[EM] Real IRV Election, Disputable Result

[Jonathan Lundell]

At 11:28 AM -0500 3/12/06, radio deli wrote:
>I saw your post on the Elections Methods List.  As a Vermont 
>legislator, we may have to decide the issue of IRV on a statewide 
>basis.  To be honest, I'm not very enthusiastic about IRV.  I would 
>prefer to support the candidate (not plural) of my choice, and if a 
>runoff must occur between candidates I didn't support, then make a 
>new decision based on the contest at hand.  

When San Francisco switched from top-two-runoff to IRV, quite a lot 
was made of the cost savings, which did turn out to be significant.

But a more significant benefit to using a single ranked-choice 
election (whether IRV or Condorcet) is participation. SF historically 
has seen a much lower turnout in the runoff election than in the main 
election.

And of course the top-two runoff is even worse than IRV with respect 
to premature exclusion; I'd guess offhand that it'd be much more 
likely than IRV to eliminate the Condorcet winner.

In practice, it's hard to justify a separate runoff election. It's 
expensive, tends to disenfranchise many more voters, and does a 
relatively poor job of finding the "best" choice by nearly any 
measure, especially in a factionalized election.

It comes down to STV (IRV) or one of the Condorcet methods (of which 
the Schulze method is worth a close look).
-- 
/Jonathan Lundell.