From: [EMAIL PROTECTED]
> Kevin, I did not understand how many voters there were in
> your simulations.
> Other than that, they look interesting and confirm my own
> (and others') widely held suspicions.
I think the way it works is that he picks a number between
0 and S for each candidate. The number of candidates is
S/10. All the other voters are assumed to vote completely
randomly. (Presumably, for this to work, S/2 should have
the highest probability, but I think all values from 0
to S have equal probability).
This is then assumed to be the total number of range votes
for that candidate.
The voter's ballot rates each candidate from 0-10. He then
adds in the voter's vote and records any increased the
utility of the election due to the voter having an effect.
Raphfrk
--------------------
Interesting site
"what if anyone could modify the laws"
www.wikocracy.com
---- election-methods mailing list - see http://electorama.com/em for list info
