[EM] MMPO's problem and proposability. IRV in Burlington.

[MIKE OSSIPOFF]

MMPO's problem and proposability:

Kevin, good point about MMPO's problem in the bad-example being more about lack of support,
as opposed to mere lack of favoriteness.

Yes, it's understandable if people don't like that. I can't say that I like it myself. But of course, for
perspective, it's necessary to consider it in comparison to other methods' problems. I concede
that MMPO is choosing wrongly in that example. But I still claim that, as problems go, that problem
isn't a strategy problem, or a problem that will prevent an electorate from getting what they want. And
isn't that the main important thing?

In that example, C isn't a Hitler. _None_ of the voters think that C is worse than both of the two most favorite
candidates.

Still, I fully concede that criticism about what happens in that example could easily be a problem for MMPO
in an enactment campaign. As I've said, opponents of progressive reform tend to have a lot more money
to spend than proponents have. Opposition to voting reform will be very heavily funded. That example will
be described on doorknob-hangers, and in broadcast-media, and people will hear about it a lot more than they'll
hear proponents' reply.

So, though I rank MMPO2 at the top of my merit-ranking, Approval is my proposal.

You know, all the advocates of better voting systems agree that Approval is better than Plurality. We disagree
only about how we'd improve on Approval. Isn't it obvious to all that Approval is the obvious thing that we all
can agree on? The obvious first proposal? It's the first, simplest, natural improvement on Plurality. Plurality
done right.

IRV in Burlingon:

The problem with what IRV did in Burlington is the effect that that sort of result can have on timid voters. 

It will make them favorite-bury. The reason I don't like favorite-burial is because many people are inclined to resignedly
compromise on a disliked lesser-evil. Never compromise on or support an evil. That should be obvious.

A good and valid strategy in IRV is to rank the acceptable candidates in order of winnability. The only problem is that
many think that a disliked lesser-evil is "acceptable". If people knew what "acceptable" means, then IRV would be ok,
and the IRV strategy of ranking the acceptable candidates in order of winnability would be ok. 

There's also nothing wrong with not using tht strategy, but, instead, just ranking sincerely. 

That apparently, is what the Republicans did in Burlington. Nothing wrong with that. If the Republicans
and Democrates were a mutual majority, then one of those would have won. If the Democrat voters ranked the progressive
2nd, then the Democrats and Progressives were the mutual majority. The mutual majority won. 

I and others have been a bit unfair to IRV in Burlington. IRV didn't do anything wrong in Burlington, except in the
sense that it did something that will make many people give elections away to a lesser-evil. Given better voting strategy,
IRV would be ok, as I've said. But, voters' strategy being what it is, IRV isn't a good proposal, and I still oppose it as
a voting system reform for the United States.

If the Republicans really considered the Democrat acceptable, not just a lesser-evil, then it would be fine for them
to rank hir 1st, strategically. If they consider it fair to not win without a mutual majority, then there's nothing wrong
with their sincere voting, and they'd have no objection to the resulting Progressive win.

(I'm going to reply to Abd's posting, but I have to go somewhere now, and so I'll probably have to reply on Monday.)

Mike Ossipoff


Mike Ossipoff

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20120303/7185ae31/attachment-0003.htm>