[EM] Scenario where IRV and Asset outperform Condorcet, Range, Bucklin, Approval.

[Jameson Quinn]

2010/5/12 robert bristow-johnson <rbj at audioimagination.com>

>
> On May 12, 2010, at 3:22 PM, Jameson Quinn wrote:
>
>  First off, I'm not generally an IRV advocate. But in the interests of
>> fairness, I think that the following (to me very plausible) 4-candidate
>> scenario should be considered.  I call it Radical Center, in honor of Thomas
>> Freidman's onanism. I'll present preferences, but you can make this work
>> with Approval (and thus also strategic Range or Bucklin) using reasonable
>> assumptions. In Approval and related systems, there is a safe defensive
>> strategy; in Condorcet systems, if both relevant groups use strategy,
>> neither wins and the whole society is stuck with significantly inferior
>> candidate.
>>
>> Candidates are Left, Right, Moderate, and the eponymous Radical Center.
>>
>> Honest preferences (6 voter groups split into 3 larger groups for easy
>> understanding.)
>> ---(35 L>...>R leftists)---
>> 18: L>M>RC>R
>> 17: L>RC>M>R
>>
>> ---(30 ...>L=R centrists)---
>> 16: M>RC>L=R
>> 14: RC>M>L=R
>>
>> ---(35 R>...>L rightists)---
>> 18: R>M>RC>L
>> 17: R>RC>M>L
>>
>> Condorcet winner is M. But if all the RC>M voters truncate before M, then
>> M does not beat R and L, so there's two cycles M>RC>(R|L)>M. Most Condorcet
>> tiebreakers, including Schulze and Minimax, would name RC as the winner. (Of
>> course, if the M voters retaliate in kind, then R or L would win Condorcet,
>> or M would win Approval, Range, or Bucklin.)
>>
>> There are two ways I can see to avoid this dilemma.
>>
>
> i'm curious to exactly what is the dilemma.  it's pretty close between RC
> and M.  M is only barely preferred over RC.
>
> is it that if M and RC voters voted against their political interests, then
> someone not to their liking gets elected?  (i can't even tell if it would be
> L or R, they look like they would tie.)  then the lesson is, if you don't
> vote for candidates (even if not your favorite) that represent your
> political interests, your political interests lose.
>
> is it that if somehow the RC voters, that cannot see into the future,
> somehow take a risky guess and announce to their caucus to stick it to M,
> that they get rewarded for it?  or that the M voters hear about it,
> retaliate, and both groups get screwed?  how is the latter worse than what
> happens in plurality where there are more centrists than anyone, but they're
> in two rival parties, and they can't get their act together?
>

No, it's no worse. And neither Bush nor Obama is/was worse than Hitler. But
that doesn't mean nobody should oppose them.

Strategy is bad. I recently gave a list of a number of different ways. It's
bad for legitimacy - that is, those who feel the system has cheated them
from their rightful win could try to overthrow the government or replace the
voting system with plurality or something crazy like that. It's bad for
cohesion: RC and M would soon be at each others' throats, unable to
cooperate on their large common ground, and even within each camp, there
would be acrimony between strategic and nonstrategic voters. It's bad for
utility - the 52/48 split in the example is small, but you can easily adjust
the numbers to widen it. It's bad for expressiveness: with strategy, it's
impossible to tell if the RC voters honestly prefer M over R or L, or if
they don't, and knowing the truth of that matter has some intrinsic value.
And it's bad because it self-reinforces: strategy begets strategy, and all
of the above bad effects are multiplied.

This particular strategy is not the end of the world. But if there were two
voting systems which were otherwise equally good, and one resisted this
strategy but the other didn't, the resistant system would clearly be better.


>
> i haven't figgered it out yet, but are you saying that IRV would *not*
> reward the RC strategy and elect M anyway?
>

Yes, that's correct.

JQ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20100512/e2d4fd77/attachment-0004.htm>