[EM] The "so called" compromise candidate

[Alex Small]

Donald wrote:

>There seems to be a mind set among a few on this list that if there is no
>majority on the first count then the winner should be the third or lower
>candidate, the so called `compromise candidate'.

Not true.  The compromise candidate, roughly defined, is whoever has the
largest coalition of core supporters, swing voters, and cross-over
supporters.  The precise definition depends on the method.  In approval all
such voters supporting a candidate are treated equally.

In Condorcet a voter ranking a candidate second place gives him advantage
over his lower choics but not over his favorite.  If somebody gets enough
of those votes he may be able to defeat all others, but you can't really
say that ranking somebody second helped him defeat your favorite.  It only
helped him defeat other candidates, and the fact that somebody else ranked
the "compromise" second and _your_ favorite third is the reason your
favorite lost.

In 2000 Gore was presumably the compromise candidate among the top 3:
Bush, Nader, and Gore.  His positions were closer to Bush's than Nader's
were, and closer to Nader's than Bush's were.  Note that Gore, the
compromise, was also the plurality winner.

Alex

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