[EM] Nash Equilibrium Outcome (NEO)

[Michael Ossipoff]

At EM, I once proposed a method that I called Nash Equilibrium System (NES).

I now call it Nash Equilibrium Outcome (NEO).

NEO:

Rank as many or as few candidates as you want. Equal ranking allowed.

Find the candidate(s) who, in Approval, based on the rankings, could win at
Nash Equilibrium.

If there's only one, s/he wins.

If there's more than one, repeat within that set.

Repeat till there's only 1 candidate who could win at Nash equilibrium.

If, at any stage, there's no Nash equilibrium, the winner is the candidate
at that stage with highest top-count.

(end of dfn)

NEO evidently has no chicken dilemma, and shares MAM's maximally
unproblematic offensive/defensive strategy situation.

So NEO appears to be another MAM-like CD method (along with MMPO).

It's inherently natural & fair to choose the outcome that no one can
improve on for hirself.

Admittedly NEO (so far as I'm aware) is new, and hasn't had the critical
discussion that MMPO has bad.

(We've heard people's best arguments against MMPO.)

Two (very roughly) related methods are DSV & Game Theory (GT).

DSV is much more complicated. Articles about GT often praise it, but don't
define it. Maybe the definition is too long or complicated to include in a
reasonable-length article.

---------------------------

Regarding the method that starts with an approval-ordered list, and then
starts sorting by pairwise defeats; and the method that sequentially adds
the most approved candidate who covers those already added:

The 2nd one evidently meets Smith, but both have chicken dilemma, and an
offensive/defensive strategy situation not as good as that of MAM, MMPO,  &
NEO.

Michael Ossipoff
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20160915/6b754e9a/attachment.htm>