[EM] Independence from clones, and Condorcet//FPP criterion failures

[Michael Ossipoff]

Kristofer:

I'd said:

> The Condorcet Criterion, Condorcet Loser, MMC, Smith, and Schwartz
> lose their meaning and value when voters experience the
> preference-distorting strategy needs of methods that fail FBC and CD

You replied:

This is the crux of our disagreement. I disagree that a method that
fails FBC and CD leads to a garbage in, garbage out situation where
every voter is so afraid of having the greater evil win that he votes
for the lesser evil.

[endquote]

I've always acknowledged that TUC advocates speculate that voters will
start voting sincerely instead of continuing to try to vote
strategically. I agree with you that their TUC advocacy depends on
that hope.

I question the reliability of their assumption about that.

You said:

If one expects the voters to be mostly honest, then FBC and CD is no big
deal.

[endquote]

With a different electorate, IRV would work fine too.

In fact, with voters who knew what they were doing, even Plurality's
shortcomings could be overcome (even if not easily).

But it would be better to automatically avoid the worst strategy
problems, instead of just hoping for sincerity in spite of strong
sincerity-distorting strategy-needs.

You said:

What evidence do I have that the voters are mostly honest? They were in
Burlington in both elections, even though the honesty didn't pay.

[endquote]

Honesty didn't pay, but you think that they'd keep repeating it, in
subsequent IRV elections.

They'd probably been promised by IRV promoters that honesty works fine
in IRV. They found out different. That's why they threw-out IRV.

I suppose we could speculate about how "honestly" they'd vote next
time, if IRV had been kept, given the lesson that they'd learned about
IRV.

You said:

 There are
other examples, too, like organizations using Condorcet methods and not
repealing them, MJ exit polls, and multiwinner methods results.

[endquote]

If Beatpath works in some organizations, that's nice, and I'm not
concerned about that use of it.

Actually, I'm more concerned about public political elections.

You said:

All considered, I think that evidence is of a little greater weight than
your "voter voting Democratic above third-party in a web poll" sample.

[endquote]

I didn't speak of only one example. I didn't experience many people's
reactions to new voting systems, but the two that I did experience and
report were consistent with eachother--Perceived favorite-burial need.

I'm only proposing Approval and Score for the U.S.'s official public
elections. I'm only recommending ICT or Symmetrical ICT for public
political informational polling in the U.S. You may be an authority
regarding how people would vote somewhere else, and that's great. But
it isn't of interest here.

Voters here believe that the winner will certainly be the Democrat or
the Republican. They believe that it's all important that it be the
Democrat instead of the Republican. It's a u/a election, with the
Republican as the unacceptable, and the Democrat as the only winnable
acceptable.

The optimal strategy of traditional unimproved Condorcet (TUC) is
clear in that situation (though it's unknown in u/a elections in which
the perceived situation isn't quite as simple):

Vote the one-and-only winnable acceptable candidate alone in 1st
place. If you want to be thorough, rank the winnable unacceptables in
reverse order of winnability, below all of the acceptables.

I'd favorite-bury if we did our official public elections by TUC, if
it were clear that only one of the acceptables was winnable.

Speculate all you want to. If you've met and spoken to more U.S.
voters than we who live here, then certainly we should take your
speculations seriously, especially if you can justify them with the
few known facts about TUC strategy.

You said:

Furthermore, let me say that an objection to a criterion failure
claiming the failure can be bypassed if only the voters keep it in
mind... it does little to convince me.

[endquote]

My purpose isn't to convince you. It's sufficient to have you post
your defenses of TUC, as you have done, in order to show how well they
hold up. Anyone can judge that for themselves. I invited people at EM
to post such defenses, and I'm glad that someone has. Your
co-operation is appreciated.

What I said was that a criterion-failure may or not be a problem.

In Approval, Score, ICT or Symmetrical ICT, the addition A2, a clone
of A, whom the A-preferrers like much better than A, could present the
A voters with a dilemma, "Should we gamble for not approving (or
top-ranking) A, when we approve A2?". I acknowledged that before.

It would be reasonable for TUC advocates to be outraged by a strategy
dilemma if TUC were free of strategy dilemma.

