[EM] query re random method rank voting

[RLSuter at aol.com]

Has there been any discussion on this list or elsewhere
about a ranked preference system of voting whereby the
method for determining the winner is decided by random
selection from among a group of methods that have been
determined beforehand to be acceptable, with the method
for each election determined in such a way that people
won't know when they vote which one will be used? If
so, what have been the main pro and con arguments for
such a system? Its main purpose would be to make it
difficult or impossible for people to decide how best
to vote strategically and thereby ensure that all or
most people will vote their true preferences.

Thanks,
Ralph Suter