Margins vs Winning Votes

[MIKE OSSIPOFF]

Rob LeGrand said:

Now Mike Ossipoff wants to discourage strategic truncation

I reply:

That wasn't what I said. I said that I'd like to keep truncation
(strategic or otherwise) from
causing the defeat of a sincere CW in violation of majority rule. And I'd 
like voters to
not have to use drastically insincere strategy in order to keep
that from happening.

But I've been quite specific about what I want from a voting system.
My criteria (and those of Steve Eppley & Russ Paielli) tell that.

What do I want from a rank-balloting method? SFC, GSFC, WDSC, & SDSC.

I've defined those in recent messages, but they're all defined at
http://www.electionmethods.org

I want to preserve majority rule, and I want to avoid the lesser-of-2-evils 
problem.

Rob LeGrand continues:

, but he doesn't find
it important not to encourage strategic equal rankings high in the ballot.

I reply:

Rob LG, let me be the one to break it to you that no nonprobabilistic
method is free of all incentive to vote strategically in some way.
You're criticizing wv because there's an incentive to vote insincerely,
and so you're asking wv to deliver something that no nonprobabilistic
method can deliver.

wv methods give some incentive for equal ranking of candidates not
regarded as equal. Margins methods give defensive strategic need
to do that, and also to reverse preference orderings, including the
burying of one's favorite.

Rob LeGrand continues:

I
don't see truncations as anything but a specific case of equal rankings. To
me, a vote of A>B when there are five candidates is simply a vote of 
A>B>C=D=E.
I don't think this case is any more special than a case like A=B>C=D>E. Mike
does, his favorite criteria reflect that, and that's why he prefers
winning-votes, which of course is perfectly fine for him.

I reply:

Wrong again. The definitions of my criteria tell you what I
consider important. But it's true that truncation, with margins,
can take victory from a sincere CW, in violation of expressed majority
wishes, and that that won't happen with wv methods.

But I don't have an opinion about which of your above cases is
more "special".



Rob LG continues:

To me the bottom line is this: Any evil strategy that a margins method 
allows

I reply:

Perhaps Rob LG has been listening to Mr. Bush's speeches and borrowing
his terminology. Instead of calling a strategy "evildoer strategy",
maybe Rob LG means offensive strategy. Can we substitute that term?

Please note, however, that my criteria don't make any mention of
offensive strategy or evil strategy. It's becoming evident that
Rob LG feels free to quote me on what I want from a voting system,
quite independently of what my criteria say.

Rob LG continues:

Any evil strategy that a margins method allows
is possible in the corresponding winning-votes method, whether by flipping
coins or by coordinated bloc voting.

I reply:

Perhaps Rob LG would like to show us an example in which a wv method
allows the strategy of truncation to elect a candidate over whom
a majority rank the sincere CW.

That's something that can happen with margins, but not with wv.

Rob LG continues:

So, as Blake has explained, winning-votes
prevents nothing.

I reply:

Well, for one thing, it prevents what I described above.

In general, it prevents the election of a candidate over whom a
majority of the voters prefer the sincere CW, if those voters vote 
sincerely, and if no one order-reverses.

SSD, CSSD, BeatpathWinner, & RP(wv) prevent the election B if
A is a member of the sincere Smith set, and B isn't, and a majority
of the voters prefer A to B and vote sincerely, and if no one
order-reverses.

WDSC & SDSC tell of other things that wv methods prevent, and which
margins methods don't prevent.

Said most generally, wv methods prevent lesser-of-2-evils problems
and majority rule violations that margins methods don't prevent.

Rob LG says that margins does much better by social utility than
wv methods. The simulation results that I've seen show practically
no difference. When margins gives away the sincere CW's win, it
loses social utility compared to wv, because the CW is typically
the social utility maximizer. Does the simulation that Rob LG is
quoting have truncation? I don't mean random truncation, as some
simulations have done. Truncation that drops sincere CWs will
, in margins methods, elect candidates whom the truncators prefer
to the CW. With any method, failure to rank the CW will hurt voters
for whom the CW is the best they can get. The resulting loss of
social utility isn't as important as that caused when the truncators
gain by taking the win from a CW.

Rob's statement about social utility is based on simulations that
don't seriously address the question.

Rob LG continues:

Margins vs.
winning-votes is obviously a religious argument

I reply:

If Rob LG considers it a religious issue, then maybe his intolerance
of those who don't share his beliefs can be explained as
religious intolerance.

I've always emphasized that there's no such things as wrong standards.
Standards are subjective & individual. Rob LG's standards aren't
wrong. But Rob LG is wrong when he claims that his standards are
better than those of others.

Rob LG continued:

Personally, I'd like to concentrate on newer,
more interesting debates.

I reply:

Yes, it would be better if Rob LG concentrated on issues on which
he might make more sense than he does on this issue.

I've always just stated my standards, and the criteria by which
I measure for those standards, and shown which methods meet those
criteria.

That isn't what Rob LG & Blake are doing here, on this issue. I
must admit that I don't know what they're talking about.

And I've always invited others to produce other criteria to measure
for the majority rule & LO2E standards.

Someone once outlined a suggestion for different
versions of my majority defensive strategy criteria, but he didn't
complete his criteria. If he ever does, that would be great.

Mike Ossipoff







_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com