[Election-Methods] Schudy--Clarifying a few things about the example
Michael Ossipoff
mikeo2106 at msn.com
Tue Jul 24 01:47:53 PDT 2007
Warren Schudy--
In your example, there's a candidate, Hitler, who is so despicable that only
two people, out of the entire electorate, vote for him. Your example would
work the same if there were 100 million voters, with only two Hitler voters.
You have one of the largest factions upranking Hitler to 2nd place. When
they do that, of course he's no longer a two-vote candidate, and those
voters mustn't be surprised if they elect him.
And, if the other large faction truncates, refuses to rank the reversers'
candidate, then the reversers will elect Hitler, if the truncators beat
their candidate pairwise. What's their chance of benefitting from the
reversal? Only if the two big factions are in a pairwise tie. If the
reversers' candidate beats the truncators' candidate, then the reversal
isn't necessary anyway. So the chance of actually benefitting from the
reversal, when there's truncation defense, is negligible. The reversers are
_incomparably_ more likely to elect Hitler than benefit from the truncation.
In your paper, and your posting, you speak of that election of Hitler as a
disadvantage of Condorcet.
No, deterrence is an advantage of Condorcet.
Yes, the fact that the Blues could even conceivablyl benefit by something
that could also elect a two-voter Hitler is a little embarrassment. But, as
I said, the best rank methods have so many valuable properties that a little
embarrassment is acceptable. Especially because the event will never happen,
because it's so well deterred. That last fact is worth emphasizing.
I like to point out that the only way you can benefit from offensive
order-reversal in Condorcet, MDDA or MAMPO is if your intended victims rank
your candidate. You can steal the election from them only if they're trying
to help you. Doesn't that make you proud of yourself?
Offensive order-reversal is the nearest thing to a strategy problem in
Condorcet, MDDA and MAMPO, but it is not a problem.
Mike Ossipoff
More information about the Election-Methods
mailing list