[EM] Student government - what voting system to recommend?

[Abd ul-Rahman Lomax]

At 06:37 PM 4/23/2007, Juho wrote:
>Another explanation to the emergence of Approval style strategic
>voting is that an individual voter might learn that, in a case where
>there are only two candidates that have chances of winning the
>election, voting A=9, B=0 instead of A=5, B=4 makes his/her vote 9
>times stronger.

The thinking is really stubborn. A voter "might learn this?" Why 
didn't the voter know this from the start. *Of course* voting the 
extremes is a strong vote. The question is why you'd cast a strong 
vote if your preferences are weak. Why? Because you want to "win"?

First of all, we think that it will be common knowledge that if you 
don't vote the extremes for at least one candidate on either side, 
you are casting a weak vote. Nobody is recommending that truly weak 
votes be cast. (But some people may want to cast them anyway, and 
they should be able to. Consider it a partial abstention, and many 
people abstain from this or that race now.)

If A was the favorite, why in the world would the voter vote A=5 in 
the first place? I'll note that the same 1/9 vote relationship exists 
between A=9 and B=8. If the voter cares strongly that A win over B, 
the voter will *not* vote 9:8.

"Strategic" voting assumes a strong desire to affect the outcome in a 
particular way. What is continually asserted here is that voters with 
weak preferences will somehow decide to vote strategically. Why? To 
please party bosses? They don't care a fig about them!

If voters have strong preferences, the kind that motivate people to 
be "dishonest" in ranked systems, under Range they will vote Approval 
style. Which expresses strong preferences. *It's honest!* What is 
continually asserted, over and over and over again, is that voters 
will turn their weak preferences (9 vs 8 in the original example Juho 
posed) into strong expressions. In order to be motivated to do that, 
they must have strong preferences! It is a total contradiction.

Look, if there is an election, and I sincerely rank A as 9 and B as 8 
(and other candidates lower than that, let's say zero) and B wins, 
I'm happy! That's an excellent outcome! The danger comes in quite the 
opposite direction from what Juho proposes. Suppose I rate B as 8 and 
C wins, with B being the runner-up. Close runner-up. I might regret 
rating B at 8.

Realistically, though, it was only 1/9th of a vote that B lost on 
account of me. Still, if the margin is close enough ....

For this reason, I prefer to have an extra voting position: Favorite. 
With Approval, I call the method A+. It is Approval for the purposes 
of determining the winner, but there is a method of indicating the 
favorite. It could have significant usefulness. This could be done in 
Range as well. In an environment where every vote or fraction of a 
vote counts -- a close election -- the voter may want to give the max 
rating to a second-favorite as well as the favorite. Allowing a plus 
expression makes, we think, this less offensive.

>  Similarly he/she could learn (maybe from experts)
>that in general voting in Approval style (as defined in the well
>known Approval strategies) in elections where there are several
>potential winners typically gives him/her the strongest voting power.

Of course it does. Voters don't need experts to tell them that! But 
the question is whether and how voters will want to exercise their 
power. If you vote Approval style, you fail to express your true 
appreciation of the candidates, and this can backfire. Your Favorite, 
say, loses to your second Favorite, and it turns out that the third 
candidate you were worried about was a big dud, not really close. 
Dag-nab-it! Why didn't I just express my honest feelings!

It is just as reasonable to consider that Range elections will move 
*away* from Approval-style as that they will move toward it. I expect 
that they will start out, actually, as close to Approval for many 
voters. Smith thinks differently, and I really don't know which of us 
is right. He's got reasons to think his way. We might both be right. 
I.e., many voters, maybe most, will vote Approval style, and it will 
be bullet voting. But there will be quite a few, from the start, who 
do something different.

I say that we are not going to really know until we see real 
elections using Range. The alleged devolution to Approval is not a 
serious harm. It would only mean that some ballot space and a 
counting effort had been wasted.

>I used this style of explanation since this explanation does not talk
>about parties, or voters belonging to them, or about the candidate
>set-up, but only about the strength of the vote of the individual voter.
>
> > Further, note that the PW candidate now gets zero from this group.
> > That's really not much different from the vote before. But it is
> > totally unnecessary. Why would these voters suddenly drop their
> > (small) support for the candidate with no chance to win?
>
>Rating the least preferred candidate at 0 reduces the probability of
>that candidate getting elected (and doesn't carry any risks with it).

But from the conditions of the problem, there was no risk of that. 
No, I don't buy it. (By the way, none of us involved with Range would 
recommend giving the "least preferred candidate" any other vote than 
the minimum. I assumed that PW was being given a 1 because voters 
somewhat liked him, there were *worse* candidates involved. If by 
"voting Approval style" Juho means that votes will move to the 
extremes, so that at least one candidate is given max rating and one 
min, we already recommend that. The question, really, is what to do 
with those in between. And what I say is that the answer to this 
depends on how strongly you care. If there are three candidates, A, 
B, and C, in order of your preference, and, let's say, there is no 
risk of C winning, then even a relatively small preference of A over 
may cause you to downrate B to zero. But if C is a contender, then 
you must balance the risk of B losing to A with the risk of C defeating B.

So it really gets interesting when there are more than two serious 
contenders. I'd suggest that in the three candidate election, you 
would think of the various scenarios: *how much* do you care about 
the two remaining pairwise elections? (We already know how you will 
vote for A and for C.) If B winning over a is a minor annoyance and C 
winning over B a major regret, then you would want to rate B closer 
to A. And if you think B is better than C, but not much, and you 
would *hate* for your vote to have elected B over A, then you would 
rate B closer to C.

And if it was six of one and a half-dozen of the other, you would 
rate B midrange. This is actually the optimum vote! (If you feel that way!)

