Range Voting

[kevin_standlee]

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.

Comments

[kalimac.livejournal.com]

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.

[thebrokenladder.livejournal.com]

Instant Runoff Voting _IS_ Single Transferrable Vote, applied to single-winner elections.

[kalimac.livejournal.com]

The application of the voting method to single and multiple-winner elections requires quite different set of voting rules. That the IRV rules are a subset of the STV rules does not make them the same, any more than the New Testament being part of the Bible makes them the same.

[thebrokenladder.livejournal.com]

STV isn't a "subset" of STV. STV can be used to select X winners out of a group of Y. If Y is 100, and X is 5, you call it STV. If Y is 100 and X is 2, you call it STV. If Y is 100 and X is 1, you called it "IRV" -- it's still STV.

The same is true of Range Voting and Reweighted Range Voting. If you use RRV to elect 1 winner, it's just the same as Range Voting. One could just dispense with the distinction, except that you might pick, say, a school board by holding six regional single-winner RV elections, or you might do it by having ONE multi-winner RRV election. But IRV _IS_ STV, make no mistake about it.

[kalimac.livejournal.com]

STV isn't a "subset" of STV.

I didn't say it was. I said that the IRV rules, not the STV rules, are a subset of the STV rules. They are different methods of counting, one intended for single winners, one for multiple winners, and I refer you to such standard textbooks as Geography of Elections by P.J. Taylor and R.J. Johnston for an explanation of the difference.

You pride yourself on being scientific; now you're being totally irrational.

[thebrokenladder.livejournal.com]

From Wikipedia:

When the single transferable vote (STV) system using the Droop quota is applied to a single-winner election it becomes the same as IRV.

The rules are no different. You use the same rules, but you just use them on a single winner instead of on multiple winners. That's my understanding based on everything I've read. I'm happy to consent to being wrong if you can show me otherwise.

[kalimac.livejournal.com]

I have shown you otherwise. Check the textbook I already referred you to, which treats them as separate systems. I consider standard textbooks a little more reliable for such classifications than wiki-fricking-pedia. You could have put that sentence in Wikipedia yourself.

If STV with only one winner is the same as IRV, so is a list system with only one candidate per list. That doesn't make a list system the proper term to describe IRV.

You do not use the same rules for the two. The act of intending it for single winners instead of multiple winners changes the rules. The Droop quota threshold calculation, for instance, which is the key feature of STV, has no place in IRV whatsoever, where you just skip it and go to simple majority. Neither does the surplus vote allocation, a concept totally alien to anything in IRV.

The original point was that IRV is the term for what Hugo voting uses, STV isn't. If you say you're using STV, you mean you're using the full panoply of STV rules to elect multiple winners. We don't. So it's not STV. Period.

Your knowledge of the systems you're criticizing is as poor as your knowledge of how science fiction conventions work, so there's no point in listening to you on anything.

[thebrokenladder.livejournal.com]

I have shown you otherwise.

On the contrary, I have shown you otherwise. IRV is the concept of applying STV to a single-winner election.

Here's the formula for STV.

http://en.wikipedia.org/wiki/Single_transferable_vote#Counting_the_votes

Now, take that and apply it where the number of seats = 1.

You get IRV.

Check the textbook I already referred you to, which treats them as separate systems.

Well, it's wrong, unless you claim that formula for STV is incorrect. I invite you to submit the corrected formula so that we can correct that Wiki page.

The book might treat them as "effectively" different methods, because for example in the Australian senate and house, the senate uses STV, whereas the house uses IRV. That is, the senate uses the results from a series of multi-winner elections, whereas the house is comprised of the winners of a bunch of single-winner elections.

I consider standard textbooks a little more reliable for such classifications than wiki-fricking-pedia. You could have put that sentence in Wikipedia yourself.

If you read a text book about STV, then feel free to describe STV, and show us how it does not reduce to IRV in a single-winner election. Show us an example.

If STV with only one winner is the same as IRV, so is a list system with only one candidate per list. That doesn't make a list system the proper term to describe IRV.

I don't know what you're trying to say here, but my intuition is that your response here doesn't make sense. ANY voting system is going to pick the same winner in an election with only one candidate. That has nothing to do with how the systems actually operate.

You do not use the same rules for the two. The act of intending it for single winners instead of multiple winners changes the rules.

The number of winners with STV can be set to whatever you like. If you pick 78 winners, it's STV. If you pick 2, it's STV. So changing the number of winners doesn't change whether it's STV.

The Droop quota threshold calculation, for instance, which is the key feature of STV, has no place in IRV whatsoever, where you just skip it and go to simple majority.

