[EM] Campaign issue: wv vs m

[MIKE OSSIPOFF]

Rob wrote:


But
when it comes to margins vs. winning-votes (also known as votes-against
or defeat-support), I don't think strategy turns out to be an issue,
unfortunately, because winning-votes doesn't actually get around the
problem.  Under winning-votes, a voter can order sincere ties in
preferences randomly and get the same expected value as if they were
counted using margins.

I reply:

Sure, one can gain insincerely in that way, and also sometimes by
insincerely voting one's 2nd choice equal to one's 1st choice, with
wv. Of course it's an individual subjective matter which things are
most important, but, to me, those strategic temptations aren't as
important as a sincere CW losing because a majority didn't
use strategy, in some cases reversal strategy. What bothers me the
most is when insincere strategy is needed to protect the win of a
sincere CW, or to enforce majority rule. Hence the majority defensive
strategy criteria.

And remember that exactly equal-rated candidates isn't common when
people know things about all the candidates. There may be some
candidates between whom I don't bother expressing preferences. But
then it's my equal ranking of them that's insincere, by laziness.

For instance, in a spatial simulation, how likely is it that two
candidates will both be exactly as far from a voter's position in
the n-dimensional issue-space?

Rob continues:

Since sincere ties are most often among the
lower preferences, voting the ties sincerely is most often punished.
But when the sincere ties are among higher preferences, winning-votes
provides a voter with strictly more strategic options than in margins.
If we can assume that the voters are intelligent (or even just
well-informed), I don't think winning-votes will do any better inducing
sincere voting than margins.

I reply:

Maybe, but there are some very different motivations & needs for
strategic voting, as I was discussing above. I personally prefer wv
because of that consideration.

Rob continued:

Mike has argued that winning-votes meets
important strategy criteria that apply when no voter "falsifies a
preference."  To me, this simply means that winning-votes often forces a
voter to reverse preferences as the best strategy.  So those criteria
don't seem to make any important assurances after all

I reply:

Well, sure, judging by wv's strategic temptations to vote 2nd choice
equal to 1st choice, and to insincerely express differences among
equally-preferred lower choices, it can be said that wv doesn't
give assurance that there won't be motive for strategy. But the
assurances that are important to me are the assurances that a
majority can get its way without stratgegy, or that a sincere CW
can be protected, without strategy under plausible conditions, and
without reversal strategy under all conditions. I'm interested mostly
in assurances to majorities, then.

Sure, SFC & GSFC only apply when no one falsifies a preference.
And yes, that means that someone who wants to steal the election from
a sincere CW is forced to attempt the risky strategy of offensive
order-reversal, instead of mere truncation. Making it difficult &
risky to steal an election counts, for me, as an advantage.

Besides, I expect that offensive order-reversal, on a scale sufficient
to change an election result, will be vanishingly rare in public
political elections. That means that SFC's & GSFC's requirement that
no one falsifies a preference is really a safe bet, and SFC's &
GSFC's guarantees are very real.

Rob continued:

, which is a shame,
because they sound great.  Mike also seems to be more worried about
preventing truncation than tied preferences in general.

I reply:

Not so much _preventing_ truncation. But merely not letting truncation
steal an election from a sincere CW, or from a member of the
sincere Smith set.

Lazy truncation is common. I've also observed strategic truncation
in a pairwise-count vote. To me, it's good if that can't mess things
up by stealing the election from a sincere CW or member of the sincere
Smith set. And again, I like the SFC & GSFC guarantees that can
be made when we prevent that.

Rob Continued:

I think of
truncation only as another case of tied preferences; in other words, I
think it's silly to consider votes like A>B>C=D=E=F as a separate case
from votes like A=B=C>D=E>F.

I reply:

But I'm mostly interested in what truncation can do to a CW,
a member of the sincere Smith set, or to a majority of the voters,
when a method is vulnerable to truncation.

Rob continued:

I also disagree with Mike's philosophical justification for
winning-votes.  He says that if A pairwise beats B, the strength of the
defeat should measured only by the support for A because the B voters
have already been overruled.  But shouldn't each side be ignored
equally?

I reply:

The losers in that pairwise contest have no right to be considered
equally with the winners. If A pairwise beats B, then the people
have spoken. The B>A voters have lost.

