[EM] a method based on candidate separation

[Forest Simmons]

Here's the MinMax(separation) method:

The score for candidate X is the separation of X from the most distant
candidate that defeats X pairwise.

The candidate with the smallest score wins.

As mentioned in a previous post in this thread, there are various ways of
defining a metric for measuring separation between candidates.

Here's a similar separation based method that occurred to me in this vein:

"Min Radius of Vulnerability"

The radius of vulnerability for candidate X is the distance from X to the
most distant position in issue space that would be preferred by a majority
of voters over X.  The candidate with the smallest radius of vulnerability
wins.

These two methods are different because the first method only uses
distances to actual candidates in the election, while the second method
takes into account all possible candidates.  If there are many candidates
scattered at random all over issue space, then the two methods are
practically the same.

Forest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140605/30618f11/attachment-0001.htm>