I've made passing mention of my STV-CLE method (the CLE stands for Condorcet Loser Elimination) a couple of times, but the recent "Proportional Condorcet Voting" thread encouraged me to finally give it a decent introduction.

Essentially, STV-CLE is just like STV, except that instead of eliminating the candidate with the fewest first-choice votes, you eliminate the Condorcet (completion) loser. I've included a vote calculator (using Beatpath ranking) in the attached file (requires Python 2.4 with GMPY).

An interesting result occurs when the voters are on a unform linear spectrum (i.e., voters = [random.uniform(0, 1) for i in xrange(num_voters)], and ballots are ranked by the distance between candidates and voters). If there are candidates sufficiently close in position to multiples of the quota (except at the end of the spectrum), they will be elected. That is, a 1-seat election will elect a candidate near 1/2; a 2-seat election will elect candidates near 1/3 and 2/3; a 3-seat election will elect candidates near 1/4, 1/2, and 3/4; etc.

Attachment: stvcle.tar.gz
Description: application/gzip

----
election-methods mailing list - see http://electorama.com/em for list info

Reply via email to