[EM] "good method" ? was "IRV ballot pile count (proof of closed form)"

[Jonathan Lundell]

On Feb 14, 2010, at 4:46 PM, Abd ul-Rahman Lomax wrote:

>> We may disagree with the counting method that is applied when
>> >> > 35:A
>> >> > 32:B>C
>> >> > 33:C
>> 
>> occurs, but it seems very clear that the Condorcet winner in this case
>> is C, as you seem to agree with me in this case.
> 
> Yes. The A voters express no preference between B and C. A is the plurality winner. But only 35 voters support A, 65 oppose A by their votes.

We don't actually know that. Suppose the B>C voters are saying, "I love B, hate C, and have no idea who A is". Granted, in this limited example, they could easily have voted B>A>C to indicate something like that. But if there are a lot of candidates who may be unknown to many voters, it's asking a lot for them to list them all (whether or not we allow equality of preference). 

I've been thinking of the possibility of handling indifference differently. Suppose that '*' means "all candidates not explicitly ranked". Then

A>B is interpreted as usual, implying A>B>*

But A>*>B mean A is best, B is worst, and all the others are indifferent, without having to rank them explicitly between A & B.