kevin_standlee | Range Voting

Current Mood: amused

Entry tags:

Range Voting

The folks advocating Range Voting contacted WSFS (actually, the WSFS webmaster,

sfrose) lobbying WSFS to change its voting system from the Instant Runoff Voting system we currently use for site selection and the Hugo Awards. Sharon told them how our rules work and suggested that if they want to change them, they come to WSFS business meetings and propose and debate the changes there, like all other rule changes. The advocate's response, in my opinion, amounted to, "Our proposal is so obviously Right that we shouldn't have to do all that hard, expensive work. You should change your rules because we tell you to do so."

I often tell people who come to me with rules-change proposals, "If you think it's worthwhile, come and submit it yourself. I'll help you with all of the technicalities to the best of my ability, but you have to make your own case, lobby people yourself, and get the votes by convincing people." Most of the time, this discourages them -- democracy is hard work! But sometimes we get people who are willing to work and debate, and sometimes we even get workable changes and improvements.

WSFS rules are intentionally designed to be resistant to change; however, they can be changed if people work hard enough at it. But it's not enough to just lobby a Board of Directors or subvert the Chairman; you have to convince the members.

Threaded | Top-Level Comments Only

sethb has a Yale math/computer science Ph.D, he also holds a BA and MA in math from Harvard... BFD about the Princeton math Ph.D. One of my bosses in the Air Force was a Princeton Ph.D. in EE, who used to disappear at lunch hour to teach at Northrup University where he was the department head (and yes, he was also simultaneously an active duty lieutenant colonel in the USAF at LA Air Force Station).

The technical term for the manner in which Hugo voting is counted used to be "alternative vote". But this wasn't a very clear term, and fans often confused it with "single transferable vote," which is actually an entirely different way of counting preferential ballots.

So I was glad when the term "instant runoff voting" arrived, because it was a lot clearer and caused less confusion. It is now the term of choice to describe this system.

Have you ever done any actual market research, including interviewing people and discussing that semantics make a huge difference, that while Person A might say, "we don't do any of that here," Person B who works ten feet away from Person A, is actually working on what Person A hasn't got a clue is being worked on in the company... it's not that what Person A is working on is any secret, it's that Person B is narrowly focused on Person B's specialty and doesn't notice anything outside that specialty or have a "systems approach." Most people don't. Or, after fifteen minutes of discussion, it comes out that Person C who said, "No, I don't work on any of that," is working on it, but doesn't think of what he or she is working on, as being of that ilk... that's the reason why e.g. the same equation gets called different names in different fields--it's the same math equation, but most people don't go cross-discipline research to notice that one can e.g. copy a solution to a problem in fluid mechanics from an electromagnetism text. Electrical engineers and programmers in particular tend to draw boxes or the code equivalent and jump down into them and never look outside the box... there was the day I was looking at a design and said to the EEs, "if these two boxes are supposed to be talking to one another there's a problem. One of them has four lines and the other five."

"Which do you want, four or five?" they asked me. "I don't care if it's four or five, I want the same number of lines on each [so that the I/O matches up]."

How about a vote on how people feel about the term "sci-fi" and being called a "sci-fi buff"

I detest the term...

Obviously you're "Missing the point" of my sarcastic comment.

We figured that would fit right in line with the goals of a bunch of sci-fi buffs.

Congratulations. You happened to hit a shibboleth of SF fandom that shows how little you know the audience to which you were trying to sell yourself. You just stepped on a semantic land mine.

That's an argument for Condorcet. Condorcet solves that problem without changing the voting method. Range voting introduces vastly uncontrollable new variables.

That chart is really baffling. What do those percentages mean? It's surely not the percentage of voters who get their preferred candidate. Is it the opinion poll results for satisfaction with the voting method? That doesn't look right either.

This is appearing more and more like an a carpetbagging ideologuistic close relative of astroturfing.

There are costs associated with administering preference polls, running elections, etc. The cost of even looking at the range stuff is time and effort.

Years ago I was involved with neural network and expert systems systems development and implementation, and vaguely recall some of the math / logic behind them involved in decision-making. Neural net systems had to be "trained" with various cases of different conditions. The training took quite a while, but then when conditions were input to the neural net system, the system analyzed in the input and generated responses based on similarity to the cases it had been trained with.

The range stuff sort of reminds me of that--sort of. One thing I notice is that there appears to be -no- calibration involved, no normalization, and that tends to create garbage for both an analysis and results of analysis....

Who are your voters who claim to be "enormously happier" with range voting? Where is the independent opinion poll that states this?

"I think [IRV] is MORE complex than RV."

Good lord.

IRV: 1) pick your favorite.

2) Now, imagine that it's vanished from the ballot (which is exactly what will happen in the counting if we get this far). Pick your favorite from the remaining.

3) Repeat.

Range voting: Decide how much you like each candidate vis a vis all the other ones. Niggle over exact numbers of points. Worry about the balancing effect of other people's votes. (This doesn't come up in IRV, since second preferences have no effect unless your first preference has been eliminated, so there's no need for strategic voting.)