But if A and A2 are both acceptables in a u/a election (which our our
official public elections are) , then there is no dilemma. Optimal
strategy unequivocally calls for approving, top-rating or top-ranking
both A and A2. (in Approval, Score, ICT and Symmetrical ICT).

In any case, offering a chance to gamble for something much better is
an unimportant "problem", as problems go. The addition of clone A2
doesn't make things worse for the A-preferrers.

You said:

By having the voters act as
manual DSV systems to patch up problems with the method, you burden the
voters

[endquote]

Translation: You don't want there to be any strategy incentive. Sorry,
but all methods have it, including TUC. Your problem regarding the
strategy-incentive of ICT when a clone is added, but not regarding the
worse strategy-needs of TUC, is reminiscent of someone fastidiously
dusting the surfaces in a room that has large and smelly dog-feces
piled up all over the carpet.

You said;

, whereas a method that passes independence from clones outright
doesn't make such demands.

[endquote]

Not in regards to that clone. But other, different and worse, strategy
demands. I've discussed them.


Also, by expecting the voters to patch the
problem by equal-ranking, you introduce something quite similar to the
Nader, Bush, Gore problem in Approval. The A-voters in my example have
to judge whether A is secure enough that they can vote their conscience
and say A > A2, or whether the other candidates are dangerous enough
that they have to vote A = A2.

[endquote]

Ok, for the person who prefers A to A2, the situation is worsened by
the addition of A2 (contrary to what I said above).  Fine, but I never
said that Approval, Score, ICT and Symmetrical ICT were strategy-free.

What you're apparently forgetting is that TUC has (at least) the same
problem, when it isn't known that everyone will vote for your
acceptable candidates as clones. As I said, your TUC top-end dilemma
exists, and can be without a solution, even in a u/a election. (I told
you about the unequivocal favorite-burial optimality, in the simple
(existing) situation with only one winnable acceptable candidate). One
difference, however, is that TUC's dilemma involves favorite-burial.

In a u/a election, Approval, Score, ICT and Symmetrical ICT have no
dilemma or uncertainty about how to vote at top-end. Of those 4
methods, only ordinary ICT shares TUC's bottom-end strategy incentive.

And strategy in Approval, Score, ICT and Symmetrical ICT never calls
for favorite-burial. Your favorite burial need in TUC is a problem of
a whole other magnitude.

Though little can be said about TUC strategy, what little can be said
shows that favorite burial is clearly, unequivocally, optimal under
the conditions that actually exist in our official public elections.

You want to liken the ICT's clone "problem" to TUC's favorite-burial
problem. It won't work.


You said:

If you want the voters to act as manual DSV systems, I have a better
suggestion. Actual DSV.

[endquote]

DSV isn't strategy-free either. And it's not a feasible proposal.
Additionally, its computation-intensiveness gives it the same
count-fraud vulnerability of other rank methods.

Thank you, but I recommend only Approval and Score as proposals for
official public elections in the U.S.

You said:

And by extending the concept of patches just a little, to outside the
system itself, even Plurality can be patched to pass independence from
clones. It's called an exhaustive-runoff primary.

[endquote]

As I said above, even Plurality could be well-used, and its
considerable disadvantages (difficultly) overcome, by an electorate
that knew what it was doing.

You said:

If you want a method that passes independence from clones

[endquote]

No, it's you who want that.

You continued:

, why not look
for one? I never claimed CD and independence from clones to be
inherently incompatible.

[endquote]

If you know of a method that meets The (preferably
legitimately-defined) Condorcet Criterion, FBC, CD and
Clone-Independence, then trot it out.

LNHe, or 0-info LNHe, would be nice, if you can manage that too.


You said:

And yes, there does exist a Condorcet method that passes Mono-Add-Top.
It's even advocated by a member of this list.

[endquote]

How long has it been since it was advocated on EM?  :-)

Is there some reason why you don't want to say what it is? What does
it lose in return (aside from FBC and CD)? Because Clone-Independence
is important to you, then I also ask if it's clone-independent.

You said:

If we're talking about self-servingly selective, I notice that in regard
to your original (pre-fix) "improved, with power truncation" way of
counting top and bottom rank, you said "That's why, in keeping with what
the voter would prefer and wishes with hir equal top and equal bottom
rankings, Symmetrical ICT interprets equal top and bottom ranking as it
does".

