[EM] MCA cut off points arbitrary?

[Forest Simmons]

On Wed, 26 Feb 2003, Gervase Lam wrote:

> Nevertheless, could MCA be tweaked a very tiny amount to get closer to
> the better fairness that Cardinal Ratings can give?  May be this could be
> done by having a different "Preferred" cut-off point.  Using an example, I
> suggested 2/3 of the votes instead of 1/2 of the votes.

The higher the cutoff, the closer MCA comes to satisfying the strong FBC.

When the cutoff is one hundred percent, the MCA finally satisfies the
strong FBC, but then the difference between preferred and approved is only
expressive, as opposed to instrumental; in other words, the chance is
negligible that anybody will ever be voted in unanimously in a public
election.

So the choice of cutoff depends on the importance of the strong FBC.

The MCA was designed with the strong FBC in mind; IRV supporters seem to
believe that IRV is closer to the strong FBC than Approval is.  Instead of
arguing with them we can offer MCA with a cutoff that will satisfy them.

Forest

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