Winning-votes intuitive?

[Rob LeGrand]

Adam is certainly the most convincing winning-votes advocate I've seen on the
list (and I've never encountered one off-list).  I like that he makes his (not
unreasonable) assumptions clear.  It's hard to imagine a better case for
winning-votes, but until there's a better one, I'm still a "margins" fan.

I come from a mathematical background; specifically, I'm a student of game
theory.  So I tend to be pessimistic: I'd rather assume that whatever voters
could do to help themselves, they will do.  The argument that any insincere
strategy that works under margins also works by flipping a coin under
winning-votes is very persuasive to me that winning-votes's advantages are
illusory.  But I can understand that some wouldn't see flipping a coin as
producing the same expected effect as tied votes under margins.

In addition, I can't think of any methods besides winning-votes methods for
which an optimal zero-information ballot is insincere.  Even under plurality,
the best vote is sincere in the absence of poll information.  I see this as a
serious defect of winning-votes, especially in a local or nonpartisan election
or if you're not one to trust polls, but again, I can understand that some
would disagree.

One thing that I never understood about Mike's reasoning (I'm sure Mike will
correct me if I'm wrong on any point) is that he likes Approval because it
never forces a voter to vote a less-liked candidate over a more-liked
candidate, but he likes winning-votes methods because, in a sense, they *do*
sometimes force a voter to engage in strategic reversal, where the less radical
strategy of voting ties might work under margins.  He likes that truncation,
voting ties among the lower ranks, isn't a good strategy under winning-votes,
but doesn't seem to mind as much that voting effective ties by coin-flipping or
outright reversal do the job for a wily voter just as well as truncation.  He
also doesn't seem worried about top-of-the-ballot "truncation".  His aims seem
somewhat contradictory to me.

It should be remembered that any possible method will have serious strategy
problems when the electorate is well-informed and there is no sincere Condorcet
winner.  Appropriately-formulated polls will show a cycle of frontrunners all
the way up to the day of the election.

When an election is definitely zero-information, finely-grained Cardinal
Ratings is undoubtedly the champ.  In fact, when no strategy is possible, I'd
rather the sincere CR winner win than the Condorcet winner.  The advantage of
the Condorcet winner is simply stability under strategy.  Given that voters
know each other's preferences, CR and Approval both home in on a Condorcet
winner, so really they give the best of both worlds.  To me, Approval's
combination of simplicity and ultimate quality is not only unparalleled, but
incredible.

Approval-completed Condercet does indeed render the margins/winning-votes
debate meaningless.  Its only disadvantage in my mind is its slightly more
complicated ballot.

--
Rob LeGrand
honky98 at aggies.org
http://www.onr.com/user/honky98/rbvote/calc.html


__________________________________________________
Do You Yahoo!?
Yahoo! Greetings - send holiday greetings for Easter, Passover
http://greetings.yahoo.com/