Of course I acknowledge that this is a subjective matter of
personal aesthetics.

I emphasize that the B>A voters _were_ considered equally to the A>B
voters, when those 2 vote totals were compared. The B>A voters lost
fair & square. After that, they're the losers in that pairwise
comparison, and the equality is gone.

When it's necessary to overrule the public's pairwise vote results,
then, we regrettably have to overrule some voters. But let's overrule
the number of voters whom we're overruleing. It's my claim that the
B>A voters have already been overruled by the public vote between
A & B. They've already lost, and it isn't the count rule that's
overrulilng them. It's the more numerous A>B voters who have overruled
the B>A voters.

Rob continued:

If A beats B 50:49, A's victory is 50 strong.  But if there's
just one more B>A voter, it's a tie, and two more B>A voters give B a
victory 51 strong.  Doesn't that seem ridiculous?

I reply:

Sure, we have to choose which wrong-looking result we want to avoid,
and we have to choose which properties, standards & criteria are
most important. It's a subjective judgement. But if we disregard
A>B 50 to 49, then we have to answer to 50 voters, when we
overrule the public vote that we've won.

Yes, and if there's another defeat that's 49 to 1, it _does_
look ridiculous to drop that instead of the 50 to 49 defeat.
But I say let's ignore how that looks, because we don't owe anything
to those 49 B>A voters in the 50 to 49 defeat. They've lost fair &
square and have nothing to complain about if we don't overrule their
defeat.

Rob continued:

It makes more sense
to me for the votes on the losing side to annihilate an equal number of
votes on the winning side, giving the margin (which is what I've always
understood the word "majority" to mean anyway).  Otherwise what good are
they?

I reply:

WV _does_ let the B>A votes annihilate an equal number of A>B votes,
when comparing those 2 vote totals to find out which one wins, to
find out what the public rule about the AB comparison. All I deny
is that the B>A voters have to still be considered after they've
lost.

Another thing: That subtraction destroys information about majority.
It thereby spoils majority rule.


Rob continued:

Shouldn't a 48:30 victory be stronger than a 51:46?

I reply:

Sure, and a 49 to 1 victory looks much stronger than a 50 to 49
victory. There's an aesthetic case, then, for acting on that.
But there's, for me, a stronger responsibility to minimize the
number of voters who are overruled, and, for me, losing voters have
already been overruled by a bigger group of voters.

Rob continued:

I would feel
cheated if I were a voter among that 46.

I reply:

But what about my claim that those 46 lost fair & square?
Isn't it worse to take victory to a winning group of voters than
to not give victory to a losing group of voters?

Admittedly there's an aesthetic case for margins. So as I said,
it's a subjective individual judgement.

Rob continued:

As much as I'd like to see a rule better than
margins at preventing strategic preference-reversal, winning-votes isn't
it.

I reply:

But it's the wv methods that make it easier to deter offensive
order-reversal strategy, by mere defensive truncation. In any case,
though, I don't think offensive order-reversal will happen on a
significant scale, and if people aren't falsifying preferences, then
SFC's & GSFC's guarantees apply, with wv.

Rob continued:

After all this, though, I still think Approval Voting is the best
public proposal, unless the public demands a ranked-ballot method, in
which case Path Voting has my support until a Condorcet completion
method that punishes preference-reversal is found.  I'm not optimistic.

I reply:

The IRVie environment in which we'd be working must be taken into
account. The IRVies are good at propagating their misunderstandings
of Approval. I've found that out from experience in LWV discussion,
where people have been exposed to the IRVie line.

And IRV is much more difficult to implement than CR. And for the reasons
in this & the previous paragraph, it seems to me that CR is more of
a winning proposal against IRV, where people have been well-saturated
with IRVie propaganda. CR has the advantage of great familiarity &
popularity. The IRVies will have a more difficult time making people
misunderstand CR.

Right now CR is my #1 choice for most promising public proposal.




(I hope it's obvious that I mean no disrespect to Mike or Blake.  I've
followed their contributions to the list for quite a while and I agree
with them most of the time)

No offense taken. We can discuss different choice-standards and I
don't take any offense.

Mike Ossipoff

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