To Blake re: strategy

[MIKE OSSIPOFF]

I'd said:

>Sure, but there are also countries that use IRV, and that's obviously
>the best evidence about IRV in action.

Blake replied:

So do you think IRV's strategy problems are worse than successive
elimination?

I reply:

You're talking about a series of ballotings, after each of which the
lowest votegetter is eliminated?

I haven't given any consideration to successive elimination. It
of course isn't the same strategically as IRV, since the voter can
make strategy decisions as the elimination votes procede, based on
current & recent vote totals & eliminations.

So I wouldn't want to say that Successive Elimination is similar enough to 
IRV for use to use it to judge IRV's merit.

Besides, you aren't very clear about exactly where it's used.

As I said, IRV is better for judging IRV.
Blake said (regarding evidence from Australian voters):

Blake continued:

We obviously have different opinions on what constitutes good evidence,
and if anyone is still reading this discussion, they'll have their own
opinion.

I reply:

You can say what you want about your judgement of the validity of
that eveidence, but it's all that's been gotten and provided here.

What can I say--"Sorry that you haven't taken the
trouble to bring any evidence"?

I'd said:

>For you, "sometimes" is good enough a guarantee. You have a gambling
>nature.

Blake replied:

If that was my argument, I wouldn't agree with it either. Saying that
IRV "sometimes" doesn't give strategic need to vote someone equal to or
over your favourite couldn't be used to justify anything. My argument
was in terms of general tendencies.

I reply:

And I just showed you that strategic incentive to vote someone over
one's favorite occurs in an ordinary, fairly typical, situation,
in IRV.

But, given your advocacy of RP(m), it doesn't surprise me that
IRV looks pretty good to you.

I'd said:

>Here there's lots of money paying for IRV promotion. And the IRVies
>are promising people that IRV will end the lesser-of-2-evils problem.
>So of course that sounds good to people. People are misled to believe
>that IRV won't give incentive to rank insincerely.

Blake replied:

It seems to me that you make much of the argument that the public is on
your side, but when it is pointed out that this is not the case, you
declare that they are just hapless dupes.

I reply:

Sometimes when you tell how something seems to you, you forget to
tell us why it seems so to you. As you've done before, you're again
quoting something that I didn't say. I never said that the public
was on my side. I said that the LO2E problem is important to very
many people. Important to voters because they admit that it dominates
their voting. Important to small party people because they say so.
Important to CVD and other IRVies because they continually claim that
IRV gets rid of the LO2E problem and its closely associated spoiler
problem.

Nor did I say "hapless dupes". All I said was that IRV promoters
mislead people to believe that IRV will get rid of the LO2E &
spoiler problems.

You, Blake, have a tendency, when quoting, to wax poetical and use 
exaggerated dramatic language that wasn't used by the person you're
quoting, and quote things that don't resemble anything said by the
person that you claim to quote.

Blake continued:

But a criterion isn't just a desired property. It's a pass/fail test.
If I think that the failers are better than the passers, I don't think
much of the criterion.

I reply:

Maybe you believe that all of the failers of Participation, Consistency
, IIAC, Heritage & Regularity are better than all of the passers.

I'd said:

>Unlike RP(m), also, Approval doesn't add new
>disadvantages, like failure of Participation, Consistency, IIAC,
>Heritage, & Regularity.

Blake replied:

But you yourself must repudiate most of these criteria.

I reply:

Not really.

Blake continues:

I mean, a
criterion like consistency is based on the idea that it doesn't make
sense for a method to behave in a particular way. But wouldn't you
agree that that is false, and sometimes it does make sense for a method
to violate consistency?

I reply:

Wbat makes sense about a candidate winning in each county, when
the election's ballots are counted in the counties, but losing
statewide when those same ballots are counted in one big statewide
count. I can't agree that that makes sense.

But what I would agree to is that of course it's well understood that
many good voting systems will sometimes do things that don't make
sense in some respect. I'd also agree that there are excellent methods,
the wv methods, that violate Participation, Consistency, IIAC, Heritage,
& Regularity, but which offer other criteria compliances that make
up for it. What makes wv good in spite of failing those other criteria
is the other criterion compliances that it offers.

IRV doesn't offer anything that even begins to outweigh the
disadvantage of failing all of those criteria. I like wv in spite
of those failures, because it complies with the majority defensive
strategy criteria. IRV's failures of the criteria in the previous
paragraph aren't balanced or outweighed by anything, and I feel
that those criteria matter, and that they do count against a
voting system that fails them.

As Arrow showed, you can't have everything in a voting
system. So you choose which criterion compliances are more important
to you.

I'd said:

>That's a bizarre thing to say. What you mean is that, while
>passing IIAC is a good thing, you feel that methods that don't
>pass it have other advantages that you consider more important.
>
>Methods that don't pass IIAC might pass CC, SFC, GSFC,
>&/or SDSC. But IIAC is compatible with FBC & SARC, as are
>Participation, Consistency, Heritage & Regularity.

Blake replied:

Well, I don't care about any of those criteria (except the Condorcet
criterion). I don't support any criterion unless I can think of a
good justification for supporting it.

I reply:

A most reasonable policy. Sometimes criteria are justified, for
a person who likes a particular standard, by the fact that they
measure for that standard. For instance, I like FBC, SARC, WDSC,
SDSC, SFC, & GSFC because they measure for the LO2E standard, and
I like WDSC, SDSC, SFC, & GSFC because they measure for the majority
rule standard.

Sometimes criteria are justified because their violation obviously
doesn't make sense. That's true of Monotonicity, Participation,
Consistency, IIAC, Heritage, & Regularity.

Violations of Monotonicity & Participation seem worse, because
they're about a voting system acting contrary to how a particular voter
votes, which means that they're about undemocratic behavior by a
voting system. No it isn't good when wv fails Participation. But
the choice among voting systems is a tradeoff, and wv has much to
offer. IRV offers nothing in return for its failure of Monotonicity,
Participation, Consistency, IIAC, Heritage, & Regularity.

Mike Ossipoff








_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx