[EM] Let's clear up some confusion

[Michael Ossipoff]

On Wed, Oct 3, 2012 at 3:28 PM, Juho Laatu <juho4880 at yahoo.co.uk> wrote:

> On 3.10.2012, at 20.37, Michael Ossipoff wrote:
>
>>> (In that case, probably you should include that difference also in the definition of what the ballots mean.)
>>
>> Wrong. My definition of Symmetrical ICT fully specifies the method and
>> its count rule.
>
Juho:

You said:

> No doubt about that. I was interested in if you expect regular voters to know that there are special rules on how the top and bottom ties are handled.

[endquote]

I answered that question. No, I don't expect or assume voters to know
those count rules. i do expect and assume that voters would be assured
that, with Symmetrical ICT, Approval, or Score, they have no incentive
or need to ever vote someone over their favoriote. They'd be assured
that, with Symmetrical iCT, Approval or Score, there is never any
reason to not vote their favorite at top. (That means with no one
voted over hir).

They'd also be assured that, when they've voted for some candidates,
there is no need to vote for additional candidates, in order to fully
help the candidates for whom they've already voted.

The guarantee in the paragagraph before last is FBC. The guarantee in
the paragraph before this one is Later-No-Help.

Another thing that Later-No-Help guarantees is that in a u/a election,
there is no need to vote for any unacceptable candidates.

By "vote for", I mean "mark, on the ballot". If you want a more
precise definition, "vote for" means "vote over someone", as I've
defined that term on EM.

Has this been repetition? Of course.

You said:

If there is anything interesting left

[endquote]

That depends entirely on what you're interested in. Maybe you're only
intrerested in margins (whilch means that you can proudly add the
Plurality Criterion to the other unimproved Condorcet
criterion-failures such as FBC, Later-No-Harm, Participation and
Consistency).

If you want something interesting, then here's something:

A few days ago I asked the advocates of unimproved Condorcet what
properties of unimproved Condorcet make it good  (in comparison to
Symmetrical ICT) as to outweigh unimproved Condorcet's failures of
FBC, Later-No-Help, and unimproved Condorcet's retentioin of the
chicken dilemma.

The interesting thing is that the advocates of unimproved Condorcet
methods (such as Beatpath, Ranked-Pairs, River, Goldfish, Kemeny, and
VoteFair)
are unable to cite any desirable property of unimproved Condorcet that
would outweigh those failures.

For a more detailed version of my question to unimproved Condorcet
advocates, I refer you to my posting in which I asked that question,
with the question mentioned in the subject-line.

Ii mention that because you implied that the subject might not be
interesting enough for you, and so I wanted to mention to you an
interesting fact.

You said:

, maybe the question if you recommend the voters to rank sincerely or
if you recommend them to sometimes use the top ties (although the
candidates are not equally good).

[endquote]

Good question. In a public election, I'd emphasize that their best
strategy in a u/a election is to equal top rank all of the
acceptables, and to not rank any accceptables.

If you dont't like that, then I'll remind you that the situation is
even worse in unimproved Condorcet. The need to top rank all of the
acceptables is still there, but, with unimproved Condorcet, it's
accompanied by a risk that one top-ranked candidate will pairbeat
another, and thereby change the winner from an acceptable to an
unacceptable.

Therefore, in the unimproved Condorcet versions, you'd never really
know what to do. You'd have to try to judge whether some acceptables
are much more winnable than others, so that refusing to top-rank Y (an
acceptable) is justified because Y isnt likely to win anyway, and
top-ranking hir could make X (another acceptable) lose.

That's a problem that you'd never have to worry about in Symmetrical
ICT. It's the old lesser-evil dilemma of FBC-failing methods.

Another problem that you'd have in unimproved Condorcet:

To fully help the acceptables against the unacceptables, you'd need to
rank the unacceptables in reverse order of winnability (to the extent
that you could guess that). That's due to unimproved Condorcet's
failure of Later-No-Help.

You might say that Symmetrical ICT's u/a strategy sounds a lot like
Approval. Yes, and the fact that the best that you can get, in u/a
strategy, in rank methods, is simply Approval strategy  is another
good reason to propose Approval (or maybe Score) instead of any rank
method.

Along with that, when proposing Approval instead of any rank method,
you get compliance with Participation and Consistency, two
embarrassment criteria whose failure will be fatal to any enactment
proposal (when the opposition makes sure that everyone hears about
it).

Additionally, you get an incomparably simpler method definition and
count rule, and an incomparably simpler and easier count (therefore
without the count-fraud-vulnerabilty of every Condorcet version)..

You asked me how I'd instruct a voter. But you know that I don't
propose any rank method for official public elecions.

I only recommend rank-balloting for informational polling. Symmetrical
ICT is my best recommendation for informational polling.

When I introduced the ICT rank-balloting poll at Democracy Chronicles,
I asked voters to rank sincerely at top. I admitted that, with any
rank method, that often won't be their best strategy (certainly not in
a u/a election). But I asked them to rank sincerely at top anyway (2nd
choice below 1st choice, etc.), because we all would like the
information that can be gotten from a rank-balloting informational
poll. Sincere ranking at top is very helpful for that purpose.

However, I suggested that people not rank any unacceptable candidates.
I said that there is no reason to. That is a completely true statement
if the rankings are counted by Symmetrical ICT.

Mike Ossipoff