[EM] 2-party domination = favorite betrayal?

[Warren Smith]

I do not agree these two things are equivalent, although they are related.

If a method exhibits "favorite betrayal" but only rarely then third 
parties might be able to flourish.   For example Coombs' IRV-like 
method exhibits favorite betrayal.  Would it lead to 2-party domination?

Really 2-party domination is an experimental question and may not be answerable
with mathematics alone, although one can often use math to become fairly confident
of the answer.

If rational voters acting in realistic scenarios with imperfect information
often find it wisest to do favorite betrayal (especially in 3-candidate scenarios
with 2 major-party candidates) then I think we get 2-party domination.
The key muddy words here are "realistic scenarios with imperfect information"
and "often" and another problem is voters are humans rather than rational beings.
wds


PS. Although approval voting does not involve favorite betrayal, there is some
reason to suspect it will lead to 2-party domination.  See
   http://math.temple.edu/~wds/crv/NurseryEffect.html
and note that range voting seems a lot less likely than AV to lead to 2-party
domination.  This is quite a subtle effect, and I certainly was not smart
enough to predict it with mathematics alone - it required an experiment to
make me see the light.