[EM] Random Method [WAS: Is strategic voting a bad thing, really?]

[wclark at xoom.org]

Bill Clark wrote:
> Voters would be encouraged to cast sincere pre-ballots by generating
> sufficient trust in the effectiveness of the strategy agents (that is, if
> voters believe that the strategy agent will do at least as good a job in
> devising a voting strategy as they themselves would, then they have
> little to no reason to falisify their ballots.)

Perhaps this is a better way of thinking about this point:

Consider an election in which the actual election method to be used isn't
known beforehand.  Each voter specifies a ranking of the candidates, and
then once all votes are cast a particular election method is selected at
random from a predetermined list of possible methods.  Perhaps it will be
IRV, perhaps Plurality, perhaps Condorcet, perhaps Borda -- precisely
which won't be known until *after* the votes are cast.

Such a system mitigates the need for strategic voting (since each system
might have a radically different optimal strategy), while doing so in a
relatively non-destructive manner (since there are *many* systems that
will perform quite well given sincere ballots.)

The "strategy agents" system mentioned in my previous post can be thought
of as defining a particularly complicated election method -- one which
changes not entirely at random, but in a chaotic manner based on the
actual relative proportions of the various voting blocs.  The principle is
the same, however: By making it difficult for voters to devise effective
strategies on their own, sincere voting is encouraged.

> It might be the case that a sort of "meta-stragegy" could evolve,

I believe comparing the strategy agent system to the random method system
described above sheds some light on how such meta-strategies might work.

Since the details of the method that ultimately results from the scheming
of the strategy agents might be predicted by knowing beforehand the actual
distribution of voter preferences, an *extremely* knowledgeable (and
clever) bloc of voters might be able to dictate which method (and hence
which strategies) will be used in determining the winner.

Random method would be invulnerable to such meta-strategies (assuming the
selection of method was indeed done at random.)  But such a system would
also be non-deterministic, which might not be acceptable to some voters.

-Bill Clark

-- 
Ralph Nader for US President in 2004
http://votenader.org/