The Allure of IRV

[Adam Tarr]

Mike wrote:

>Yes, because IRV has been pushed everywhere as the rank-count,
>there's little hope of getting any other rank-count accepted,
>except maybe in a city where the IRV promoters haven't been at work.
[...]
>So IRV should be easy to defeat, and, instead of getting into the
>debate of one rank-count vs another, Approval is the proposal that
>should be offered as a replacement for IRV.

I disagree with these sentiments.  We have to recognize that there is a 
very strong intuitive allure to being able to rank your favorite in first 
place, and rank other candidates after that.  Even though Approval is 
strongly preferable to IRV for a variety of tactical reasons, it is very 
hard to shake people of the conviction that honest ranking in IRV will do 
the job.

This is where Condorcet comes in.  If someone is already sold on rank 
balloting, then you say to them, "If you like rank balloting, you should 
really go with Condorcet voting in stead."  You can then provide a simple 
left/right/middle example that shows how middle frequently loses in IRV, 
even though middle would beat either candidate head to head and wins the 
Condorcet election.

Approval, on the other hand, makes the most sense to those who have only 
been exposed to lone mark plurality.  It is an obvious improvement once you 
get a decent amount of information about it, and of course it costs nothing 
to implement.  I would push for approval voting in a place which has no 
exposure to ranked balloting.

So I think Approval is the best method to push in places where IRV has not 
reached, but Condorcet is the best method to push in places that are 
already moving toward ranked ballots.  So, in a sense, we are of opposite 
minds on this.  It's really a question of psychology, marketing, and tactics.

-Adam

----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em