[EM] SITC vs [what?]

Mon Sep 24 21:56:58 PDT 2012

This is part 1 of a 2-part reply:

On Mon, Sep 24, 2012 at 4:11 PM, Juho Laatu <juho4880 at yahoo.co.uk> wrote:
 > I will not comment the Dodgson and changing vs. adding votes
related misunderstandings.

It is solved. The misunderstanding had nothing to do with changing vs
 adding votes. The misunderstanding was about the ambiguity of "any".
 Dodgson looks at the _sum_ of the margins against you, to find out how
 many pairwise votes would have to be reversed, ignored, or added (take
 your pick)  to make you the CW. MinMax(margins), instead, looks at the
 single largest margin against you.

As I said, it depends on how "any" is meant. Now I realize that you
 meant MinMax(margins). The only part that I missed was regarding its
 advantage(s). I spoke of its disadvantages, and about the natural
 authority of SITC's choice, when no there isn't exactly one unbeaten
 candidate.. Someone else might like the idea of MinMax(margins)'
 circular tie solution. Probably lower in social utility though. Most
 would probably agree that there's no such thing as one single best
 sincere winner, when ranking is sincere and there's a circular tie or
 no CW. It's an arbitrary matter of individual preference, unless
 someone argues in terms of SU. Even then, it would be difficult to
 prove that one particular circular tie solution is the best.

In any case, though, MinMax(margins), as i said, uses an illegitimate
 definition of "CW".

I hope that misunderstanding is now solved. My example best sincere
 winner criterion was meant to refer to the Minmax(margins) philosophy.

 >
 > On 24.9.2012, at 16.33, Michael Ossipoff wrote:
 >
 >> If you think that
 >> MinMax(margins) or Dodgson is better than Symmetrical ICT, under
 >> sincere voting, you have yet to tell why.

>> Do you really think that would help hir
 >> status against opposition in office better than being the most
 >> favorite candidate in the top cycle?
 >
 > I don't know what "most favorite" means here.

Then I'll tell you. By "most favorite", I meant voted favorite, voted
 in 1st place, by the most people. The usual meaning of "the most
 favorite" is "favorite to the most people".

You said:

Minmax(margins) can elect outside the top cycle if such a candidate is
 closest to being a CW (measured in number of required additional
 votes)

[endquote]

Now,  you see, that's exactly what I was talking about. Now you're
 back to Dodgson again, aren't you. "Closest to being CW" is Dodgson.
 "Closest to being CW" is not MinMax(margins). "Closest to being CW" is
 about the _sum_ of the margins against you, rather than about the
 largest margin against you.

But it doesn't make any difference whether you really mean Dodgson or
 MinMax(margins), because what I've said about one applies to the other
 too.

You said:

I don't claim that this criterion would be the best one for all
 elections, but it is one that sounds usable for some needs.

[endquote]

Pretty much anything is _usable_, at least for some needs. That hardly
 makes for an argument for Dodgson or MinMax(margins) vs SITC.

If that's your most favorable and strongly-worded recommendation of an
 alternative to SITC, then there has indeed been a misunderstanding,
 because i thought that you were saying that a different method would
 do better under sincere voting. What you're saying seems to merely be
 that, because SITC is strategically good, then something else must be
 better when voting is sincere. I've twice told you why that isn't
 necessarily so. It's a fallacious way to try to show that SITC isn't
 good when voting is sincere. What you have is a "maybe".

>> So you're saying that different voting systems should be used for
 >> different elections. But, as each new election comes near, who decides
 >> which method will be used for that particular election?
 >
 > I'd expect one series of elections to stick to one method (and be
based on one stable understanding on what kind of a candidate is the
best sincere winner).
 >
 >> Should we use different voting systems for presidential and
 >> Congressional elections? If so, then which one would be better (by
 >> ideal sincere winner) for the presidency,and which would be better for
 >> Congress?
 >

You said:

> Those two elections are very different by nature, and therefore they could well have different targets / understanding of whom to elect with sincere votes. The question on which method and which sincere winner criterion to choose is very difficult since changes to the current system may mean changes to the very basic concepts of the system. There are multiple options. One interesting question is if the president should be from a large party of if he/she could be a compromise candidate that has no major party behind him/her.