Interestingly, IRV has led to two-party duopoly in EVERY country where it has been used on any scale: Ireland, Australia

That is not true. Don't you know anything about Irish or Australian politics, or do you just look at a list of the party affiliations of the prime ministers?

I'll attempt to respond to a few of the ridiculous arguments in Clay's post, because I've been a Hugo administrator and am therefore a masochist.

every time you tried to tell him you wanted sandwich C, you got sandwich B. ... THIS is EXACTLY what you are doing by using IRV instead of Range Voting.

Oh, for God's sake. "Every time"? This is actually a distorted version of a rare paradox that forms a tiny technical argument against IRV but is not therefore an argument for range voting. And the bad-with-English storekeeper analogy is so inane and a bad analogy that it's hardly worth commenting on.

Have you never seen how IRV works? You total all the votes, over and over again, in rounds,

Have you never seen how IRV works? We have a computer program that, once you get all the raw data in it, counts the entire Hugo ballot and spews out the whole result, first second third &c &c places, in the complete iterations, plus checking for a special rule we have for balancing winner preferences against No Award preferences - everything you could possibly want to know - in less than 20 seconds. I've used it personally. It's such a good program that Hugo voting results are usually published in the form of its immediate output. This is not in the slightest way more difficult than counting up range voting points.

You tell me ANY way in which you think IRV is better than Range Voting, and I'll prove you wrong.

Since you elsewhere define things like 1-5 star movie ratings as a form of range voting (which I think is wrong, because in those you're not trying to advocate a winner), then I can tell you this: I've voted both ways, and I personally find casting range votes much harder than casting IRV with a limited number of candidates. Much harder.

Now, you can talk about voters prefering range voting, though I want to see your proof that they do. Has Gallup done a survey, and how did they count it? And I'm only one voter (though the other Hugo voters who've responded here sound as if they'd say the same thing). But you said you'd dispute "ANY way" in which IRV claims to be superior. And one voter's preference for it is a claimed superiority.

So come on. Tell me that I'm mistaking my own personal preference.

Yeah, there are a lot of smart people out there. If Seth has the credentials you say he does, he should be able to analyze our claims and find flaws in them if those flaws exist. If he can find any methodological errors in our utility efficiency calculations, for example, is is welcome to point them out, or even work on our OPEN SOURCE utility efficiency calculation program, which presents the results in terms of Bayesian regret I should add.

Maybe he could solve some of our unsolved puzzles http://www.rangevoting.org/PuzzlePage.html

Well, no, it's an argument _against_ IRV. Both Condorcet and Range Voting solve problems like this, so you can't just say that it's an argument "for Condorcet". And Condorcet cyclic ambiguities show that Condorcet is deeply flawed. For instance, consider this scenario:

35 A > B > C
33 B > C > A
32 C > A > B

Who wins this election? Perhaps your Condorcet method of choice picks A. Now, if B drops out of the race, but not a single voter's preferences change, C beats A by a huge 65-35 margin. Hence the utter broken-ness of the "majoritarian" idea. We propose something better, the _utilitarian_ idea. The most good for the most people. And if you know even the most basic economics, you want the highest expected value. Range Voting gives that to you. Condorcet, with such paradoxical behaviors, often doesn't. And under strategy, it falls apart, whereas Range Voting degrades, under total strategy, approximately to Condorcet. That is, Range Voting at its WORST is about the same as Condorcet at its best.

See the DH3 pathology: http://www.rangevoting.org/DH3.html

Why Range Voting is better than Condorcet:
http://www.rangevoting.org/CondorcetExec.html

Range Voting kind of IS a Condorcet method
http://rangevoting.org/AppCW.html

I began my trip into the election methods debate as a Condorcet enthusiast. It didn't take long before I saw the light. I encourage you to study the facts objectively, and make the same move.

Clay

Why should he bother? Are you going to pay him more than his day job?

The discussion on this has gone far and wide, and I see that it's been read by a bunch of people who are not on LJ, some who can't read it (work restrictions) and have made do with messages forwarded to them, some who can read but can't post for various reasons.* From among this last group I got some good comments, which I'll partially quote here:

...A voting system can be as brilliant as you want, but if people don't understand how the result is reached then they won't believe it is fair. As far as most people are concerned, the definition of a "fair" voting system is "one in which my favorite candidate wins". If their candidate doesn't win, and they can't understand the voting system, then they will claim that the voting system is unfair. (As you noted, this happens when the first round leader doesn't win in the Hugos.) [...]

...The issue with selecting the "best" voting system is not selecting the one which results in the best reflection of people's desires, but one that achieves the best balance between that aim and people having confidence in the voting system. If a voting system is sufficiently hard to understand then people will not have confidence in it and will keep assuming that the result is flawed.

This all makes sense to me. Indeed, it partially explains why any system other than "first past the post" is a hard sell -- if the people using the system don't have sufficient confidence in it, the system is a failure. To have sufficient confidence, they have to be able to understand it. It doesn't matter if you're mathematically perfect; if you can't explain it in terms that Joe Six-Pack understands, you're doomed.

