[EM] RE : Re: Clone proofing Copeland

[Chris Benham]

Kevin Venzke wrote:

>Brian,
>
>--- Brian Olson <bql at bolson.org> a écrit :
>  
>
>>If I understand this, it's a Condorcet cycle resolution system based  
>>purely on who was 1st pick on each ballot. So, count up the virtual  
>>round robin matrix, and count 1st place votes separately for later if  
>>needed.
>>
>>It's incomplete. A Condorcet method can elect someone no one put in  
>>for first place. It can even have a cycle of people no one voted for  
>>first place.
>>    
>>
>
>It counts first preferences of candidates defeating the potential
>winner pairwise. A CW will always have a score of zero under this
>method.
>
>However, you are right about there being a little cause for concern:
>
>34 A>E>F
>33 B>E>F
>33 C>E>F
>
>I believe this is an E-F tie. E's win over F is worthless due to E's
>lack of first preferences.
>
>Kevin Venzke
>  
>
On  the pure method Forest suggested,  I make this an A-E-F tie (all 
with a score of zero).

But what is wrong with my suggestion of first dropping from the ballots 
the non-members of  the
Schwartz set?

Chris Benham



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20070104/2f6fdea0/attachment-0003.htm>