[EM] viciousness to a voting system?

MIKE OSSIPOFF nkklrp at hotmail.com
Fri Apr 27 23:17:12 PDT 2001

Rob LG wrote:

But I feel justified in using
the word "vicious" to describe Mike's attacks on margins.

I reply:

That's the 1st time I've heard of "vicious" being applied to how someone
speaks of a voting system. "Vicious" is a strongly emotional word
to describe my disapproval of a lousy voting system.

Rob continues:

He seems to see
margins methods as betraying all that's good about Condorcet methods.

I reply:

I didn't say it betrayed anything. I said that it _lacks_ all of what's
good about the good Condorcet methods. And it has all of the worst
problems that pairwise-count methods can have.

You know, the IRVies are right about a few things: Approvalists &
IRVies agree that the pairwise-count methods are a strategic mess.
And it's true--except that they don't usually realize that there are
exceptions. Margins isn't an exception. The IRVies are also right when
they say that Condorcetists live in an ivory tower, not even within
hailing distance of the voter and his/her concerns. Again, there are
a few exceptions, but only a few.

Rob continues:

the importance he ascribes to his strategy criteria

I reply:

I ascribe importance to certain standards, because I've noticed
the adverse results of failing those standards.

I often hear people complaining about the results of the failure of
those standards.

My criteria are
my effort to measure for those standards. If Martin adds some new ones,

Rob is all concerned that people will be able to gain under some
conditions by unequally ranking some equally-preferred lower choices,
or insincerely upranking their 2nd choice, when it's a natural circular
tie. I have no idea why Rob considers those as problems. What
political problems will they cause? What voter complaints will remain
unsolved due to adopting wv?

Rob's finding of strategy possibilities in wv isn't an astounding
discovery--I mentioned that Gibbard & Satterthwaite have shown that
every method will sometimes have situations where someone can gain by
strategy. So I'm instead concerned about the gross strategy problem
known as the lesser-of-2-evils problem, or the strategic need for
drastic insincerity. That person who insincerely upranks his 2nd choice
equal to his favorite in wv isn't doing so to protect a CW. It won't
do him any good unless there's a natural circular tie. He's trying
for a slight increase in expectation, which can only pay off if there's
a natural circular tie. I don't regard him as victimized or suffering
a serious strategy dilemma. Rob is bothered by a strategic temptation, 
hoping for payoff if there's a natural circular tie. And this is only
about 0-info elections, something that doesn't happen in public elections.

I'm more concerened about strategic need to abandon a favorite.

As for insincerely ranking unequal some equally-preferred lower choices--
again, that's a temptation, not a need. Someone can get some gain of
expectation by doing so in a 0-info election (which doesn't exist in
public elections). I'm sorry, but I don't feel sorry for someone who
is tempted to rank unequally some candidates among whom he has no
preferences. What is he losing? What's the problem? Who is wronged?
A CW who is so poorly supported that indifferent voters can defeat him
by falsification? Maybe Rob is worried about indifferent, marginally-
supported CWs.

So Rob's supposed strategy problems of wv don't sound much like
problems. Meanwhile, his Path Voting will sometimes elect your last
choice because you didn't vote some lesser-evil compromise alone in
1st place, over your favorite.

Rob says that the guarantees of my criteria are too easily
gotten-around. For instance, SFC & GSFC don't apply if there's falsification 
of preferences. But I claim that's unlikely on a scale
sufficient to change the election result. And with Path Voting, there's
no need to do anything to get-around those guarantees--they don't exist
at all with Path Voting. So a method that doesn't offer those SFC &
GSFC guarantees at all is as good as one that can't offer them if
there's large-scale falsification?

Falsification is how those guarantees are gotten around. But how is
SDSC gotten around in wv? It isn't. For the methods that meet SDSC,
the guarantee is absolute. Contrary to Rob, SDSC can't be gotten-around.
Again, though, its guarantee doesn't exist for Path Voting. With
Path Voting, there are times when a majority can make someone lose
only if a large number of them vote some lesser-evil alone in 1st place,
over their favorite. That sounds a lot like Plurality, IRV, & Borda.
SDSC? Path Voting & RP(m) don't even meet WDSC.

Rob continues:

There are a lot of really smart people on this list.  It's a shame mutual
respect isn't more apparent.

I reply:

What, did I disrespect Margins? I admit that it meets as many
defensive strategy criteria as Plurality, IRV, & Borda.

I carefully make a point of at least matching other people's respect level.

Mike Ossipoff

Get your FREE download of MSN Explorer at http://explorer.msn.com

More information about the Election-Methods mailing list