[EM] Re: equal rankings IRV

[Bart Ingles]

Kevin Venzke wrote:
>
> The simplest FBC failure scenario I can think of looks like this:
> 
> 6 A
> 2 B
> 2 C=B  <<< (sincere is C>B)
> 3 C>B
> 
> A and B are the lesser evils.

Note: I re-ordered the lines above to more clearly show B as the
apparent centrist, between A and B.  I didn't alter the contents of the
lines.

I don't think this is a particularly obscure example.  Suppose A is the
incumbent, with B and C members of a split challenging party.  This
scenario seems all too common.  The only thing missing are the A-leaning
swing voters, who lean B>>A>C.


> B is eliminated and A beats C, when the two voters vote C=B.  However, this
> only happens because the 3 voters buried their support for B.  If the lesser
> evil candidates are always promoted to (equal) first place, on everyone's
> ballots, then I don't see how there would be any FBC failures.
>
> Reasoning: If lesser evils never have lower preferences, only first preferences,
> then A is either going to have more votes than B or he isn't.  Eliminating
> other candidates isn't going to change that.

Right-- if they vote as though it were straight approval voting.  I
agree.  This is why consider approval voting a better system than
ER-IRV(whole).


> I think for the C=B voters to vote B>C, they would have to believe that C
> can't beat A, and that despite this, more voters will vote C>B than B.

Well, it's true, isn't it?  So all they need is an accurate poll.

It's possible that of the two sincere C>B groups, the B=C voters have a
weaker preference between the two, and/or a belief that B is better
positioned to defeat A.  I can't think of another reason to
differentiate them from the C>B voters.

It might be reasonable to assume that if the B=C voters didn't have the
ER option, some would strategically rank B first and some would
sincerely rank C first.  If half-and-half, ER-IRV is neither better nor
worse than plain IRV.

On the other hand, the fact that they voted B=C might indicate that they
were inclined to vote strategically anyway.  It's clear that these two
voters would have been better off voting B>C, so maybe they would have
voted that way.

With approval voting, it might be reasonable to assume that some of the
C>B>A voters would have voted C=B (with the remainder bullet voting for
C).  If two of these voted C=B in the example above, the outcome would
have changed.

Regardless of the outcome, the existing B=C voters would not have been
committing a strategy error under approval voting (hence FBC).


> I think the voters risking the election are those who DON'T use approval
> strategy.

I don't think that's a valid generalization.  The statement is true
enough for the more extreme voters, who voted C>B, but in this case the
voters who DID use approval strategy were the ones who blew the
election.  Note that the C>B voters were ranking sincerely, while the
B=C voters were apparently attempting to use strategy (they just didn't
use a strong enough strategy).


> --- Bart Ingles <bartman at netgate.net> a écrit : >
> > What would convince me otherwise would be a set of strategy equations
> > comparable to those used for calculating optimal strategy in approval
> > voting, or possibly simulations with ER-IRV(whole) showing that
> > "lesser-evil or better" strategy is as good or better than
> > "lesser-evil-only" in terms of social utility efficiency.

Another thought-- has anyone calculated the best zero-info strategy
under ER-IRV(whole)?  Is it to rank sincerely, use approval strategy, or
something in-between?


> Well, I can't produce these, at least not at the moment...  But my thoughts
> would be:
> 
> "Lesser-evil-only" will only elect lesser evils.
> "Lesser-evil or better" could elect a more broadly appealing candidate, just
> as in Approval.

The problem is that the C>B>A voters mistakenly believed that ranking B
second was enough.  This was a kind of a "sucker bet", much like giving
a partial rating under Cardinal Ratings.  Then again, the C=B>A voters
were also commiting a "sucker bet".  It's not that they were wrong-- you
can be wrong in approval voting too-- it's that they were'nt maximizing
their chances.


> Again, if people are using "lesser-evil or better" strategy, I don't know
> where the FBC failures come in.

You did a good job of illustrating one above; I think where we differ is
the likelihood of it happening.

We haven't touched on nomination strategy, but if the B and C were
members of the same party, it would be in the interest of the party to
make sure that C was kept off the ballot.  Hence the Duverger effect.

Bart Ingles