[EM] Why the concept of "sincere" votes in Range is flawed.

[Juho Laatu]

--- On Mon, 26/1/09, Raph Frank <raphfrk at gmail.com> wrote:

> On Sun, Jan 25, 2009 at 11:10 AM, Juho Laatu
> <juho4880 at yahoo.co.uk> wrote:
> > Another approach to offering more
> > flexibility (maybe not needed) and
> > more strategy options (maybe not
> > wanted) is to allow the voter to
> > fill the pairwise matrix entries
> > in whatever way. This means that
> > also cycles can be recorded.
> 
> More flexibility can be obtained using a range/score like
> ballot.
> 
> For example, assuming a rating of 100 counts as 1 vote, if
> I rate one
> candidate at 150 and another at 25, then that counts as a
> full vote in
> favour of the first candidate.  However, if the difference
> is less
> than 100, then the value entered into the matrix is only a
> fraction of
> a vote.  The condorcet winner is then determined based on
> the sum of
> all the matrices, as per normal condorcet.
> 
> What is nice about this method is it gives the voter full
> control over
> what kind of voting method they would like.
> 
> It is possible to emulate
> - approval
> - range/score
> - condorcet (equal ranks allowed)
> 
> This assumes that all the voters fill in the ballots in a
> certain way
> for each method.
> 
> Allowing full flexibility (subject to a max value in each
> square) on
> filling out the matrix itself would give the voters more
> freedom.
> However, I think that would yield to massive voter
> overload.  Also,
> there might be strategic issues.

Here's one approach to collecting all
kind of data at one go. Also this one
doesn't allow cycles nor all
combinations of preference strengths.
A ratings based ballot with values
from -inf to inf and some (voter
given or fixed) threshold values.

A = 1000
B = 200
max_support = 100
C = 50
approval = 30
D = 1
min_support = 0
E = 0
F = -100
max_preference_strength = 10

Approval interpretation is A=B=C>D=E=F.
Range interpretation is A=B=100, C=50, D=1, E=F=0.
Rankings interpretation is A>B>C>D>E>F.
Rankings interpretation with preference strengths is A>B>C>D>(0.1)>E>F

I'm not sure if anyone really wants
all this. The first practical
implementation might be a Condorcet
election that covers also ratings
data for polls.

Juho