[EM] IRV vs Condorcet again

[robert bristow-johnson]

On Jul 27, 2010, at 12:58 AM, Dave Ketchum wrote:

> IRV allows voters to express their thoughts, but does not read all  
> that the voters say - see Burlington in 2009, where IRV counted as  
> third the candidate human counters saw as first:
>     After disposing of two minor candidates we see 2981 Kiss, 3294  
> Wright, 2554 Montroll.
>     With M losing, add 1332 M>K & 767 M>W and Kiss wins, BUT IRV  
> does not see what Condorcet would see:
>     Add 2043 K>M & 371 K>W and see M liked better than W or
>     Add 1513 W>M & 495 W>K and see M liked better than K,
>     Meaning human counters, and Condorcet, can see Montroll deserving.
>
> Are we in a cycle?  No, for M is CW.
> Near a M>K>W>M cycle?  Change some M>W to W>M.
> Near a M>W>K>M cycle?  Change some K>W & M>K to W>K & K>M.
>
> Condorcet counts all that the voters vote.

preach it, Brother.  (but i'm already in the choir.)


now it's another election season in Burlington and Vermont, and  
another battle.  the GOP have one candidate for guv (they got the 11th  
Commandment).  5 different Dems are gonna beat each other up for the  
nomination next month.  hmmmm, i wonder who's gonna be in better shape  
for November.

but i'm still stinging from Town Meeting Day last March (and the  
bruising "Question 5").

--

r b-j                  rbj at audioimagination.com

"Imagination is more important than knowledge."