Re:[EM] Sincerity for criteria, Blake

[Alex Small]

Blake Cretney wrote:

>So, Approval has a very similar problem to IRV. I can certainly see why
>Green supporters wouldn't see why IRV is much worse from this perspective.

Approval and IRV would probably never elect a Green, since they're
radical.  Same for Libertarians, Buchananites, etc.  They should push for
PR.

A centrist but distinct third party, however, would probably do quite well
with IRV or Approval.  Such a party might be somewhat left on social
issues, somewhat right on economic issues, but never too far from the
center.

With IRV it would have to be one of the top 2 (after all of the minor
parties are weeded out).  Otherwise one of the 2 polar opposites would win.

With Approval they could be the first choice of the smallest faction (e.g.
40% Dem, 40% GOP, 20% 3rd option) and still win.  Most Dems and Republicans
wouldn't give an additional vote to the other side.  The third option,
however, with positions appealing to both sides, might command enough cross-
over votes from the Dems and GOP to win.

If they were the 1st choice of the largest or second-largest faction they'd
be almost guaranteed a victory, assuming that they get some cross-over
votes, and that cross-over votes cast by their faction split roughly even
between the Dem and the GOP.

So, Approval would favor whoever gets the largest coalition of core
supporters, cross-over voters, and undecided swing voters.  Add in
advantages like monotonicity, FBC, IIAC, plain old simplicity, and having
only n tallies to keep rather than n! and you have an excellent method!  (I
know that IRV can be done with O(n*2^n) tallies but many people might want
all n! tallies to verify that there's no foul play.)

Alex Small

--
"Frodo gave his finger for you."