[EM] favorite betrayal in Condorcet(wv, =permitted, no partial votes)

[Warren Smith]

The original general-purpose 19-voter FBC example from the Center for Range Voting web page
http://math.temple.edu/~wds/crv/IncentToExagg.html:
8:B>C>A
6:C>A>B
5:A>B>C
B wins under Condorcet Voting [Ranked Pairs variant, winning votes,
equality-ranking permitted] according to Eric Gorr's calculator at
http://www.ericgorr.net/condorcet/.  (Incidentally, I think this is also
the same as what has been called "Steve Eppley's MAM method.")

Modification I: In original, 3 of the 6 C>A>B voters switch to A>C>B:
8:B>C>A
3:C>A>B
3:A>C>B
5:A>B>C
New result: AB tie.  This is an improvement from the switched voters' point of view.

Modification II: In original, 3 of the 6 C>A>B voters switch to A=C>B:
8:B>C>A
3:C>A>B
3:A=C>B
5:A>B>C
B wins.  (This is no change.)

CONCLUSION: these 3 voters, by "betraying" their favorite third-party candidate C
so that they could strategically exaggerate the major-party candidates A and B
to "top and bottom", with C "middle", caused the election result to improve.
But if they had only partially betrayed C by voting A=C>B, then that would not
have provided enough power to change the election result.  A full-power betrayal
was necessary.

I conclude from this, that either
(1) Adam Tarr's contention that the "winning votes"
and "permit equalities" enhancement of Condorcet, could avoid this kind of
2-party-domination-inducing problem, was wrong.
or
(2) Eric Gorr's calculator program is wrong.

-Warren D. Smith
[let me know n which of 1,2 you think it is.  
let me know if this is really the same thing as Eppley MAM. Thank you.]