[EM] voter strat & 2-party domination under Condorcet voting

[RLSuter at aol.com]

In a message dated 8/13/05, Warren Smith writes:

> Given that this is the case, we now can take it to be 100% certain that 
> Condorcet voting methods will lead to 2-party domination, just like the 
> flawed plurality system those methods were supposed to "fix", and just
> like experiemntlly is true with IRV.

I just can't buy this. It's not even 100% certain that plurality voting
methods will lead to 2-party domination. The reason is simple though
unfortunately not widely known. It is that just as voters can vote
strategically, candidates can campaign strategically (or at least
much more so than they have in the past). The fact that very few
have done so until now, especially in prominent races like the U.S.
presidential race, doesn't mean strategic campaigning won't
become common in the future if plurality voting continues to
be the voting method.

As an example of strategic campaigning, Ralph Nader could have
used a strategy in either 2000 or 2004 involving campaigning
strongly up to and through the fall TV debates but promising to
withdraw after the debates if polls had shown that he had no
chance of winning. An even better strategy would have been to
promise a negotiated withdrawal involving throwing his support
after the debates to whichever frontronning candidate offered
the best incentives. Moreover, he could have begun negotiations
with both candidates well before the debates and publicized
their progress, thereby giving his supporters and other voters
additional information about what kinds of compromises the
frontrunning candidates would have considered in order to
increase their chances of getting Nader's post-debate
endorsement. And if Nader had demonstrated great skill
as a negotiator, that could have greatly boosted his standing
with voters in a very short time, possibly making him a (if
not the) frontrunner.

One problem with your mathematical calculations is that they
are based on the assumption that the "frontrunningness" of
candidates will be more or less stable throughout a campaign
or at least during a campaign's final weeks, but that is actually
a very uncertain assumption. (In the 1998 Minnesota governor's
race, for example, the ultimate winner, Jesse Ventura, was
polling well behind the two major party candidates at10% only
a few weeks before the election.) And it would become far
more uncertain if any third party or independent candidate,
especially one as well known as Nader or Ross Perot in
1992 and 1996 or John Anderson in 1980 had campaigned
strategically. With widespread use of strategic campaigning,
2-party domination could conceivably be ended or greatly
diminished even with plurality voting.

-Ralph Suter