[EM] Correction of false statements by Ossipoff & Schudy about range voting.

[Abd ul-Rahman Lomax]

At 07:28 PM 7/22/2007, Paul Kislanko wrote:
>
>Steve Eppley wrote in part:
> >>Warren Smith's
>example, in which a voter has total knowledge of all other votes before
>casting her own vote, is implausible in the elections we're interested
>in reforming.<<
>
>I might be mistaken, but when I was introduced to this group it was more
>about studying methods than "reforming elections."

Mr. Kislanko is correct to point this out, though election reform is 
also relevant here. That is, in studying election methods it is 
relevant to consider how they would function in public environments. 
This, however, is not an advocacy list, so Mr. Eppley's use of "we" 
was casual and doesn't refer to the members of the list in general, 
even though *many* of us are interested in reform and some of us are 
actively working for it.

The general topic of election methods, however, includes all kinds of 
methods for choosing between alternatives. Usually we assume a 
single-stage, deterministic method, but it has been quite proper 
here, I believe, to, for example, refer to deliberative process or 
negotiation as "election methods;" indeed, the conformance of an 
election method to what could be expected from full deliberative 
process is a possible criterion for judging methods.

I've noted, for example, that standard deliberative process (Motion, 
Second, Debate, Amendment (which process may continue repeatedly), 
Close of Debate, and Vote Yes or No) is indeed an election method, 
one that we ordinarily don't consider on a large scale because it 
becomes, we have generally thought, impossibly cumbersome. There are 
several ways around this though, the general principle being a 
boiling down of the group which actually participates in the 
deliberative process, which, I have come to realize, is severable 
from the actual voting at the end. This comment, all by itself, would 
be enough for some readers, albeit unusual ones, to derive a major 
part of what I write about over and over.

>Folks who get all on board with election reform when their candidate loses
>tend to not discuss things in terms of "methods" but of "practices" and
>we're spending an inordinate amount of time on matters that mix voter
>behaviour with the mathematics underlying an election method. I've proposed
>before that there should be a way to axiomitize the distinctions (I'm not
>smart enough to propose one, but I joined the list to learn about how all
>the different methods work, not to try to impose one that I like...)
>
>For what it's worth from all I've learned about methods on this list if I
>were going to "reform" anything about the mess that is US national elections
>I'd pick approval for party primaries and some Condorcet-compliant method
>for the general elections.

Mr. Kislanko should take a look at the Range 2 method used in MSNBC 
polls. It would make an *excellent* tool for political parties to 
discover the best candidates, even if the final party decision is 
made by some Condorcet-compliant -- or even Plurality -- process. The 
MSNBC Range polls cut through and made plain some very interesting 
facts that have largely escaped notice in the media. There is only 
one Republican in those polls with a net positive rating, and all the 
others have a net negative rating, a substantial one. That Republican 
is almost not a Republican, as seen by other Republicans: Ron Paul. 
Indeed, Ron Paul's positive rating was quite comparable with the 
Democratic frontrunners, who were Edwards and Obama. Clinton had a 
slight negative rating.

Now, there is certainly polling bias; these polls were open to 
anyone; cookies prevented multiple voting, but any informed computer 
user could have gotten around that and voted more than once. But what 
I found interesting was that the polls seemed to reflect the truth 
about the electorate, as well as I can understand it from other 
sources. There were three Democrats who had higher ratings than all 
those considered frontrunners in the Republican stable. Only Ron Paul 
was in range of winning, if the election were Range and the election 
had been held with these poll results.

(These Range 2 polls were expressed in a form that I've often 
recommended: the votes were -, 0, +. 0 was the default vote, if you 
didn't fill in a circle for a candidate, that was a zero rating. It 
affected the percentages for each vote, which is what was reported, 
but not the net rating (positive minus negative).)

When there is sufficient pre-election process, I would seek something 
even stronger than Condorcet for the final race, I would require a 
majority vote for the winner. In an Approval election. With 
sufficient process, I don't consider Range necessary. However, there 
often is not sufficient process, and Range functions better, I 
believe, in that environment. It is best in the zero-knowledge case, 
clearly, with the optimal vote shading into Approval style *for some 
candidates in one's preference sequence* as knowledge of the election 
probabilities increases.