On 2/17/23 04:46, robert bristow-johnson wrote:

Nice! Two comments:

• The ballot format allows for equal rank. I'm not familiar with legal
  language, but from the document, it seems like everything needs to be
  precisely described. Thus it might be a good idea to specify how
  equal-rank first preferences are counted: does A=B>C give one point to
  both A and B, or half a point to both?

• There's a chance it would backfire if someone engineers a Condorcet
  cycle. Perhaps better is that the highest Plurality scorer among the
  candidates with the highest number of pairwise wins, wins instead?
  That's Copeland,Plurality and thus Smith.

Of course, if the legislative document is meant to implement
Condorcet,Plurality specifically (and it's past the "which method do we
choose" stage), then disregard what I just said and go for
Condorcet,Plurality - we'll see how it goes :-)

-km

I think we want our laws to be precisely described.  Otherwise we leave too much to the judges.

This isn't Borda.  There are no "points".  We want to make sure, as I pointed out (in my paper) with the North Dakota Supreme Court ruling of 1911, that what we are ultimately counting are people.  Not points, not marks, not scores, but people.

So we got this definition:

(5) “Paired comparison” or “head-to-head comparison” means an evaluation of two candidates for the purpose of determining which candidate has the greater number of ballots ranking that candidate over the other candidate in the pair.

and we got this provision:

(d)(1) Additional tabulation. Upon tabulation of the ballots, if no candidate receives a majority of first-ranked preferences, the ballots shall be tabulated again by paired comparison and examining every possible paired comparison. In each paired comparison, the presiding officer shall note the winning candidate in each paired comparison or if there is instead a tie.
(2) Condorcet winner. If a candidate is the winning candidate in every
paired comparison, the candidate shall be declared the winner of the election.
(3) No Condorcet winner. If there is no candidate that is the winning candidate in every paired comparison, then the candidate having the plurality of first-ranking preferences is declared the winner.

Now, perhaps the language here:

"... the presiding officer shall note the winning candidate in each paired comparison or if there is instead a tie ..."

needs to be more explicit about what the meaning of being a winning candidate in a particular paired comparison.

It's only if one candidate receives more votes than the other, that they are a winning candidate.  Tying with the other candidate does not mean either is a winning candidate.

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

On 2/17/2023 10:48 AM, robert bristow-johnson wrote:

= = = = =
[KB]  Do I understand paragraph (3) correctly: If there's no Condorcet
winner, then you default to a First Past the Post winner?

  - Ken Bearman, Minneapolis MN

Yes, given current usage data in the U.S., this is what happens about 0.2% of the RCV elections.  Out of over 500 RCV elections in the U.S., only once did the election demonstrate a cycle.  And that was Minneapolis Ward 2 in 2021.  (And BTW, Nicolaus Tideman analyzed this and thinks that this evidence of a cycle is inconclusive because they had only 3 ranking levels and 5 or 6 candidates.  But, as the ballots were marked and assuming none of the voters would have ranked anyone below 3rd-choice if there was more ranking levels, then it was a cycle.)

Now, in my opinion, the problem here becomes a political problem that might eclipse the technical problem.  The technical problem is about how to keep any cycle from being gamed or incentivizing tactical/strategic voting in future elections.  This is a quite esoteric technical problem and I would invite Markus or Nic or anyone else with their Condorcet-consistent method to explain it to a bunch of legislators, let alone the public.

We explain Condorcet this way: If a simple majority of voters mark their ballots preferring Candidate A over Candidate B, then Candidate B is not elected.

Why should Candidate B be elected?  The IRV people twist themselves into pretzels trying to answer that.  (Their answer is either "we've been doing it this way for several decade and IRV is well tested" or, when challenged that the "well tested" has shown failure in 3 out of 500 cases, then they trot out the old "Candidate Milquetoast" argument, saying, essentially that IRV saved us from getting Montroll in 2009 or Begich in 2022 because they are both milquetoast and "kiss the baby" candidates.  A real horseshit argument.)

So, in the less than 0.2% of the cases where it is impossible to satisfy the simple ethical principle of the Condorcet criterion, then we have to explain to the public how and why the candidate who was elected was chosen.  What is it about this winning candidate that makes him/her a more appropriate choice for election to office than any other candidate in the Smith set?  What, that the public can see and discern, indicates electoral support?  How would we answer that question?

That California bill (from a previous session, I believe it's a dead bill now) used the Hare IRV winner as the contingency winner if no Condorcet winner. https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202120220AB2161  I think that, for less than 0.2%, defending this choice over plurality is difficult.

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

The candidate left unranked on the fewest ballots may be a better choice of
default, because plurality is subject to vote splitting ... the main
motivation for voting reform in the first place.

On Fri, Feb 17, 2023, 11:47 AM robert bristow-johnson <
rbj@audioimagination.com> wrote:

Forest, it seems to me that, in the one case outa 500+ RCV elections in the U.S., that because it goes to a cycle, then the vote is already split.  It's split pretty evenly three ways.

