Mr. Eppley wrote: If it doesn't take much time, how about doing these 3-candidate random simulations: ****** 3. Every voter ranks all 3 candidates. Some rank two candidates as equal first choices (e.g., A=B > C) and the rest have no equal rankings. (There are 9 possible orderings.) ----- D- did 40 simulations. Results- 37 Condorcet winners, 3 circular ties Average winning percent of all winners 59.5 (range 50.2 to 84.6) 32.5 percent of the voters voted a A=B>C type vote (range 13.1 to 51.3 percent of the voters) 67.5 percent of the voters voted a ABC type vote (range 49.7 to 86.9 percent of the voters- reverse of the preceding since vote for 2 option is not in this simulation). Again, such 32.5-67.5 split is highly suspect for a real election.
