[EM] equal rankings IRV

[Tom Ruen]

James,

I'm not at all willing to judge theoretical differences of IRV versus ER-IRV, but I I have strong opinions on application of approval votes. I take the "one person, one vote" ideal seriously in politics. I can only seriously support the option of split-votes for tied rankings as a natural extension to IRV.

I can accept ER-IRV as an experimental method to consider, but I disapprove of the name "Equal Rankings IRV" since it suggests nothing about how equal rankings are counted, and may equally apply to "Approval votes" as well as "split votes" as far as the name implies. I'd defend the name ER-IRV as best reserved to imply "split votes" (since IRV is based on "single vote" STV which definitely could only count equal rankings as split votes). I'd most certainly prefer a completely different name for whole votes counted, or if not, then something like "Approval IRV" which to me more accurately described the hybrid counting scheme.

I see two political arguments against approval counting:

1. "Single vote" methods allow clear measures for candidate strengths even if there's a strategic betrayal of favorites that prepoll as weak and unlikely to win. When there's no (zero-sum) "cost" to offering extra votes, voters will offer them more as strategic protests or encouragements to losers, and will (if wise) withhold them to the degree they can affect the winner. A candidate that gets 5% of the vote in a "single vote" count have 5% of the population supporting them. 5% in an approval count means nothing since it comes from an unknown combination of "favorite choice", "pity votes", "insincere tactical votes", and general mischief votes. I don't know about the average candidate, but if I was running I'd not want "pity votes" or "protest votes" in my count. I want to know how many voters are willing to stand behind me alone over all others. I would gladly accept a "half-vote" by voters willing to offer me equal to another candidate since it represents a sacrifice.

2. "Single vote" methods (plurality/runoffs better) promote ideological opposites as the top two candidates. These two combined give a clear coverage of a single integrated "majority" and single "minority" position. Those top-two rank prestege will influence how the parties will move in future elections and how the general population lies when limited to a single choice. In comparison Approval votes, will, in 3-way races at least, tend towards promoting two similar candidates into the top-two positions, and say little about the real "center-of-mass" of the voters in general.

I see the logic of "single vote" as a defendable right, and "approval vote" counting as, at-best, an indulgence allowed under the whim of a kind majority. That is, if a majority of voters are disinterested in supporting more than one candidate, then it doesn't matter that a minority would like this option. In contrast, a "split vote" option, if counting isn't impeded is more likely to be judged as a harmless option, even if a majority may still reject it as unnecessary.

I can't offer any serious interest in "approval count" methods for politics. I judge them "mostly harmless" for affecting winners, and don't expect there ever to be a political majority will to test it.

Tom Ruen

Date: Sun, 13 Jun 2004 06:32:28 -0400
To: election-methods-electorama.com at electorama.com
From: "James Green-Armytage" <jarmyta at antioch-college.edu>
Subject: [EM] equal rankings IRV

Dear Voting Gurus,
...
I'm wondering how everyone else feels about equal rankings IRV (ER-IRV).
Really, I'm welcoming everyone to give their opinion on the matter, even
if they're just agreeing with what someone else has said. I guess you
could say I'm trying to take a poll or something.
...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20040615/2c0be44f/attachment-0003.htm>