The Condorcet criterion just says that a candidate that pairwise beats
*all* the other candidates must win. (Obviously such a candidate is
also the Copeland winner,
which is why Copeland is a Condorcet method.)
But the Condorcet criterion doesn't say anything about counting and
comparing numbers of "pairwise wins."
Right. I guess my line of thinking was....Condorcet seems to put a high emphasis on pairwise wins, in the sense that if you pairwise beat everyone, you win, regardless of how much you beat them by. It would seem logical that that be carried through, so that those who have more pairwise wins beat those who have fewer. But your example showed why that would not always be a good thing.
---- election-methods mailing list - see http://electorama.com/em for list info
