[EM] Poll of single-winner election methods

[MIKE OSSIPOFF]

Don asked about the poll condicted on EM some years ago. The methods
that had been proposed here at that time were PC, Smith//PC, IRV,
Copeland, and some proposals that Demorep made. I don't know whether or
not Approval was a proposal in that poll, or whether it was only
a poll on rank-counts.

Smith//PC was the winner, and PC was the 2nd finisher. That was true
not matter which of the proposed rank-counts were used to determine
the winner & 2nd finisher.

Since then we've found out about other methods that are known to
meet more criteria than PC & Smith//PC are known to meet, but
PC is still tops when the greatest brevity of definition is needed.
And Smith//PC can easily be gotten from PC by adding one short clause.
That can even be added in a subsequent enactment campaign, after PC
is enacted.

So, even though, based on known criterion compliances, Tideman(wv),
SSD, & BeatpathWinner look better, PC & Smith//PC still might be
very good proposals for public elections.

(As Markus recently pointed out, it hasn't been demonstrated that
PC & Smith//PC pass GSFC or SDSC, or that they fail those 2 criteria).

Mike Ossipoff

_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com