The Wiki on "Droop quota" says:

The brackets denote the operation of rounding down. This gives the Droop quota the special property that it is the smallest integral quota which guarantees that the number of candidates able to reach this quota cannot exceed the number of seats. In a single winner election, in which STV becomes the same as Instant Run-off Voting, the Droop quota becomes a simple integral majority quota–that is, it will be equal to an absolute majority of votes.

But I know, you probably think I edited this myself as well. I promise you, I didn't.

Neither does the surplus vote allocation, a concept totally alien to anything in IRV.

That's a fallacy. In a two-winner STV election, for instance, after you pick the first winner, any surplus vote goes to the second winner. After you pick that last winner, you don't do anything with the surplus vote..you just halt the election, because you have no more candidates to select as winners. With IRV, you just do that same halting process, just one candidate sooner. Single-winner STV is to two-winner STV what two-winner STV is to three-winner STV.

The original point was that IRV is the term for what Hugo voting uses, STV isn't.

That's like saying, "ape" is the proper term for what you and I are, "mammal" isn't. We are both mammals and apes, just as single-winner STV is both STV and IRV.

Stop being so stubborn, and accept that the Wiki entries are correct, and you are just misunderstanding what your text book says.

[kalimac.livejournal.com]

The question is not whether the IRV rules are a subset of the STV rules. I already agreed that they were, before you ever brought Wikipedia up. Yes, applying the STV rules to one seat amounts to IRV, I said so specifically, and I never said otherwise. That you apparently think I did shows your poor grasp of the argument.

The question is whether they are classified and terminologically described in political science as the same system. They are not, and I offered proof in the form of a poli sci citation. You offered Wikipedia, and the fact that you could say something like "proven wrong by Wikipedia" shows your ignorance and poor hold on both reality and argumentation. Wikipedia has a lot of useful and accurate information, but it is completely unreliable as proof of anything whatsoever. You'd be thrown out of class for using it as a citation of anything.

So if you "apply" STV to a one-seat election, you get IRV. AND YOU SHOULD CALL IT IRV, NOT STV. That was my original point in this thread. My original comment was not addressed to you, but to a curious bystander who was clearly uninterested in polisci trivia. And I'm sorry the fact that I didn't bother to mention to that person the complex relationship between the STV rules and IRV rules has confused your - again - poor overtaxed mind.

In the name "Single Transferable Vote," the word "transferable" refers to the surplus majority transfer. This has no place in IRV and consequently the name STV is totally inapplicable to IRV, despite the fact that all the rules of IRV do have a place in STV.

I dropped the discussion of the comparative virtues of systems not out of lack of counterarguments, but because you are completely impervious to the significance of the points that make you mistaken. See kevin_standlee's last comments, and see his next post. And indeed, I see you've gone on to claim that I am mistaken in my own personal preferences, so you've gone beyond all previous benchmarks of arrogance.

I thought I could continue to clarify this simple matter of terminological classification, but even there it turns out that your bewilderment is too great to overcome. Sorry.

[thebrokenladder.livejournal.com]

The question is whether they are classified and terminologically described in political science as the same system.

Yes, they are. They ARE THE SAME SYSTEM.

[thebrokenladder.livejournal.com]

So if you "apply" STV to a one-seat election, you get IRV. AND YOU SHOULD CALL IT IRV, NOT STV.

So there should be a different name for every different number of winners that you can use STV for? So if we're picking a council of 10 winners, that should have its own name, and if 20 winners, then that should have its own name too. You disagree? Then why arbitrarily decide that STV applied to a single winner race is somehow qualitatively different from one with two or three winners? It isn't.

Furthermore, this has nothing to do with the fact that IRV IS STV.

[thebrokenladder.livejournal.com]

you are completely impervious to the significance of the points that make you mistaken.

Name a single point that you believe makes me mistaken. Chances are I've already heard it and refuted it. There is not a single point where you could say that IRV is better than Range Voting. Not one, at least that's ever been presented yet.

And indeed, I see you've gone on to claim that I am mistaken in my own personal preferences, so you've gone beyond all previous benchmarks of arrogance.

Well, it's the truth. You'll derive significantly greater satisfaction if you use Range Voting. So to say you think IRV will make you more satisfied is wrong - unless you (bizarrely) feel that using IRV would be so much "fun" that it would make up for your dissatisfaction with the election results.

I thought I could continue to clarify this simple matter of terminological classification, but even there it turns out that your bewilderment is too great to overcome. Sorry.