This is a divided electorate and also a little schizoid or goofy.  You have a divided electorate, an extremely close 3-way race, no one elected will have a sweeping mandate of the vote, and Arrow prevails.

So you gotta elect someone anyway and you have to explain to people why that person was elected.  What is it that candidate has that the other two (or more) candidates don't have?

If it were up to me, I'd just say Schulze or Ranked Pairs (margins) and leave it at that.  But then I will have some trouble explaining that to some other schlub, and especially if there was a cycle.

IRV tried to simplify the debate with a method that essentially has no overall principle.  Condorcet has a pretty simple and clear principle.  First we need to get people hooked into that.  Then when trouble happens (and Arrow assures us it will eventually), then we have to reach a little further and explain, convincingly, to the public how and why the winner was identified as such and elected.

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

Suppose the whole election was to choose one of the three candidates ... no
noncycle members were in the election at all.

Would you recommend FPTP Plurality over Approval?

On Fri, Feb 17, 2023, 6:07 PM robert bristow-johnson <
rbj@audioimagination.com> wrote:

I dunno.  I would recommend Condorcet RCV and, only in the super-rare case of a cycle, I would recommend plurality of first-choice votes because it's no worse than we already have now with FPTP and it's simple to explain to voters why that one candidate (the one with more first-choice votes than anyone else) is, in some sense, uniquely more preferred by the electorate over any other candidate.

In a competitive 3-way race, I dunno if Approval would be better than FPTP.  I do know, if it were Approval, that I would have a tactical decision to make regarding my second-favorite candidate.

But with RCV, I would have no tactical decision to make.  I would know immediately what I would do with my second-fav.

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

Robert, ordinarily I not only find most of your posts very interesting
(when I understand them, as I'm not a mathematician or well-versed on
all the technical stuff that most everyone here posts) but I also admire
your patience.  But in the case of your last two posts (below) I object
twice.  Then I have a question about cycles.

 (1) Your wrote, "If a simple majority of voters mark their ballots
preferring Candidate A over Candidate B, then Candidate B is not
elected.  Why should Candidate B be elected? The IRV people twist
themselves into pretzels trying to answer that." My objection is this: 
You describe a Condorcet election rule and then ask why [implicitly in
an IRV election] the Condorcet rule isn't followed.

Your question isn't entirely honest.  If you're talking about an
election where the criteria/rules are Condorcet, then no IRV (or FPTP)
proponent I know of will say B should be elected.  If you're not talking
about a Condorcet election, then your question is inapt and the
"pretzels" comment is just a bit insulting, I hate to say.

 (2) In your second post, responding to Forest Simmons, you wrote, "IRV
tried to simplify the debate with a method that essentially has no
overall principle."  My objection is this: IRV proponents employ a
simple principle to explain the process, viz. that the vote counting in
the absence of a first-count majority is a series of runoff elections
using one ballot/precinct appearance.

If I misunderstood what your "no ... principle" comment was about,
please clarify for me.  (I know you detest IRV, but that comment also
was just a bit insulting.)

 (3) As I wrote above, I don't grok a lot of the technical stuff you
all discuss in the EM list.  I think, though, that I understand what a
cycle is.  So I'd like to give you more information about the
one-in-500+ you've referred to, the Minneapolis Ward 2 election in 2021,
before I ask a question.

There were five candidates on the ballot.  (Scroll down to "Original
tabulation" here:
https://vote.minneapolismn.gov/results-data/election-results/2021/council-ward-2/)
The five were, in descending order, a Democratic Socialist; a Democrat
(called Democratic-Farmer-Labor or DFL in MN), who fell just short of
party endorsement; a Green; another DFLer; and a Republican.

The top three differed very little on issues or policies. However, there
were three Charter amendments on the ballot*; Question 2 would have
abolished the police department and replaced it with a Department of
Public Safety.  (Q2, the only one that failed, was extremely
controversial and probably helped increase city-wide turnout.**)  Of the
top three, Arab opposed it and Worlobah and Gordon supported it.  That's
the only substantial issue difference I know of among the three.

I'm giving you all these local details because I wonder, Does "cycle"
mean much when the candidates in the cycle are essentially
interchangeable?  I.e., whichever of the three won, the supporters of
the other two would have approved of most official policy outcomes
anyway,so what does "cycle" mean outside of election theory?  OTOH, if
that interchangeability is irrelevant in discussing cycles, please
explain for me.

Thanks for any responses.
  - Ken Bearman, Minneapolis MN (not in Ward 2)

 *
https://vote.minneapolismn.gov/results-data/election-results/2021/ballot-questions/
** You may have seen a photo of nine Minneapolis Council members above a
big "Defund Police" sign.  Q2 came out of that event, as did a lot of
hot national political chatter.

On 2/17/2023 8:07 PM, robert bristow-johnson wrote: