[EM] STV and weighted positional methods

[Raph Frank]

On Sat, Jan 31, 2009 at 6:47 PM, Kathy Dopp <kathy.dopp at gmail.com> wrote:
> On Sat, Jan 31, 2009 at 5:41 AM, Raph Frank <raphfrk at gmail.com> wrote:
>
>> That isn't the case in plurality.  Lots of people vote for one of the
>> top-2 rather than their favourite.
>
> That is called Strategizing. In plurality it is ALWAYS TRUE that if I
> choose to vote for a candidate in any election, my vote will increase
> that candidate's chances of winning, not decrease it.

That is likely true in IRV too.  Unless you know the exact
probabilities, giving a candidate a higher ranking is more likely to
help than hurt the candidate.

I think the likely situation with IRV is that there will be 2 major
candidates.  Voting for anyone else will have little effect.

It doesn't elect a condorcet winner.

Again what is your view on condorcet methods and approval?

> Clearly IRV/STV never allows me to vote for a candidate first and know
> that it will help that candidate win, whereas in plurality I always
> know that if I vote for a candidate it increases that candidate's
> chance to win.

I am not sure chance to win is the right term here.  If ranking a
candidate first has a 70% of helping and 30% chance of hurting, then
it also increases the candidate's chance to win.

I would disagree that absolute monotonicity is required for a voting
system to be considered fair.  As long as the chances of participation
failure/non-monotonicity is low enough, then that is acceptable.

The fact that plurality is monotonic doesn't outweigh the tendency of
the system to 2 party domination.