Dear Eric, you wrote (7 Jan 2003): > The algorithm implemented is Beatpath Winner, now mentioned on > the site.
I have suggested that the strength of a pairwise defeat should be measured primarily by the number of winning votes and secondarily by the margin. An implementation can be found here: http://www.topica.com/lists/RankedPairs/read/message.html?mid=1601349191&sort=d&start=34 When there is no unique beat path winner then random ballot should be used to choose from the potential winners. I recommend that the beat path method should not be restarted with the potential winners since this could lead to a violation of monotonicity. Markus Schulze ---- For more information about this list (subscribe, unsubscribe, FAQ, etc), please see http://www.eskimo.com/~robla/em
