[EM] Range-Voting article--basically right, but a few disagreements

[MIKE OSSIPOFF]

I liked Professor Smith's Range-Voting articles. I was one of those who felt 
that Range-Voting (CR) was a little bit less desirable than Approval because 
sincere voters can be taken advantage of by strategizers. But, as Smith 
points out, it won't really be like that. Everyone will have benefit of 
strategic advice (look at all the strategic advice we get now, from honest 
individuals (even if their lesser-evil strategy is based on false 
information)). Some will choose to vote sincerely. Sometimes I probably 
would.

I'd mentioned some time ago that, if Nader-preferers who feel the need to 
vote for Kerry vote sincerely, and rate Kerry below Nader, I like that.

IRV promoters mistakenly repeat endleslly that Approval voting will revert 
to Plurality. No, if people know what they're doing, they'll always vote for 
their favorite. And if Nader preferrers now feel a need to vote for Kerry, 
they'll do so in Approval too, till they notice that Nader outpolls the 
Republican. So, in the 1st Approval election, they'll vote for Kerry and 
Nader, in Approval. But if it were CR, and if they vote sincerely, they'll 
help their sincere favorite against their lesser-evil, and that's a good 
thing, not only for Nader, but also for honesty and democracy.

But what if there are 2 genuine progressives, and you like one better than 
the other, and Approval strategy says to vote for both, to defeat the 
completely unacceptable candidates? Do you downrate one of those 
progressives in CR? I hope not. What's a good thing with Kerry/Nader 
wouldn't be such a good thing then. But Democrat vs Nader is the situation 
now, and in that now-exsting situation, sincerely-voted CR would be a good 
thing.

Smith expresses that in terms of Schwartrznegger's election, but I prefer to 
put it in terms of Kerry & Nader, as I did above. Anyway, his argument makes 
sense.

Smith's information about 67% sincerity in CR is encouraging, at least when 
it means progressives will downrate the Democerat.

It seems to me that he reported that people voted Approval like Plurality. 
There's nothing wrong with that strategy, but I believe that Smith is 
mistaken if he's saying that that's how Approval will be used in a public 
election by people who know how Approval works. As I said, progressives who 
now think they need the Democrat will act on that belief in Approval too, 
and will vote for Nader and the Democrat.

Why would his survey show otherwise? Maybe he spoke to sincere voters who 
vote by principle. Maybe he spoke to voters who didn't believe that they 
needed the Democrat. Maybe he spoke to people who didn't understand how 
Approval works:

I've encountered such people, and it's one reason why CR is probably a 
better public proposal.
One person objected to Approval, saying that if you vote for the 
progressive, that weakens your vote for Democrat over Republican. Of course 
it in no way affects your Democrat>Republican vote.

Another person told me that, though Nader is more honest than Kerry, and 
offers better policies than Kerry, that person would vote only for kerry in 
Approval. Due to a mistaken impression of how Approval works? Maybe. Or 
maybe due to the completely misguided progressive anger at Nader for 
allegedly spoiling the corrupt Democrat. This may sound surprising, but many 
progressives confuse  stratgegy with candidate merit, and will telll you 
that Nader is less desirable--and when asked why, can only give you the 
alleged strategic reasons for not voting Nader. But somehow they begin to 
perceive that as a matter of candidate merit rather than strategy. "But 
isn't Nader egotistical" (because takes votes from the corrupt 
media-anointed candidates)? "He's whacko" (Further questioning reveals that 
he's whacko because he runs though he isn't viable, not because of his 
policies).

If Smith's survey said that progressives will vote only for Kerry in 
Approval, and if that's what will happen, it isn't just a fault of Approval; 
it's a hopelessness about voting. That 2nd person that I spoke of, when 
voting in an Internet straw poll (at the excellent website
http://www.masquilier.org/agora ) ranked Kerry first, and ranked Nader far 
down the ranking, below Dean and a few other Democrats, even though that 
voter said that Nader was more honest and offered better policies. Could it 
be that we're wating our time by offering better voting systems?

