A What-If: Condorcet in 1996

[Steve Eppley]

Donald wrote:
-snip-
>It is not clear if the author is another supporter of Condorcet -
>but he thinks that Ross Perot would have won the Presidential
>election if Condorcet would have been the election method used. 
>I'm betting on the local pairwise guys to manipulate the election 
>so that their hero, Nader, will be the winner. 
-snip-

I think Perot's cachet in 1996 was considerably diminished from 1992 
and he wasn't taken seriously by most voters this year.

Perhaps Perot could have won in 1992, if a lot of Clinton supporters
preferred Perot more than Bush and if a lot of Bush supporters
preferred Perot more than Clinton.  This isn't totally implausible.  
Here's an approximation to illustrate:
   37% Bush > Perot > Clinton
   10% Perot > Bush > Clinton
   10% Perot > Clinton > Bush
   43% Clinton > Perot > Bush
   (Perot beats both Bush and Clinton pairwise.)
Of course, for a significant fraction of voters Perot would have been
ranked after Bush and Clinton, and the speculation is whether that 
fraction would have been large enough to tilt the election away from 
Perot.  (According to Perot, exit polls in 1992 indicated Perot was 
victimized by lesser of evils voting and would have won head to head 
against either Bush or Clinton.  But I don't know how credible those 
exit polls were.)

It's certainly unrealistic to think there wouldn't have been other
viable candidates running as well, if a good voting method had been
used in 1992 or 1996.

I don't think Nader (the $5,000 man) could have beaten Clinton head
to head in 1996 but maybe someone else could have.  Remember, because
of the existing voting method's spoiler dilemma (and other barriers
like money) potential Democrat rivals of Clinton, such as Dick
Gephardt and Al Gore, chose not to compete against Clinton in the
primaries or general election. 

How much did you lose on that bet?  :-)

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)