IRV is STV. If I started calling it "Instant Duo Voting" whenever STV was used for two-seat elections, that wouldn't change the fact that we were still using STV, even if it was also called IDV. I'll keep calling myself a chordate, even though I'm a mammal, and a human. I hope that doesn't bewilder you too much.

[kalimac.livejournal.com]

And no, the ape/mammal comparison is inapplicable. That is not how these systems are classified. And I have not misunderstood my textbook. Here is how that book - again, it's Geography of Elections by Taylor and Johnston - classifies the systems it discusses.

1. The Plurality System

(i) The plurality system in single-member constituencies

(ii) Multi-member plurality sytems

(iii) Weighted plurality systems

2. Preferential systems

(i) Single-member preferential systems

(a) The alternative vote [i.e. IRV, which did not yet have that name]

(b) The double-ballot

(ii) Multi-member preferential systems

(a) The single transferable vote

3. List systems

(i) The simultaneous list

(ii) The local list

(iii) The party list

4. Mixed systems [which it doesn't enumerate in a list]

In the description of STV, page 59 of the Penguin edition, it says, "If, as with our hypothetical example, no candidate exceeds the quota of 651 and there are still seats to be filled, the procedure used in the alternative vote [that is, IRV] is operated."

So it acknowledges the relationship between the two, but you see that it depicts STV as borrowing the procedure for IRV - which, in fact, it does not always need to do in actual cases - not as IRV being a special case of STV.

You can look at them that way if you want, but it is not inaccurate to say that they are different systems and that a single-winner election using it should say it is using IRV, not STV.

[thebrokenladder.livejournal.com]

And no, the ape/mammal comparison is inapplicable

IRV is a subset of STV, so it's totally applicable.

..it depicts STV as borrowing the procedure for IRV - which, in fact, it does not always need to do in actual cases - not as IRV being a special case of STV.

STV doesn't "borrow" from IRV - IRV is just STV applied to a single-winner election.

From FairVote.org, the ring leaders of the IRV movement:

Instant runoff voting (IRV) was first proposed in 1870 by William Robert Ware, the first professor of architecture at the Massachusetts Institute of Technology. Ware developed the idea of IRV for elections for single winner offices after reading the works of Great Britain’s John Stuart Mill and Thomas Hare about choice voting (also known as “the single transferable vote” or STV).

The only point being made in your quote is that situations can apparently arise in multi-winner STV that have to be accounted for, but that they don't arise in single-winner STV. That doesn't mean these systems are different.

Let me make this perfectly empirical. You show me any set of rank-order ballots you can devise, where IRV picks a different winner than STV.

Clay

[kalimac.livejournal.com]

I looked at this again, and your misinterpretations are so amazing I couldn't resist saying more.

You quote the Wiki on Droop quota to counter my statement that Droop quota threshold calculation has no place in IRV. But in fact the Wiki on Droop quota calculation says just what I said: that you don't calculate it in IRV, you just go straight to simple majority.

Same thing is true with surplus vote allocation. The concept of a relative percentage calculation is essential to STV. It has no place in IRV, and your own argument says so.

A counting system designed to produce a single winner is not the same thing as a counting system designed to produce multiple winners, even if the number of multiple winners varies. That is a standard concept which you are fruitlessly trying to deny.

[thebrokenladder.livejournal.com]

you don't calculate it in IRV, you just go straight to simple majority.

The Droop quota is (votes/(seats+1)) + 1. For a single seat, this equals "half the votes, plus 1". That is the definition of a "simple majority".

You could use a different quota, where all a candidate would need is 40% of the vote to win, so in the following scenario:

40% Bush > Gore > Nader
29% Nader > Gore > Bush
31% Gore > Nader > Bush

Bush would simply win in the first round, whereas with the Droop quota (IRV) he would lose to Gore after Nader was dropped.

This isn't just speculation. There is at least one country that uses top-two runoff (hold a second election between the top-two finishers if no one gets a quota) where the the quota is not 50%, but 40%.

So let me repeat - IRV IS STV, with a Droop quota. The math is identical.

But again, I welcome you to try to present me with a hypothetical election scenario where STV with a Droop quota would pick a different winner than IRV (or do anything different at all).

[kalimac.livejournal.com]

"A preferential list system", I should say. Lest I confuse your overtaxed brain further.

[thebrokenladder.livejournal.com]

"Confuse my overtaxed brain". Why say silly things like that when we can have an intellectual debate? Clearly I'm not confused, because as my citations above indicate, your claim was wrong, and mine was right. I was just asking you what you meant, because your statement sounds vague to me. I still don't know what exactly you're trying to say. By list system, you mean STV, or does that include ALL ordinal voting systems, such as Condorcet? I can't address your claim if you use vague terminology.