Frankly, I think that most voters, with Range, will not beat their 
heads against the problem. They will simply vote sincerely; the fact 
is, that by the analysis I just made, a simple sincere set of ratings 
(normalized to min and max range) is not only the easiest vote to 
cast, it is the most likely to affect the outcome desirably.

In the two-party environment, Range strategy, like Approval, is 
pretty simple. It's the three-way race which is trickier.

> > If you are going to propose that Range will *reduce* to Approval,
> > you will have to use reasonably likely scenarios.
>
>I think the vote strength argument that I presented above is quite
>generic and applies in all typical elections - assuming that we talk
>about competitive elections where the voter wants to do his/her best
>to make his/her favourite alternative win.

But this contradicts the assumed initial sincere vote! If you want 
this, why would you vote A=9, B=8 in the first place? By voting this 
way, you are saying that B winning is almost as satisfactory to you 
as A winning!

> > And if it *does*, under some difficult-to-anticipate circumstance,
> > reduce to Approval, that isn't a bad outcome!
>
>Approval is not very bad. There are different ways of describing
>Range to the voters. I think a description that advices voters to
>indicate their sincere utility values of the candidates in the ballot
>is not a good description since that makes those voters that vote
>strategically (Approval style) and not as told more powerful than
>those that vote as told. Defining Range as "like Approval but with
>option to give only weaker fractional preferences" would be more fair.

I wouldn't describe Range, on the ballot, using terms like "sincere 
utility values." That is a description of "sincere Range," and 
indications are that if voters vote that way, election outcomes will 
be maximally satisfactory. Yes, this means that some people will see 
their favorites lose to someone who wasn't their favorite, but they 
will also see someone they prefer, over someone they disliked, win. I 
think that people can and will understand that democracy is often 
about making compromises. It is *not* about crushing the opposition!

I would describe Range 11 as "You have a maximum of ten votes to cast 
for each candidate. You may cast no vote for a candidate ("zero"), 
which means you will be contributing nothing to the election of that 
candidate, and you may cast up to 10 votes, 10 votes being the 
maximum contribution you may make to a candidate's election. All the 
votes will be totalled and the candidate with the most votes wins."

(I'm suggesting sum-of-votes Range, which is not the current 
"official" Center for Range Voting proposal. But it looks like the 
basic proposal is going to change to the simpler sum-of-votes method. 
In sum-of-votes, abstentions are given a default value, zero in what 
I described above.)

But I'd actually like to see Range start with Range 3, with the 
ratings being -1, 0, +1. And the instructions would be quite simple:

"Vote -1 to vote against a candidate, vote +1 to vote for the 
candidate, and vote zero or leave a candidate unrated to have an 
intermediate effect. The candidate with the greatest sum of votes will win."

In this system, abstentions are midrange. They pull the vote toward midrange.

But I don't think we really know what form of Range is optimal. We 
need experimentation.

[I suggested that there be a runoff between the Range winner and a 
Condorcet winner, if they differ]
>Let's assume that a Condorcet winner exists. In this case this method
>could be said to be a method where the voters are given a second
>chance to think again if the Range winner could be seen as a "good
>compromise" even though the majority could easily vote as in the
>first round and elect the Condorcet winner.

Yes. That is, the original ballot analysis showed that this C. winner 
was rated higher than the Range winner on a majority of ballots.

>  I'm not sure this method
>would be a very practical method in real life large elections but in
>principle the idea of "recommending" the Range winner to the voters
>is a positive idea. Some strategies where people would try to
>influence who the Range winner will be could take place (i.e. the
>Range winner of the second round would not be the sincere range winner).

I think Juho means that the Range winner of the *first* round would 
not be the sincere Range winner. If there is a second round, it is 
not held as a Range election. It is a straight 
which-of-these-two-shall-be-elected vote. Voters will know, this 
time, if the first election was sincere, which candidate will be most 
broadly acceptable. Which is more important to them, for their 
preference to win or for the most broadly acceptable candidate to 
win? Majority rule.

I'd suggest that if their preference was weak, the majority might 
prefer the Range winner, on reflection. But if their preference was 
strong, they might insist upon it.

My actual preferred election method is deliberative. Standard 
Robert's Rules applied to an ordinary motion to elect so-and-so. With 
a Range poll taken at some point in the process. I may be the first 
person to point out that this standard process, followed with care, 
will always select the Condorcet winner. (And maybe not! -- anyone 
see this before?) But it is a Condorcet winner after a process which 
informs the voters regarding all aspects of the election. Including 
the social utility winner. Do we want to maximize social utility? Yes/No!

The runoff I mentioned above was originally suggested as a safeguard 
against unanticipated election results. In particular, it was one 
response to the possibility of using Average Range instead of 
sum-of-votes. Average Range needs a "quorum rule," CRV calls it, to 
deal with the problem that a relatively unknown candidate could get a 
high average rating, the extreme being, of course, one vote at max 
rating. The debate has been long over this among Range advocates.... 
my position is that Average Range introduces difficult issues that 
could interfere with Range implementation. Sum-of-votes is far 
simpler to completely explain.

I never have actually understood the significance of the present 
proposed Quorum Rule.... it's rather arbitrary. So as a safeguard, a 
runoff. And then the idea became a runoff between the Range winner 
and the Condorcet winner, which, if it was part of the system, would 
pull the rug out from under complaints that Range fails the Majority Criterion.

Range will usually choose the Condorcet winner, Warren's simulations 
show that. But, of course, there are exceptions. We claim, generally, 
that the Range winner is, under these conditions, a *better* winner 
than the Condorcet winner. But I prefer to allow the electorate to 
make that decision!

(In full deliberative process that is exactly what happens with every 
aspect. The original motion is amended or not....)

Impractical in large groups, directly. But quite practical with 
delegable proxy!