[EM] Why Not Condorcet?

[Dave Ketchum]

I see deciding to use Condorcet as important.  To go with that we  
would need to decide how to resolve cycles.

On May 13, 2010, at 3:37 PM, Kevin Venzke wrote:
> Hi Robert,
> --- En date de : Mer 12.5.10, robert bristow-johnson <rbj at audioimagination.com 
> > a écrit :
>> On May 12, 2010, at 7:40 PM, Kevin Venzke wrote:
>>
>>> I think approval-completed Condorcet is a better
>> proposal because there's
>>> a greater chance people would understand how it
>> works.
>>
>> what exactly is that?  the winner being the candidate
>> in the Smith set with the most 1st-choice votes?
>
> My preference is that when there is no CW, the candidate with the most
> votes in any position is elected. But you could limit it to the Smith
> or Schwartz set.

Any Condorcet method means the voters doing ranking of candidates and  
the counters identifying the CW if there is one.

If no CW the counters find a cycle of candidates and the particular  
method decides which cycle member shall win (some methods use most  
votes as mentioned above; some use margins - difference in liking of a  
pair; more vs less).

Either way, each member of a cycle would be CW if the other cycle  
members were omitted.

While quality of resolving a cycle is important, an important detail  
of that is being able to describe to voters meaningfully what is being  
done to/for them.
>
> Kevin Venzke

Backers of other methods can claim theirs are better than Condorcet.   
Consider what Condorcet offers as to ease of voting for the one OR  
MORE candidates you like best, including indicating liked a little vs  
more liked - and ease of counting such votes

I read of arranging ballot data in a triangle, rather than in a matrix  
as usually described.  A minor detail, but what would be easiest for  
ballot counters is most important while they count, though rearranging  
for later processing would be possible.