[EM] Fwd: FW: IRV Challenge - Press Announcement

[James Gilmour]

Kathy
Can you please provide a link to a directory where we can find all of the relevant documents, and with their correct titles, as
referenced in the various legal texts?
I found my way to this page of links:	
  http://electionmathematics.org/em-IRV/

I have looked at all of the documents listed there and have read some parts of some of them in detail.  The language used is
confusing in places because the word "vote" is frequently used when the reference is to "ballot paper".  For example, a "vote" has
only one value, namely "1", but a ballot paper may, at different times, have different values, the maximum of which is "1 vote".
This distinction between "ballot paper" and "vote" is helpful, if not essential, to a proper understanding of how STV-PR works and
is applied.  (But perhaps the use of the word "vote" in this way is to comply with some relevant USA legal language?)

>From the definitions given in this document:	
  http://electionmathematics.org/em-IRV/AMENDED_REV_OrdinanceMNTaskForce.pdf	
it is clear that for multi-winner STV-PR elections the City of Minneapolis would use the Weighted Inclusive Gregory Method (WIGM)
for transferring surpluses.  This method uses fractional transfer values ("Gregory"), based on transferring all of the relevant
ballot papers ("inclusive"), each at the correct value ("weighted").  Thus at no time can any ballot paper ever have a value
exceeding 1.0000 vote (as they are working to four decimal places).

I cannot comment on the accuracy of the various examples and counter-examples of the WIGM calculations because I cannot be sure I
have been looking at the correct documents.  It is very difficult to know what is what when a small part of a calculation is
presented without its full context (but that is typical of legal documents that make lots of cross-references to other documents and
productions).


Leaving all the arithmetical details aside, it would appear from this document:	
  http://electionmathematics.org/em-IRV/ReplyMemoJG10-6-08.pdf	
that the Plaintiffs' main argument against STV-PR is that its use would be "unconstitutional" because the second preferences used to
transfer surplus votes are examined and given effect before any other second preferences are considered.  This is portrayed as
creating an inequality of one person's vote compared to another.

It is essential that the surpluses are transferred BEFORE any other action because the destination of the surplus could change what
has to happen next in identifying the required number of winners from among the larger number of candidates.  Specifically, the
surplus could reverse the order of the bottom two candidates and so change the one who will be excluded at the next step in the
counting process (always assuming an exclusion is the next step).

It is a novel idea that giving effect to the contingency choices of the voters in a way that ensures proportional representation
would be "unconstitutional".

James



> -----Original Message-----
> From: election-methods-bounces at lists.electorama.com 
> [mailto:election-methods-bounces at lists.electorama.com] On 
> Behalf Of Kathy Dopp
> Sent: Wednesday, October 08, 2008 5:30 PM
> To: Jonathan Lundell
> Cc: election-methods at lists.electorama.com
> Subject: Re: [EM] Fwd: FW: IRV Challenge - Press Announcement
> 
> 
> Yes.
> 
> Thank you Doug. I woke up this a.m. realizing that fact.
> 
> However, then the City must admit that all voters who voted 
> Doug>Meg and did not list a third choice, have their votes 
> diluted to less than one vote, since the vote values these 
> voters retain is 1 - 0.0434
> 
> In either case, no matter which way excess vote values are 
> calculated, voters' votes are not valued nor treated equally, 
> nor are the power of voters' votes equally applied to 
> determine outcomes.
> 
> Cheers,
> 
> Kathy
> 
> On Wed, Oct 8, 2008 at 9:45 AM, Jonathan Lundell 
> <jlundell at pobox.com> wrote:
> > On Oct 7, 2008, at 5:17 PM, Kathy Dopp wrote:
> >
> >> Yes. I was the person who pointed out that the City's own example in 
> >> its Memo shows how some votes are valued at more than one (1) for 
> >> some voters in the City's example (and if the City's example were 
> >> more realistic, it would show how some voters' ballots would be 
> >> valued at less than one(1) vote.)
> >>
> >> If you actually take the time to read my affidavit and the City's 
> >> example in its Memo, you will see that Exhibit G and the City's 
> >> example clearly mathematically prove the truth of the Plaintiffs' 
> >> arguments.  The mathematics is irrefutable, despite any argument you 
> >> could try to make to divert attention from the mathematical facts.
> >
> > You've made a miscalculation there, by the way.
> >
> > At 
> http://electionmathematics.org/em-IRV/ReplyMemoJG10-6-08.pdf page 
> > 5, you write:
> >
> > "Doug's electors carry a weighted vote — .6667 + .3333 + 0.0434 = 1.0434."
> >
> > You neglect that fact that Meg does not retain the entire 3333 weight 
> > from the second choices of her transfers from Doug, but rather (.3333 
> > - .0434), just as Doug doesn't retain the entire 1.0000 of his votes, 
> > but rather (1 - .3333).


No virus found in this outgoing message.
Checked by AVG - http://www.avg.com 
Version: 8.0.173 / Virus Database: 270.7.6/1713 - Release Date: 07/10/2008 18:40