I think margin is the more natural criteria. I think relative margin is the best criteria to represent the people's will. I think winning votes is the best criteria to provide sincere rankings.
I have yet no way to compare their relative quality, so actually I can live with any of these. Just tell me in advance how my unranked choices will be interpreted... However, I prefer Ranked Pairs as a Condorcet method. Steph. PS: relative margin is mathematically coherent. Even if A(7)>B(1) locks before C(400) > D(365), A cannot win. Try it!! ---- For more information about this list (subscribe, unsubscribe, FAQ, etc), please see http://www.eskimo.com/~robla/em