____________
*I preemptively rule that a discussion on why people can or cannot read or post messages here is out of order. Take my word for it. I'll delete comments on the subject of ability-to-read-or-post. If you want to discuss it with me, write to me directly or go start your own topic in your own LJ.

This is based on social utility efficiency calculations.

See: http://RangeVoting.org/BayRegDum.html
and http://RangeVoting.org/vsr.html

You can't use real world polls to calculate this, because you can't read people's minds, and you can't run hundreds of millions of elections and poll every single voter anyway. Instead what social utility calculations do is take scenarios of the utilities candidates have for a range of choices, and then determine how voters would vote based on those levels of preference for the options. We then look at the average utility produced by the selection of the winner, scaled as a ratio of "distance" covered from the utility produced by a random winner, to the utility produced by the ideal winner. You do millions of simulations, using hundreds of different models of voter behavior, from strategic, to honest (honest is trivial, strategy is more complicated), and from totally ignorant to totally informed, with 2 candidates, with 3, with 4, with 5, etc. You then see the average satisfaction produced by the various voting systems. Range Voting dominates the other methods, in every combination of 5 fundamental parameters we tried.

You can run our software for yourself, and try to find bugs if you like.

http://RangeVoting.org/IEVS/

Also, we often get asked "Why are you using a computer simulation instead of real humans?" So, http://rangevoting.org/WhyNoHumans.html

CLAY

debgeisler said, the burden of proof is on the zealots.

I've been the target of that sentiment myself, and I haven't always won my arguments. Remember that there are probably large swaths of the WSFS Constitution that would not bear up well under the examination they would receive as new proposals.

I suggest that the social utility of RV is actually very low because the counting process is too opaque for most of society to have any confidence in its accuracy or fairness. Your models appear to assume that everyone understands what's going on and has confidence in the system.

Any system in which the electorate does not have a high degree of confidence is doomed because the results will not be seen as a legitimate reflection of the common will. (Just look at American politics; the election mechanics are simple, but voter confidence is low.)

An argument against one system is meaningless unless it's an argument for a better one. (Remember Churchill saying that "Democracy is the worst form of government, except for all the others"?)

You acknowledge that both Condorcet and range voting solve the particular problem you described earlier. Therefore your criticism of IRV is an argument for Condorcet for the reason I stated: that it solves the problem without introducing new variables which might create new problems. And the uncontrolled point system of range voting emphatically does.

Some of the people involved writing various parts of the WSFS Constitution aren't around anymore to ask why they're written the way they're writter. Other people are still around. The WSFS Constitution, though, started off as a document for a gathering a lot smaller and less complicated than modern Worldcons, with a more homogeneous population. Also, the speed and accuracy of contemporary telecommunications, of sending electronic identical copies around to thousands of people in a short timeframe, didn't exist until relatively recently. Compare e.g. the current incarnation of Ansible to something run off a hektograph (spelling) and send around getting chewed up the the mail delivery system and received in Australia months later by slow ship...

And then there were those lyrics by Barry Gold (I think they were his).

"The song of fandom's a sad song...
"No cons... for the next ten weeks!"...

These days there can be more than six just on the same weekend, forget about gaps of months!

Ah. So your argument of voter preference for this system is entirely specious. Thank you, that tells me all I need to know.

It's not bugs in your software, it's the assumptions made when writing the software and creating the data. Social utility/cost benefit/"economic man" arguments are always liable to drift away from reality, unless cross-checked against real-life preferences. (See the staggeringly inane "Everyday Economist" columns on Slate magazine for an example.)

If range voting is kind of a Condorcet method - and I can see that - then I definitely prefer Condorcet.

By the way: you write,

I began my trip into the election methods debate as a Condorcet enthusiast. It didn't take long before I saw the light.

Do you have any idea how much that makes you sound like a Scientologist? We in science fiction fandom have good and long reason to be wary of that kind of starry-eyed insistence that perfection has been found, and that accounts for a lot of the immediate bristling your posts have been generating. (Have you ever heard of a fellow named Claude Degler?)

So we have the same argument coming both from surveying methodology and neural net training, just using different language.

I could see a "-2/+2" scale dealing with the calibration issue well. It wouldn't be too difficult to say "-2=hate, -1=dislike, 0=neutral, +1=like, +2=love" but that doesn't mean that I think it's worth replacing Hugo and site-selection voting scheme in the first place.

Threaded | Top-Level Comments Only

Page 4 of 8

Range Voting

Re: Supermajority vote?

no subject

no subject

Re: Range Voting

Re: Missing the point

Re: Range Voting

Re: Supermajority vote?

Re: Supermajority vote?

Re: Missing the point

Re: Erroneously?

no subject

Re: Missing the point

Re: Missing the point

Re: Supermajority vote?

Re: Supermajority vote?

no subject

no subject

Re: Erroneously?

no subject

Re: Erroneously?

Re: Supermajority vote?

no subject

Re: Erroneously?

Re: Supermajority vote?

Re: Missing the point