[EM] IRV vs Plurality

[Abd ul-Rahman Lomax]

At 03:01 AM 1/17/2010, robert bristow-johnson wrote:

>On Jan 17, 2010, at 12:53 AM, Abd ul-Rahman Lomax wrote:
>>
>>There is a common error here, which is to assume that Range
>>"requires" too much information from the voter.
>
>well, it does force the voter to consider the questions "oh, i hate
>this guy 28% more than i hate the other guy, so how do i rate each
>candidate in range?"  the range rating values are a superset of the
>adjacent integer rankings from a ranked-order ballot like one for
>Condorcet, IRV, Borda.

Part of the problem is the way in which Range has been presented. It 
isn't really "rating" candidates, though that can be part of the 
process. It's *voting.*

The simplest way to describe Range, and to think about it, is that it 
is Approval Voting with fractional votes allowed. Not required.

There is a whole debate among students of Range about using average 
vote rather than sum of votes. The difference is that with sum of 
votes, we have a traditional Approval voting system, which always 
uses sum of votes. Not average vote. (Average vote is meaningless, 
really, unless the ballot asks for Yes or No or Approve/Disapprove 
for each candidate). Average vote doesn't consider majority at all. 
Naturally, I support sum of votes, and though average vote is 
interesting (in terms of understanding the future of a candidate), it 
isn't *voting*, which in it's basic form, is seeking for a majority 
of voters to vote for a candidate for the candidate to win. Is voting 
1/100 vote for a candidate voting "for" a candidate?

I would try to make ballot instructions make it clear that the voter 
is casting fractional votes, and probably shouldn't vote for a 
candidate at all if the voter isn't willing to support the candidate 
against others. That makes the decision much easier.

>   in the ranked-order ballot, all the voter has
>to decide is who she would vote for in adjacent candidates: A>B>C.
>she doesn't have to decide how much more she likes B over C than how
>much A is over B.  one is a quick set of qualitative decisions.  the
>other makes it a quantitative issue, and that's when a lot of us get
>out our dartboard.

Sure. But you don't have to make those quantitative decisions if you 
don't want to. It's optional, and, in fact, I prefer that voters not 
cast fractional votes unless they are easy for them to decide.

>   i don't think making threshold decision based on
>the precise sum of a bunch of noisy numbers (which is what Range is
>when we use our dartboards to score a candidate) does much other than
>to add the means of the noisy numbers and a sum of zero-mean random
>numbers which throws a little bit of dice into the mix before using
>the threshold comparison and determining the winner.

The numbers can be noisy, but surely you know that adding certain 
kinds of noise can improve the accuracy of a feedback system! They 
aren't actually noise, they are "noisy." The averages provide 
information, and the very fact of the existence of fractional votes 
-- even just one! -- improves the utility of the system, that's been shown.


>so it requires thinking that we wouldn't have to do otherwise.  if we
>don't feel like thinking that seriously, it becomes a big noisy
>threshold on the means of stable ranks.  that's sorta like Borda and
>does become the equivalent if people's evaluations of candidates
>sorta "linear".
>
>>First of all, Approval is Range, simply the most basic Range method.
>
>it's Range with 1-bit binary values.

Yup. Range 1, I call it, which means that there are two possible 
votes, generally with one being the default. Approval voting is 
Plurality voting, with the *option* of voting for more than one. Most 
voters, under normal conditions in the U.S., don't need to do it!

>>So what you have is a contradiction: "Range" requires both too much
>>and too little information. Surely it depends on the specific Range
>>implementation.
>
>yes it does.   of course the answer is (if i may appeal to an audio
>image) that what we *normally* mean when we say "Range" is were the
>sliders for each candidate are either continuous or have many
>discrete values (say 10 or 100).

Only Smith considers a continuous "slider," and I prefer, simply, to 
consider that what separates Range from Approval is the ability to 
cast fractional votes. Freedom from the voter. Sure: if you have 
freedom you have more choices and, gosh, you might even be tempted to *think*!

Tell me, do you want voters to think or do you think of voting 
systems as a device that extracts information from voters without 
them thinking about it? And making actual decisions?

>a two-position slider is what we call a "switch". requires one bit of
>information.  that's getting qualitatively different.  either you are
>at the minimum number of levels (or bits of information in the slider
>position) or you're not.

And all the switches are off by default. So, don't want to do much work?

First option: don't vote at all! Leave it to others who know more and 
care more. And this is actually a rational strategy, it works. 
Usually. It's the exceptions that are killers. But all the Get Out 
The Vote activists pretty much are in denial about the usual 
situation. Turnout is not an unmitigated good. It can be a problem, 
it depends. And it should be up to the voters, voting should never be 
mandatory, as it is in Australia.

>perhaps a 3-position slider can be "Actively Disapprove", "no opinion
>- neutral", and "Actively Approve"

Range 2 with midrange default.

>perhaps a 4-position slider can be "Actively Disapprove", "no opinion
>- neutral", and "Actively Approve" and "Hey, I really like this guy!"
>
>perhaps a 5-position slider can be "This guy is crap", "Actively
>Disapprove", "no opinion - neutral", and "Actively Approve" and "Hey,
>I really like this guy!"
>
>we can continue on like this with more discrete levels and all we'll
>get are gradations of the above.  it's all a matter of degree.
>
>but the 2-position slider is a 1-bit piece of information:
>"No","Yes", that's the minimum a voter has to judge.  that's
>qualitatively different.  here's why:  with the multi-level (3 or
>more), then order has to be considered with candidates that you
>approve or disapprove.

This form is plurality with a tweak. Or, really, if we were to think 
about what's natural, it's vote-for-one-only that is the tweak. A bad 
one. Why would you want to throw out the vote of someone who votes in 
a way that says, "A or B, either one is okay with me, given the 
circumstances. Thanks for asking!"

>but the multi-level or continuous slider (3+) requires *more* than
>just ordering information (who is preferred to whom?), it requires
>*spacing* information.

You are assuming no default, i.e., average explicit range. Bad Idea.

And you are confusing a freedom with a requirement.

don't vote at all.
cast one vote
cast one or more votes
cast fractional votes.

It is really that simple, and each of these options can be rational. 
And, in fact, easy.

>   like "i hate candidate D worse than i hate C
>whom i dislike more than B whom i like less than A."  you have to
>decide that D is twice as badder than C than C is badder than B or
>some other value judgement.  what if you just don't feel like making
>such a precise judgement?  then you get your dartboard.

Then if you don't feel like it, don't do it!

Approval style voting in Range is perfectly reasonable, and, in fact, 
gives the voter maximum power to affect the result, if.

If the voter understands the frontrunners and makes sure to vote for 
or against them. And if there are three, gosh. The voter has a 
problem that might actually be difficult. If you want to protect 
voters from making difficult decisions, set up a two-party system, or 
a hierarchical system where the voter only makes binary decisions in 
a series of set-wise choices.

Think it *all* the way through, Robert. We are here to help you do that.