Regretted Turnout. Insincere = ranking.

[Mike Ositoff]

On Sat, 13 Jun 1998, Markus Schulze wrote:

> You wrote (13 Jun 1998):
> > If initial conditions change (i.e. by adding more
> > voters with different rankings), then the results
> > can be expected to change.
> 
> If it is not possible to create an election method,
> that never rewards tactical voters, isn't it senseful
> to ask, whether there is at least 1an election method,
> that never punishes sincere voters?
> 
> Markus


Yes, but that probably isn't possible without holding a 2nd
election. Any 1-balloting method, such as Smith//Condorcet or
Schulze, could conceivably make some regret not having used
retaliatory defensive strategy, if a large group of voters succeed
in electing their favorite & defeating the Condorcet winner via 
order-reversal. Well I should add that retaliatory defensive strategy,
isn't the only defensive strategy available. In any pairwise-count
method, ranking the Condorcet winner 1st (if you can guess which
alternative it is) protects it against order-reversal if enough
people do that. So the Condorcet winner then wins.

So SC & Schulze can conceivably make you regret sincere voting,
but some of the 2-balloting methods that we've discussed here
get rid of strategy dilemma, in my opinion--something not possible
with any 1-balloting method I've heard of.

But there are different ways methods can "punish sincere voters",
some more serious than others. The method can allow offensive strategy
to succeed because you didn't use defensive strategy. Or it can make
someone sorry that he didn't use an offensive strategy that was
available to him. That happens when a method violates No Show,
or Positive Participation, for instance, or when SC or Schulze
would have given advantage to someone had he insincerely ranked his
upper choices equal. It seems to me that the offensive strategy 
possibilities resulting from No Show or Positive Participation 
violation is virtually impossible to predict & exploit, and so
it wouldn't be a problem. Aside from that, it doesn't seem so bad
'punishing" someone for not having & using the predictive knowledge
needed to offensive exploit a No Show or Positive Participation
violation.

p.s. I've been offline since my last posting. The computer broke
down & I had to replace it & find out how to set-up the new one.

Mike Ossipoff

 
> 
>