[EM] Reply to Moe's Multiple Winneres Revisted post:

[I like Irving]

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 03/22/01
Dear Moe St. Evergreen and cohorts,

     So, you are still looking for a multi-seat proportional method that is
a bit better than Cumulative Voting.
     Do I have a deal for you?  Yes, indeed I do, in fact, I have five
better models for you to look over.

           1)  Pure STV
           2)  Single Transferable Vote (STV)
           3)  Choice Runoff
           4)  Bottoms Up (IRV)
           5)  Single Non-Transferrable Vote (SNTV)
           6)  Cumulative Voting
           7)  Limited Voting

     That's it, there are no more non-partisan methods.  If you were
looking for some super-duper method that involves a combination of
Condorcet and Approval, forget it, does not exist, it is merely a myth.
You have been mislead.  The only way in which Condorcet and Approval can be
used in multi-seat elections is via single-seat districts of multi-seat
elections.

     Anyway, getting back to reality, the next method up my ladder of
quality methods is, `drum roll please', Single Non-Transferrable Vote.

     SNTV is a very simple method, the voter has only one vote for only one
choice that he can cast for only one candidate.  The candidates with the
highest number of votes are the winners.
     In the 1995 Cambridge election, SNTV would have elected eight of the
same nine candidates that STV did actually elect.  So, you can say SNTV was
ninety percent as good as STV for this one election.  I have not checked
any elections since then.

     But you're asking for a good method that allows preference choices.
     That would be good thinking on your part, but first you should know
about the `Golden Rule of Preferences' that applies to all preference
methods, single or multi-seat.  Please be aware that I did not write this
rule.
     The Golden Rule is: "Later preferences cannot help nor harm earlier
preferences."
     In other words, your lower choices are not to be used to help defeat
your current top choice.
     Or, in other words, none of your lower choices are to be used in the
calculations while your current top choice is still a contender.
     Keep the Golden Rule in your mind while you are looking for the best
preference method for your needs.  Let it be your guiding light.

     The next method up the quality ladder of methods is `Bottoms Up' and
it is the first preference method in the stable.  It works the same as
Irving (IRV) except it is used for multi-seat elections.  The runoff cycles
stop at the number of candidates you want elected.  In the 1995 Cambridge
election, Bottoms Up would have elected the same nine candidates that STV
did elect.
     There is a slight improvement that you could add to Bottoms Up, and
that is to use a quota.  The transferring of votes must work within the
confines of this quota.  I call it Choice Runoff, but it is actually the
last part of STV.  It's use will require you to learn how to do fractional
transfers of votes, but as long as you must learn that, you may as well go
up the ladder to STV, or better yet, up to Pure STV.
     STV and Pure STV are more complex than SNTV, but they are
understandable.  After all, this is not rocket science.  A high school
student, good in math, would be able to understand STV and be able to work
three or four hundred ballots in a few hours using only a hand calculator.

You wrote:
Moe:  "STV/hare seems rather complex, and there seems to be some confusion
on what the fairest implementation of STV is, and on the simplest
procedural language."

Don:  There is no confusion for me nor for the high school student on what
is the fairest implementation of STV.  Pure STV is STV without the dirty
little secrets that STV has picked up over the years.  You do not need to
use any of the dirty little secrets in your STV, if you don't use them,
they are not a problem.

Moe:  "It also appears to be more arbitrary on ties, where an
random/arbitrary tie break could, if I read it right, completely eliminate
a candidate, that might otherwise have large majority approval."

Don:  This policy of eliminating all the tied bottom candidates is one of
the dirty little secrets that STV has picked up over the years.  You don't
need to use this policy and you shouldn't use this policy as an excuse to
reject the best proportional method, Pure STV.

Moe:  "It seems to me that the Condorcet winner should always be among the
winners in a multiple winner system."

Don:  You are in error, Condorcet is not a valid method for any election,
single nor multi-seat. Condorcet is not to be used as a standard to judge
any other election method.  Condorcet fails the Golden Rule of Preferences.
Don't use your love of Condorcet as an excuse to reject Pure STV.

Moe:  "I think we would prefer rankings to CV, but I am not sure exactly
what to advocate, and I am not sure I can trust the so-called Center
for Voting and Democracy when it fails to balance its out biases."

