[EM] Majority Winners and 3-Level Approval

[Forest Simmons]

When I proposed three level Favorite plus Acceptable Alternatives (aka
Bucklin done right) I was thinking of counting the top level by the rules
of cumulative voting, so that it would be impossible for two candidates to
each get more than fifty percent of the top level votes.  That would give
incentive to keep the top votes to truly top choices (tied for first
place) so that voters wouldn't feel constrained to dilute their top votes
with compromise votes in any way.

That thinking was in an effort to satisfy typical voter psychology ...
wanting to show unique support for favorite (without strategic penality)
when there was only one favorite.

For myself, I prefer Alex's version where two candidates could both
surpass 50 percent approval before having the middle level kick in.  I
just wonder about average voter psychology.

Forest

On Tue, 30 Apr 2002, [iso-8859-1] Alex Small wrote:

> Demorep has argued quite strenuously that majority approval should be
> necessary to avoid sending the decision to a legislative body, holding a
> second election with new candidates, or leaving the office vacant (the
> Libertarian side of me wouldn't mind if nobody is in a position to spend
> tax dollars...).
>
> Also, in the last few days I've developed an infatuation with 3-level
> approval.  Sorry, regular approval, but my new voting method gives me the
> same rush of excitement that I felt when I started studying election
> methods.  Can we still be friends? ;)
>
> In approval voting, if there are 3 candidates with roughly equal-size
> camps, most voters might gamble that the other camps will cast cross-over
> votes.  They might cast only one vote so as not to dilute support for
> favorite.  This leads to a minority winner, outrageous Playboy interviews,
> a stint with the XFL, and soon it's all gone to hell and thousands are
> dying at Gettysburg...
>
> (Sorry, I'm in a whimsical mood...)
>
> 3-level approval should greatly increase the odds of people casting cross-
> over votes, making a majority winner far more likely.  In fact, in a 3-way
> race, if almost everybody casts both a "preferred" vote and an "acceptable"
> vote then a majority is almost guaranteed.
>
> If this method is more amenable to voter psychology, and it prevents
> another XFL then I'm all for it...
>
> Alex
>
> P.S.  Be nice to Jesse.  He strikes fear into the hearts of the
> Republicrats, as well as anybody in his immediate vicinity... ;)
>
> ----
> For more information about this list (subscribe, unsubscribe, FAQ, etc),
> please see http://www.eskimo.com/~robla/em
>
>

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