[EM] Richard's neat idea (was Tyranny of majority)

[Richard Moore]

Forest Simmons wrote:

> Now having buttered you up a little bit, I want to point out that at least
> one thing I said below is true.  Your great method is indeed strategically
> equivalent to ordinary Approval.

Forest,

Thanks for your kind comments and suggestions.

I wasn't trying to give the impression that this idea doesn't collapse
strategically into Approval. In fact I wasn't thinking of it as an election
method at all (like CR), but as a theoretical standard for evaluating
election results (like SU). It really isn't the same as the mythical
"absolute SU" either, first because each voter's evaluation of the
background candidates is different, and second because of the way it is
summed.

If you did a simple summation of the voter's ratings according to this
method, then I suppose the result could be considered a "calibrated SU".
But instead of the simple summation, I suggested a count of the pairwise
majority wins over background candidates. That honors the concept of
majority rule, but in a way that doesn't give just one particular majority
the only say.

So hopefully I kill two birds with one stone. At least it takes care of
two of my own concerns: the need for a better standard than the
uncalibrated relative SU and the nonexistent absolute SU, and some
method of evaluating majority potential that doesn't neglect all the
other majorities that didn't win the election.

By the way, I was thinking of the calling the rating of a candidate
by this method the candidate's "majority win potential", or maybe
just "majority potential". Do you have any other suggestions for
nomenclature?

Richard