Quoting Dave Ketchum <[EMAIL PROTECTED]>: > Maybe time for new traditions. A and B had core > support. They also each > had core enmity, and ranked ballots allowed both > support and enmity to be > expressed.
Well said! One of the attributes with IRV by supporters is that it shows depth of support. What it does *not* show is depth of enmity. Given three equally-strong parties, a centrist candidate will typically be the winner under Condorcet (or mean Kemeny order, which is my favorite Condorcet-complete method). Under IRV, either the liberal or conservative wing can win by crowding the center. (One could argue that this shifts candidates toward the center anyway, but why pick a faux centrist when you can choose a real one?) Under our present system, it's like using a steering wheel with only two settings, hard left and hard right. I'm convinced that the enmity toward opposition parties would be reduced if our representatives were "split the difference" moderates rather than "love them or hate them" conservatives or liberals. Michael Rouse ---- Election-methods mailing list - see http://electorama.com/em for list info