[endquote]

Best to let the voters decide that. If they want a large party, then
 give they will choose a large party. If they prefer smallness as a
 party-attribute, then they will choose a small party.

But, guess what? What if they don't evaluate parties based on how
 large or small they are? Could it be that maybe they'd instead
 evaluate parties with regard to honesty, un-corruptness, and the
 desirability of the parties' policy proposals?

But, in any case, it isn't for your or me to decide whether the
 winner's party should be small or large.

Maybe voters will compromise, maybe not. Again, that's their choice.
 People will compromise to get their best outcome, but hopefully they
 have good judgement about what is unacceptable to them.

I recommend that, if someone is progressive who prefers a progressive
 other than Jill Stein, for November, they should vote for Stein as a
 compromise. As a favorite-burial compromise. U.S. voters are hardly
 strangers to favorite-burial.

Of course rank methods compromise automatically. But they all
 nevertheless require additional compromise on the part of the voter,
 for optimal strategy.

You'd like to believe that in most societies, rankings will be
 sincere. I'm going to make a general statement about most societies:
 People like to get the best outcome possible. In the U.S., in England,
 and in Australia (with IRV), it's common knowledge that lots of people
 say that they favorite-bury, because it's optimal strategy. And you
 know what? They're right. It is. I myself would favorite-bury in an
 IRV election. In November, if the election were IRV, I'd rank Stein in
 1st place, and Roseanne Barr in 2nd place, though I prefer Barr to
 Stein.

l don't disagree with the use of favorite-burial in Plurality and IRV
 (or in unimproved Condorcet). I merely disagree with some other
 progressives' notion of what "acceptable" means. One nice thing about
 FBC methods is that the consequence of misjudgement about that aren't
 so bad. Both sides of the "Democrats acceptable?" question would agree
 on that. Everyone would agree on that, except of course the mass
 media.

Maybe you think that my evaluation of candidate and party-platform
 acceptability is wrong. Maybe I claim that your evaluation of it is
 wrong. But we both agree that we don't want the other's
 acceptability-evaluation to have really bad consequences. So we both
 would like the other to not have to drastically distort due to those
 mistaken acceptability-evaluations.

As I said, the owners of the mass media here are very strongly
 motivated to get people to favorite-bury. They're quite consistent,
 continual, tireless and relentless about it. I've describe the virtual
 news blackout regarding non-Republocrat candidates, parties and
 policies. In the U.S., it's as if they don't exist, because the media
 say so.

Combine that with people's natural inclination (based on observations
 and personal reports in widely-separated countries) to vote to get the
 best result that they can. What does that say about your sincere
 ranking ideal? La La Land.

 If you think that most societies would rank sincerely, under these
 conditions that I've described above, then you're living in La La
 Land.

But you needn't worry about the choice between large and small
 parties, or between compromise and favorite--that's for the voters to
 decide for themselves.

I agree that sincere voting is good. That's why I like voting systems
 that don't drastically distort voter sincerity.

The societal damage resulting from favorite-burial is drastically,
 dramatically, worse than any adverse result of not using a method that
 would choose the best ideal sincere winner, if voting were sincere.

..quite aside from the fact that I've told you why SITC does well when
 people rank sincerely.

...quite aside from that fact that you don't even really make a
 serious claim that your MinMax(margins) or Dodgson would do better.
 You only have a _speculation_ that maybe...

...quite aside from the fact that I've told you where people would
 rank as sincerely as you expect: La La land.

You said:

In the Congress one has to decide e.g. if one wants to keep the
 two-party approach or not. The end result might be two very different
 election methods.

You see, you're still talking like a Soviet-system-advocate. It's
 entirely antidemocratic to purposely choose a voting system that will
 elect a number of parties chosen by you.

It's blatantly obvious that Plurality preserves a (phony) 2-party
 system that is really a 1-party system.

Your question is, "Gee, I don't know--Should we use a method that
 forces preference-distortion that will preserve a '2' party system?"

Wrong question. With Approval, Score, or SITC, the voters will decide
 that for themselves. If they like the Republocrats best, then they'll
 keep on electing them, no matter how good the voting system is.
 They'll do so even with Approval, Score, or SITC. Don't worry. Those
 methods won't make a multiparty system if people really prefer the
 Republocrats. Stop worrying.

