[EM] VoteFair representation ranking recommended for Czech Green Party

[Raph Frank]

On Sun, May 2, 2010 at 8:48 AM,  <VoteFair at solutionscreative.com> wrote:
> See above.  Also, based on this comment I've added three links to this
> answer in the FAQ.

Great.

> Previously I had presumed that election-method experts would go to the pages
> that contain the rigorous descriptions (as above).  I had intended the FAQ
> for average folks (who run away from mathematics and even too many numbers).

Well, being able to find stuff on a website is not necessarily the
same skill as election methods knowledge.

I hit the "about votefair ranking" link, as I just wanted a description.

If you search for web-sites on other methods, they normally give a
description of the method on the front page or no more than 1 link
away.

>> I think this method might be proportional even when electing more than
>> 2 candidates.  Does it meet Droop proportionality?
>
> The method does not get repeatedly used the way other methods keep reducing
> the weight of each voter's ballot.  In what I recommend, all ballots return
> to full weight after each two successive candidates are chosen.

It only gets Droop proportionality for 2 seats.

Effectively, a faction with 1/3 of the vote would be able to get half
of the seats, i.e.

Party A: 40%
Party B: 40%
Party C: 30%

Party A+B will get all the seats, assuming the supporters vote as a bloc.

>> However, if there is a condorcet tie, then S might be one of the
>> candidates which defeats W pairwise.
>
> I'm not sure what you are saying here.

I mean if there is a condorcet tie, then the winner might end up with
a minority.

7) A>B>C
4) B>C>A
5) C>A>B

A vs B: 12 - 4
A vs C: 7 - 9
B vs C: 11 - 5

I think A wins here under Votefair?

I created a survey on your site with the above votes:
http://www.votefair.org/cgi-bin/votefairgenballot.cgi/votingid=59588-57285-34547

This gives A as the most representative/popular and B as the 2nd most
representative.

If we remove ballots which vote for A first, we get

4) B>C
5) C>B

C become (temporary) runner-up.

We then compare A to C

Ballots which mark A above C
7) A>B>C

Ballots which mark C above A
4) B>C>A
5) C>A>B

We now have to de-weight the first group of 7 ballots.  However, they
represent only 43.75% of the votes.

Presumably, this means that they are de-weighted to zero?

The other 9 voters would then select C as the 2nd most representative
candidate.  This is not the result that your site gives.

However, maybe your site de-weights the ballots by

(Va - 0.5)/Va

Va = fraction of the votes held by A's supporters.

(0.4375 - 0.5)/0.4375 = -0.1429

This gives:

-1) B>C
4) B>C
5) C>B

If it considered the -1 as +1, then that is a tie between B and C.

I tried with

7) A>B>C
2) B>C>A
5) C>A>B

and it still gives B the 2nd place.  I think maybe the negative number
of causing a problem.  Maybe the vote totals are unsigned so -1 is
equal to the max possible number of votes?