[endquote]

Yes, equal top or bottom ranking voters' preferences, intent and
wishes justify and call for the (unfixed) Symmetrical ICT's
interpretation of equal top or bottom rankings.

Regrettably, unfixed SICT doesn't work with the CD compliance solution
that I intend for such methods.

So I made one small change, to fix Symmetrical ICT.

You said:

You also spoke quite warmly about (unqualified) LNHelp.

[endquote]

Though bottom-end strategy is less important than top-end strategy,
it's desirable to reduce need for it, as LNHe compliance does. I
previously thought that SICT meets LNHe. It doesn't strictly meet
LNHe, but it meets 0-info LNHe and Strong 0-info LNHe, which is
effectively just as good.

You continued:

Then you
find out that keeping power truncation would lead to CD failure, and
oops! Out it goes. So much for "the voter's wishes" of how his ballot is
interpreted.

[endquote]

I made it clear that it was regrettable that the top-count solution
for CD compliance necessitated giving up the equal-bottom-voter-wishes
advantage of SICT.

But it turns out that it isn't necessary to entirely give that up.
It's only necessary to stipulate that two candidates can't beat
eachother, and that, if unfixed SICT says they would, then the one
that beats the other is the one ranked over the other on more ballots
than vice-versa.

That's a small and reasonable abridgement of the rights of
equal-bottom-ranking voters.

You continued:

So much for plain LNHe, which was so desirable up until
that point.

[endquote]

Nonsense. LNHe didn't stop being desirable. I said that I wouldn't
trade CD for LNHe. At that time, I repeated my statement about the
desirability of LNHe, and the regrettableness of having to give it up.

Actually, SICT never did meet strict LNHe. But, as I said, it meets
0-info LNHe and Strong 0-info LNHe, which, for practical purposes, is
just as good as meeting unqualified LNHe.

You said:

It's now okay to weaken both.

[endquote]

It's regrettable when it's necessary to give up one desirable property
for another. I made it clear that giving up bottom-end respect for the
wishes of equal-ranking voters was regrettable (and it turns out that
it isn't necessary to do so after all--with that small SICT fix that I
described), and that giving up LNHe was regrettable too.

And no, Strong 0-info LNHe isn't a significant weakening of LNHe.
Compliance with it is, for practical purposes, just as good as
compliance with unqualified LNHe.

Forget about the "groupthink" remarks. I wanted to hear your defense
of TUC, and I have.

As you said, it isn't productive to comment about others' possible motivations.

As for "weaken[ing] the reform movement", of course there isn't any
cohesive reform movement. There can't be one, unless advocates of
methods and criteria become more willing to justify what they
advocate. I wouldn't count on that from TUC advocates.

You said:

Oh, and about the value of independence from clones: it lets voters know
there aren't going to be any split-votes problem from similar
candidates. You know, like what happened in South Korea, or in 2000 in
the US.

[endquote]

Wrong. Nader and the Democrat weren't similar candidates or clones, as
you defined a clone-set.

In general, in reality, the lesser-of-2-evils giveaway favorite-burial
problem isn't a clone problem. Clone-Independence won't solve or avoid
that problem.

You said:

It'll also assure the voters that powerful parties can't run a
bunch of copies to win.

[endquote]

I guess you're comparing clone-independent methods to Copeland,
because you're citing a problem that doesn't exist in the methods that
I recommend.


You said:

Kinda like FBC lets the voters know how they
don't have to vote anyone above their favorites.

[endquote]

...except that clones are a special particular kind of candidates,
whereas the favorite-burial problem is more general. And
favorite-burial is a greater distortion of preferences than any
strategy needed in Approval,.Score, ICT or SICT.

You continued:

And that will happen
without the voters having to play the manual DSV roulette.

[endquote]

If strategy need is "having to play the manual DSV roulette", then
voters in Beatpath and other TUC methods do indeed have to play the
manual DSV roulette, for the reasons that I've been describing.

You said:

(And please don't say that slight violations of independence from clones
don't matter while slight violations from FBC do. Or if that's your
preference, don't expect others to share it.)

[endquote]

Obviously it's for the individual to judge for hirself which problems
are worse, or which criteria are more important.

Mike Ossipoff