Don:  I support the Center on IRV, but I do not suport their multi-seat
proposal of three seat districts using the Droop quota, two more dirty
little secrets are the Droop and small odd number districts.  I tried to
talk Rob Richie out of that setup, but to no avail.
     The best thing you can do in this search for a better method is for
you to turn the search over to someone else.  Turn it over to one of you
cohorts, Larry or Curly, either one cannot but help do a better job than
you.
     Better yet, that high school student would do a far better job of it
than all three of you guys put together.  Anyway, you are not fit for the
job, you have too many biases.  Your misguided dislike of IRV is causing
you to not consider the best method, Pure STV, only because it contains IRV
runoffs.

Moe:  "Any suggestions would be welcome, especially if they include
pointers to well described procedures that can be easily tested and
advocated."

Don:  You are not being honest, what you really mean is that "Any
suggestions would be welcome" as long as they do not include the runoff
cycles.  You will accept any weird method that is not STV.
     I repeat, you are not qualified to lead this search for a better
method, turn the job over to some high school student if you want the job
done right.

Moe:  "This list is called, "election-methods", but perhaps it should be
called, single-winner-methods, as that seems to be the focus of this list."

Don: `You cannot see the forest for the trees.'
     It is true that most of the discussions on this list involves
single-winner-methods, but what you do not seem to comprehend is that
single-winner includes single-seat districts which is the method of choice
of most of the major multi-seat elections in the world today.  The game
plan is that of keeping the single-seat districts, but to corrupt them with
some weird method like Condorcet or Approval Voting.
     While new single-winner methods will be used in single-seat elections,
you must understand that the lovers of the weird methods, Approval, Borda,
and Condorcet (ABC), are mainly interested in multi-seat elections because
that is the area in which they hope to realize the most gain for small
third party candidates. The ABC Weirdos favor the lower parties, and will
allow their candidates to get elected, now and then, from time to time.
Under the right conditions, Approval will even propel a candidate, with
only a few votes, up and ahead of all the other candidates, cannot you see
that to be true?  I'm only telling you this so that you can start your real
education on election methods.

Moe  "The only response I received to asking about multiple winner
elections, was one specifying STV/Hare, and which was posted by what
appeared to be an IRV advocate, who referred me to a rather loudly
biased IRV advocating website, the so-called Center for Voting and
Democracy (should be called the Center for IRV/STV, or else remove
their personal bias from their Center's media outreach)."

Don:  I am that IRV advocate, and you got what you asked for, so stop your
bellyaching.
      You asked: "Could you send me a pointer to a clear description of an
STV system..."
      I pointed you to CVD's website, for sure they must have a description.
      Did you not find a description of STV?
      If you did find a description, then what's your problem?
      You got what you asked for.
      The quicker you turn this job over to someone else the quicker your
group will have a better method of electing their members.  You are too
biased.

Moe:  "Only members of a committee can vote for the delegates representing
that committee."

Don:  When we have a small number of votes, we run the risk of too few
votes for too many candidates.  A good solution is to require nominations
as follows:
- - - - - - - - - - - - - - - - - - - - - - - - - - - -
Rules of Nomination:  by Donald Davison

     It would be helpful if the number of candidates could be reduced
without harming the rights of anyone.  This can be done by setting some
rules of nomination.
 (1) A candidate must be nominated by another council or board member.  If
a member wishes to be considered for the position that member should start
with at least enough support to get nominated.
 (2) The nominator can only nominate one candidate that accepts the
nomination.  The nominator must make a commitment to only one candidate.
 (3) The nominator cannot be a candidate and a candidate cannot be a
nominator.  The nominator cannot run against the person he nominated -
again he must be committed to the person he nominated.  And each candidate
must be committed to themselves as candidates.  These rules would limit the
nominations to a maximum of five for a ten member council or board.  These
rules still allow any member to be nominated with only the low support of
one other member.
     We can further improve on this by requiring a second nomination.  The
person who is the second would also come under the same rules of
nomination.  Using a second would limit the nominations to a maximum of
three for a ten member council or board.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Donald Davison, who is now running seven days behind with his email.