[EM] Unranked IRV versus Approval - divergent winners exist!

[Tom Ruen]

Bart said:
> Why would you ever support unranked IRV over approval voting?  It might
> be useful in multi-seat elections, but is there anything that would
> warrant iterative counting in single-seat districts?

Perhaps you're right. I'm not sure yet.

I note that recursive "instant runoffs" and Condorcet agree with the
Approval result on unranked ballots, and unranked Condorcet will obviously
never show an uncomfortable cycle.

I am also excited for using Unranked-IRV to handle multiseat elections. It
becomes a cumulative voting method which better frees voters to be able to
support more candidates than their "voting bloc" can clearly help win since
weaker candidates will be eliminated and their vote moved to their stronger
remaining candidates. It should also be possible to use quota surpluses to
reweight ballots, like STV.

My main defense for and attraction to Unranked-IRV is that it satisfies the
one vote/seat rule of our current elections. It is a good compromise in my
opinion since it is just another way to count approval ballots. I like that
approval votes could still be used for measuring support for each party, and
split votes can be used to determine elimination order.

Also, if IRV is passed, I really like the option of tied ranks, especially
if it is agreed that an approval count of top ranks is a more accurate
representation of support for each candidate than a single vote. Ties are
just as secure as ranks when elimination order is clear, and a tie sends a
message that I would like my major party to listen to my minor party.

It can become a sort of embarrassment when an instant runoff must be
performed. A candidate that gets a majority in one round has a more
legitimate majority support than one that required the elimination of a
weaker party to get a majority. As a voter, I would actually gain some
satisfaction knowing my split vote helped force a runoff (instant is not
instantaneous!) when my favorite of the top-two candidates was less than my
perfect choice. Approval makes it a little too easy for my compromise to
win.

About the instability of elimination among 3 strong candidates, in ranked or
unranked IRV, I'm still not overly afraid. Small spoilers are the more
common enemy. When a candidate has enough core support to become a big
spoiler, as measured in the polls, that candidate will get much attention
from the media and the other two parties and they will start talking about
the issues that helped this third party to rise. Whether or not a third
party wins doesn't matter as much as having more discussion of the neglected
issues that helped the third party become popular. Maybe we can actually
have some more serious debates between candidates rather than who is going
to give the biggest tax cuts or entitlements to voters.

Tom Ruen

----- Original Message -----
From: "Bart Ingles" <bartman at netgate.net>
To: <election-methods-list at eskimo.com>
Sent: Thursday, March 29, 2001 2:32 AM
Subject: Re: [EM] Unranked IRV versus Approval - divergent winners exist!


>
> Tom Ruen wrote:
> >
> > Finally, I'll ask a bigger question than my original one (Unranked-IRV
or
> > Plurality):
> >
> > If we are limited to unranked ballots, for single seat, single ballot
> > elections, which method would you prefer: Plurality, Unranked-IRV or
> > Approval voting? (or some other method?)
>
>
> Why would you ever support unranked IRV over approval voting?  It might
> be useful in multi-seat elections, but is there anything that would
> warrant iterative counting in single-seat districts?
>
> Bart