[EM] compromising strategy in Condorcet...

[James Green-Armytage]

Jobst Heitzig <heitzig-j at web.de> writes:
>Since the above
>situation was exactly the reason for me to propose non-deterministic
>methods which will elect either of A,B,C with a positive probability.
>Then no majority has an incentive for compromising, at least not when we
>assume that noone will vote strategically if s/he thereby risks to get a
>worse outcome than before...

	That's not a valid assumption. Voters might be said to weigh the
probability of the worse outcome and the worse-ness of the worse outcome,
against the probability of the better outcome and the better-ness of the
better outcome. 
	For example, if the voter considers the worse outcome to be not much
worse than the status quo, and the voter considers the better outcome to
be vastly better, and the probability of both outcomes increases equally
given a certain strategy, then we are not on safe ground in assuming that
the voter will not use that strategy.

my best,
James