[EM] Oops! Squeeze-effect.

[Michael Ossipoff]

Raph:

When you mentioned IRV's squeeze-effect, I said that unfavorite CWs
will get eliminated. But of course the squeeze-effect refers to
something more than that:

Say it's a 1-dimensional political spectrum. A CW could be favorite to
many. S/he could have a plurality. And s/he could be eliminated, if
the s/he's the most favorite, and the favoriteness tapers gradually,
away from that CW.

Elimination would start at the extremes. Transfers would be sent
inward, until the candidates adjacent to the CW would have collected
all of those inward-transferred votes, enough to eliminate the CW.

Of course gradual tapering of favoriteness, away from the CW is hardly
an unusual or contrived situation.

So it's safe to say that IRV isn't at all good at electing CWs or fair
compromises.

All the more reason why IRV would be no good under the existing
conditions, with our existing electorate.

...at least not without JITW.

In that squeeze-effect scenario, of course the voters preferring
candidates to one side, plus those preferring the CW, must add up to a
majority. So there are two majorities, and the one that prevails will
be the one on the side that the CW's voters prefer and transfer their
votes to, when the CW is eliminated.

That majority is a mutual majority. The other one is not. IRV is fine
if you're in a mutual majority. But, as I was saying before, though
I'm sure a majority want something more progressive, there might not
be a progressive _mutual_ majority, and so IRV might not be a good
idea for progressives. IRV is a special purpose method that should
only be supported by befeficiaries of a mutural majority.


Mike Ossipoff