - see http://electorama.com/em for list info