[EM] Some July 4 comments

[Bart Ingles]

Falling behind, but at least it's not Labor Day yet...

Chris Benham wrote:
> 
> Previously I wrote:
> Assuming for a moment that a political landscape dominated by 2 major
> parties  is a normal and happy situation, isn't it VERY important that
> the party which is prefered by an absolute majority is the winner ?
> To which Adam Tarr replied:
> " Absolutely. But prefered by a majority to what?"
> To which I answer:  to the OTHER major party, of course. ( If  you are
> offered a menu with  2 items, A and B, on it  and you say "I prefer A" ,
> obviously you mean "I prefer A to B".)

A better question might be, "A majority _of_ what?"  Of all voters, or
only those preferring A to B?  Expressed or sincere preferences?

Taking the most probable intended answer-- a majority of all voters'
sincere preferences but not of 1st-choice preferences-- if there are
known to be 2 major parties, then the correct voter strategy to support
A or B is perfectly obvious.  The only possible reason for not voting
for either A or B is that you either have no preference between them, or
you wish to cast a "protest vote".  The equivalent action for a
3rd-party voter with IRV is to bullet vote for the minor candidate.  So
it's not clear how much difference IRV would make in a partisan two-way
race (James Carville's hysteria notwithstanding).

Where a method like IRV would be most useful-- and where it is most apt
to fail-- is when the best plurality strategy is not obvious, such as a
competitive three-way race, or a local election with several candidates
and no reliable polling data.


> I don't deny, and wasn't arguing, that Condorcet is better than IRV. I
> was arguing against the proposition that  "IRV is nearly  as bad as
> Plurality". I know that some Approval proponents have boasted of their
> roles in helping to scuttle the introduction of  IRV (to replace
> Plurality) is some parts of the US.

I haven't really heard of any approval proponents, myself included, who
have done anything other than attempt to provide information that had
been omitted from a pro-IRV pitch.  If IRV proponents believed in their
product enough to fully inform the people they are trying to influence,
there would be less chance of anyone queering the deal.

As for IRV being nearly as bad as Plurality, one could make the case
that plurality has at least a few things going for it.  In a 3-way race,
the plurality winner is guaranteed to have over 33% first-choice
support.  IRV can only guarantee 25%, and the figure declines faster
with additional candidates.  And IRV effectively neutralizes small 3rd
parties in exchange for the pure symbolism of higher first-choice vote
totals.

One could argue that 3rd parties and their supporters actually have
_more_ power with Plurality than with IRV, provided they use intelligent
strategy.  If they actively run against what they perceive to be the
worst of the major-party candidates, while stepping aside in support of
the best, they can attempt to gain concessions from the major party.  Of
course most 3rd parties don't do this, and instead influence the outcome
in a negative direction, but the opportunity is there.