[EM] Strategy in IRV/elimination

[MIKE OSSIPOFF]

Now I understand Blake's statement better, after reading Bart's
reply to it, so let me comment again:

> >Blake Cretney wrote:
> >  [...]  Strategy in IRV is actually to place a centrist
> > lesser-evil small party in front of a more favoured large party.

I'd misunderstood you. You were referring to the situation that
we've always shown in our 3-candidate IRV examples.

If we knew enough to say which is the bigger candidate, then
there wouldn't be strategy dilemma. The whole problem is that we
don't know that, and so people can misjudge.

Obviously the only time when we'd need to insincerely rank
Middle over Favorite is if Favorite is bigger than Middle, and
can eliminate Middle. I'm sorry, but that isn't a new revelation
here. We knew that. And wouldn't it be nice if we could always
know when that's the case. The problem is that we might believe
that that situation exists, that Favorite will eliminate Middle and
then lose to Worst, when actually Favorite had a win, and we gave
it away by insincerely ranking Middle in 1st place.

The situation that you describe is the one that I've been
talking about all along, and the one demonstrated in all of our
3-candidate IRV badexamples. When Favorite is bigger than Middle,
and would eliminate Middle then it's strategically advantageous
to insincerely rank Middle 1st.

It sounds like we agree on what IRV's strategy problem is.

Mike Ossipoff

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 
http://profiles.msn.com.