[EM] Viability (finding the odds after you've voted)

[Forest Simmons]

Here's a rule that agrees with the MinMax (third refinement) rule for five
and fewer slots, but might be more clone proof in general:

We seek to find two adjacent slots to merge in a list of slots of various
viabilities.

Let X be the slot not in the center of the list with the greatest
viability.  If X is above center, then let L be the list of slots below X,
otherwise let L be the list of slots above X.

Apply this procedure recursively to L.

Stop when the recursive list L gets down to two slots.

This rule is motivated more by the "approve down to the front runner,
inclusive only when the second place guy is below front runner" strategy
than by Weinstein's max voting power strategy.

Forest

On Sun, 3 Aug 2003, [iso-8859-1] Kevin Venzke wrote:

> Forest,
>
> I tried out my suggestion and it worked incredibly badly.  The given odds
> seemed very exaggerated for unviable candidates; it suggested to almost
> every faction that its expectation was near midrange.  I tinkered with exponents
> to exaggerate higher-ranked candidates, but I didn't get much improvement.
>
> However, I wonder if, using MPCR as you've defined it, except that the viability
> calculation each round took into account the total number of points awarded by
> a ballot, MPCR might be made more clone-proof.  (It would not make any difference
> in my trials, though (every faction votes one candidate per rank), so I can't test
> it.)
>
> I think you understand what I mean by "clone-proof."  As it is, I wonder if an
> ideology would have an advantage by running clones, and differentiating their
> slots.  They presently get comparable viability per candidate, and could force
> other factions to merge their own candidates too soon.
>
>  --- Forest Simmons <fsimmons at pcc.edu> a écrit :
> > And we know that we have the right rule for the three slot ballot!
>
> I wonder if it still works with more than three candidates for those ranks?
> Maybe you've already thought this through, but might there be a clone problem?
> I'll think about this.
>
> > So in the case of the Three Slot Max Power Cardinal Rating method, we
> > could justly claim the name MVP, the "Max Voting Power" method.
> >
> > Note how this corresponds with the technical definition of voting power,
> > the probability of your vote being pivotal.
>
> OK, but I'm trying to remain optimistic that we'll figure out "Better Than
> Expectation" at some point.
>
> Kevin Venzke
> stepjak at yahoo.fr
>
>
> ___________________________________________________________
> Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français !
> Yahoo! Mail : http://fr.mail.yahoo.com
> ----
> Election-methods mailing list - see http://electorama.com/em for list info
>