[EM] Yee diagrams

[fsimmons at pcc.edu]

Dear Ka-Ping,I recently posted a messagehttp://lists.electorama.com/pipermail/election-methods-electorama.com/2008-November/023242.html to the election methods list in which I gave a link to your Voting Visualization websitehttp://zesty.ca/voting/sim/I suggested including Random Ballot as one of the basic methods for comparison.Also I would like to suggest a less abstract interpretation of the diagrams:Each circled point represents a potential place for a regional conference.  The cost of getting there is proportional to the Euclidean distances from the attendees' homes.  Each non-circled point represents a different possible center for the distribution of these homes. Etc.  For a non-deterministic method L, the point P is colored RGB(p,q,r) iff the respective winning probabilities (under method L) for the three conference centers are p, q, and r, given that the distribution of attendees' homes is centered at P.As I have said many times, your diagrams are the most convincing argument we have against IRV for people that take the time to understand their significance.If people can understand a uniform distribution more easily than a normal distribution, then I would suggest that we use voters uniformly distributed inside circles of a radius equal to the distance between the two most widely separated options (conference centers).My Best,Forest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20081113/4150d5be/attachment-0002.htm>