>> Of course, judging by how well they choose the ideal sincere winner
 >> assumes that you still think that there won't be a chicken dilemma,
 >> and can tell why.
 >
 > I see the sincere winner criterion and strategic concerns as two
separate topics.

What is the sincere winner criterion? The methods that I advocate are
 the most likely to encourage sincere voting, or relatively sincere
 voting, in comparison with other methods.

And that _is_ a strategic topic. That's because certain strategy-needs
 are what can and does distort sincere voting--the only thing that can
 distort and prevent sincere voting.

You said:

The method that will be eventually used may deviate from what the
 sincere winner criterion points to if there are strategic concerns
 that must be addressed by selecting a method that has the required
 strategy related properties.

[endquote]

You're repeating your speculation. I refer you to my explanation that
 the respect for voters' preferences, intents and wishes that makes
 SITC choose better under sincere voting, also results in less strategy
 need. You think that strategy-freeness and good sincere results are
 mutually incompatible. That's an unsupported speculation of yours.
 You've been repeating a speculation.

>> If there will be defection in situations like the chicken
 >> dilemma examples, then can you still advocate Beatpath,
 >> MinMax(margins) or Dodgson over SITC, by saying they will get sincere
 >> rankings?
 >

> You have to pick the method so that strategic concerns will be properly addressed. I don't want to take position if one of those is >absolutely better than others (since that is not relevant to my claim).

I don't know what that means.

>
 >>> I tried to cover all the questions in your mail. You may point
out the unanswered ones, so I can check what I can do with them.
 >
 > I don't think the following four questions that you gave as a
response are ones that I left unanswered, but new questions or new
formulations. I'l check them anyway.
 >
 >> 1. What makes you think that MinMax(margins), Dodgson, or Beatpath
 >> won't have a chicken dilemma?
 >
 > I already said that I do believe that basic Condorcet methods are
not very prone to this problem. I know that you disagree. Maybe you'll
find one day a proof that will convince me.

I'm going to repeat this all over again for you:

The A and B voters detest C. They consider both A and B to be much
 better than C. But the A voters like A better than  B. And the B
 voters like B better than A. Each of those factions (A and B) would
 rather elect their own candidate.

Sincere rankings:

Numbers at left represent percent of voters:

27: A>B>>C
 24: B>A>>C
 49: C (no preference between A and B, for simplicity)

The A voters know that co-operation is needed to defeat C. So they
 co-operate, because someone must.

The B voters know that the A voters are co-operative and responsible,
 and will surely co-operate by ranking B in 2nd place. So they choose
 to take advantage of the A voters by not ranking A.

Actual rankings:

27: A>B
 24: B
 49: C

The B voters have allowed C to pairbeat A.

C beats A, 49 to 27.
 A beats B, 27 to 24.
 B beats C, 51 to 49.

It's a circular tie.

With 3 candidates, the unimproved Condorcet(wv) versions are the same
 as eachother.

The smallest votes-against is the one against B.

B wins.

Evidently with margins, the defection-success isn't as guaranteed.

But no doubt we can adjust the margins to make MinMax(margins) elect
 whomever the example-writer wants it to elect.

Next time, I'll post for you an example in which MinMax(margins) and
 Dodgson fail too (they're the same in a 3-candidate circular tie).

>> Must I do that, to show you their
 >> chicken dilemma? Request it and I will.
 >
 > No need since I don't expect that to change my opinions. It could
be a wasted effort. I'm interested if there is something really
convincing, but maybe better leave this topic this time, with the
assumption that I would not believe it anyway.
 >
 >> 2. What makes you so sure that the United States won't have a
 >> significant amount of favorite-burial, when unimproved Condorcet, such
 >> as Dodgson, MinMax(margins) or Beatpath, is used?
 >
 > I'm not "sure" but my best guess is that basic Condorcet methods
would work well enough.

Repetition of that isn't the same thing as support for it.

You said:

My confidence is based on theoretical studies, experiences with
 Condorcet in non-political elections and experiences with IRV in
 political elections

[endquote]

You mean like in Australia, where people say that they're
 favorite-burying? And when Australians say that favorite-burial is
 optimal in IRV, they're right. It's optimal. I'd favorite bury in an
 IRV election.

To be continued...

Mike Ossipoff