[EM] The worst about each system; Approval Preferential

[Abd ul-Rahman Lomax]

At 09:33 PM 5/26/2010, Kathy Dopp wrote:
>Abd ul,
>
>I agree with virtually everything you say here.

Thanks.

>  However, I would also
>consider that an excellent system for electing one winner would be
>"approval, every voter votes for up to two candidates, followed by a
>runoff of the top two vote getters".  It solves some of the problems
>of a simple runoff election, avoids the spoiler effect I think, and is
>very fair.  Although it does seem to always require a runoff election.


Would you agree that if only one candidate gets a majority in this 
approval election, and it is a form of approval, a runoff is 
unnecessary? A runoff might only be needed if (1) there is no 
majority, or (2) there are two majorities, which is more iffy.

There is already an excellent system that is not terribly different 
from this, and the only difference between how it was used is that it 
wasn't coupled with a runoff for majority failure. And I'm suggesting 
a couple of tweaks. I'd be thrilled just to see original Bucklin 
restored, it worked, and it's a much better system than the 
propaganda has claimed.

Perfect, almost, for a primary election in a runoff system, because 
it's much better at finding *real* majorities than IRV.