[Election-Methods] Simple two candidate election

[Ian Fellows]

Sorry, I meant to post to the list....

I'm not aware of much in the way of serious opposition to the two-party
majority vote. in fact any reasonable methodology should reduce to the
majority vote when there are two candidates. The only method (to my
knowledge) that doesn't do this is range voting. www.rangevoting.org. Range
voting reduces to something similar to what you are saying. The problem lies
in giving numerical value to something as intangible as "satisfaction." How
do I measure my satisfaction relative to yours? Also, regardless of strongly
I feel (perhaps I only have a weak preference for one candidate),
strategically, I should give the maximum rating to the candidate I like to
maximize my happiness. And could this even be called strategic voting,
because there is no universal yardstick of happiness? Perhaps my weak
preference is comparable to your strong preference.

Ian Fellows
Statistician
University of California, San Diego
http://thefell.googlepages.com

p.s. The arguments on rangevoting.org are pretty fast and loose

-----Original Message-----
From: rjbrown at gmail.com [mailto:rjbrown at gmail.com]On Behalf Of rob brown
Sent: Thursday, December 20, 2007 2:00 PM
To: Ian Fellows
Subject: Re: [Election-Methods] Simple two candidate election


I was going to post a reply to the list but noticed your reply was
just to me.  (bummer, I was kinda hoping you'd go on record with that
opinion)

Yes I am no fan of range voting, and it is a range voting fan (clay
shentrup) who inspired my post.  What is particularly disturbing is
now there is a new book coming out (gaming the vote) that, while
otherwise excellent, endorses range voting.

Interestingly, I did get Clay to admit that in range voting, it is
expected that even honest voters will "scale" their vote so that at
least one candidate gets a zero, and one gets a 100 (or whatever is
max).  In which case it would "reduce to the majority vote when there
are two candidates."  However, the rest of his arguments would not
make sense if it is assumed that a two candidate election works fine
with majority vote.

> p.s. The arguments on rangevoting.org are pretty fast and loose

Ha.  A lot of people seem impressed with Warren's math, and while I
haven't scrutinized it, I sure have seen their logic, which strikes me
as craaaaaaaazzzzeeeeee. :)

-rob