[EM] Re; Proportional Lottery Application

[Forest Simmons]

"Hahn, Paul" <manynote at wustl.edu>
Wrote ...
"I am very likely missing something here, and please explain to me what it
is if I am?but is this a long way of saying ?Condorcet//Plurality has more
going for it than we first realized??"

Paul,

no, but-good question.

You are thinking of Plurality as a fallback method when there is no
Condorcet Winner.

But we're talking about Plurality as a measure of which pairwise defeats
are strong enough to keep, and which ones need to be discarded.

The greater the first place count of the defeater, the greater the priority
of the defeat.

Suppose A defeats B, which in turn defeats C, which defeats A, and that the
respective Plurality counts are 20, 35, and 45. Then the two defeats with
the highest defeater scores are kept, and other defeat (A defeats B with
only 20 plurality votes) is annulled in deference to the higher priority
defeats.

We end up with B>C, and C>A, which makes a beatpath of B>C>A. So B wins,
not the Plurality winner C.

I hope that clarifies the difference. The rationale for the defeat priority
measure is a longer story, not so obvious. (If it were obvious, I suppose
it would have been discovered long ago.)

It has taken many decades of combined experimentation as hinted at in the
"convergence of ideas" paragraphs, to find and understand the advantages of
this method compared to other equally simple, intuitive methods.

You can expect to see lots of discussion (on this EM list) that will
gradually illuminate these advantages.

The method is simple. The theory is not so easy. Most people get their
understanding through many worked examples ... not by deductive logic.

So it takes patience.

I think it's worth it!

-Forest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20221022/daf46d9e/attachment.htm>