Smith said that to vote probabilistically in Approval, to simulate CR, 
required specially-construced dice, and much complexity. Not really. Can't 
decide whetiher to vote for someone? Flip a coin. How complicated is that?

If the method is Approval and you want to cast a sincere CR vote, you don't 
need custom-made dice: Cut 10 identical squares or rectangles from a piece 
of paper or cardboard, roughly a square inch or two each. Number them from 0 
to 9. Put them in a paper bag. Say you want to give to a candidate .79 
votes. Take a number out of the bag, and write it on a piece of paper. 
Return the number to the bag. Take a number out again and write it directly 
after the first number, as the 2nd digit of a 2-digit number. If that number 
is less than 79, vote for the candidate.

Smith mentioned Bayesian regret. I've heard of it, but not heard it defined. 
If I read his definition correctly, it's the SU (social utility) of the SU 
maximizer minus the SU of the winner. Is that right? One way to minimize it 
is to prevent high SU candidates from running. Another way is to maximize SU 
of the winner. Smith says that CR does best at that. But that would only be 
true under sincere voting, and under the special conditions where the CW 
doesn't maximize SU.

As I mentioned here before, the CW maximizes SU under a wide variety of 
conditions.

1. When there's 1 issue dimension.

2. When there are more issue dimensions and:

  a. Distance (representing disutility) is measured by city-block distance

  b. Distance is measured by Euclidean distance, and the 
population-density-distribution has the
     following property:

     Along any line through some point, the population density along that 
line, starting from that
     point, is the same in both directions from the point.

    That property is had, for one thing, when density is normallly 
distributed in each issue-dimension
    as is the case in pretty much all simulations. Normal distribution is 
widely, and reasonably,
    assumed for various social distributions.


So then, we can expect that the CW will maximize SU. That doesn't depend on 
sincere voting. I suggest, then, that Condorcet maximizes SU more reliably 
than CR does, and therefore does better at minimizing Bayesian regret--if 
Bayesian regret means what I believe it does.

There may be more that I want to say about the article. If so, I'll mention 
it in the article's context:


Smith wrote:

6.BAYESIAN REGRET (FOR STATISTICS NERDS):
Approval voting does quite badly (measured by Bayesian Regret)
compared to many other more expressive voting systems such
as Borda and Black, when there are 4 or more candidates and we have honest 
voters.

I reply:

He means nonstrategizing voters. Strategizing isn't dishonest. Borda isn't 
going to give us nonstrategic voting. It can't be counted on in any method 
other than maybe Condorcet.

But if voters are divided between  strategy and sincere rating, maybe CR 
will do better than Approval by SU, if the Approval strategy is based on 
incorrect information. I don't expect CR to do better than Condorcet by SU 
under actual conditions.

Smith writes:

Approval is better than these systems [Borda, Black, CR]- for strategic 
voters - but is worse for honest ones.

I reply:

We have many insincere voters now. Probably some, maybe many, would vote 
sincerely in CR.

Summary:

Smith seems right about CR giving better results than Approval for our 
current political situation, with Republicans, Democrats, and progressives 
who (think they) need to help a lesser-evil. But when there are 2 or more 
genuine progressives, and some completely unacceptable candidates, sincere 
CR voting by progressives could be detremental, and Approval voting for them 
all seems better.

Bullet voting in Approval by people who now vote lesser-of-2-evils won't 
happen if they know how Approval works.

Of course in the 2nd Approval election, when they notice that Nader 
outpolled the Republican, votinlg only for Nader would be good strategy.

Though CR with some sincere voting might very well give better SU than 
Approval when Approval stragtegy is based on incorrect information, I'm 
skeptical about Smith's claim that CR gives better SU than Condorcet.

Though, if we disregard public acceptance,  I like Condorcet better than CR, 
and like Approval better than sincere CR when it's several genuine 
progressives vs some completely unacceptable candidates, I don't share 
Smith's stong opposition among those methods. I like Condorcet, Approval, 
and CR.

Many advocates of Condorcet, CR, or Approval strongly oppose the other two 
of those methods. I like them all.

Mike Ossipoff

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/