Gaming the Vote

I'm sure Bush and his cronies would find a way to cheat in that system as well.

Well, it would be harder to cheat with Range Voting, because the margin of loss between Bush and Gore would have been greater without the spoiler effect that Nader caused. And lots of other candidates could have run as independents, or even in their own parties, and with Range Voting, they could have had a fighting chance even if they didn't appear to be "electable".

Also, "cheating" (election fraud) is a smaller problem than our bad voting method, in a methematical sense:
http://rangevoting.org/RelImport.html

by the way, we suggest a 0-based range, although it's pretty arbitrary. so 0-99 or 0-9 or 0-10 is better than 1-10.

http://rangevoting.org/Why99.html

I think Range Voting is the most important issue in the world, and Gore needs to shift his focus to Range Voting. It doesn't matter how much you convince people that global warming or whatever other calamity is a big deal. If they can't elect the leaders they really want, who will address their concerns, then they can't respond to their concerns.

What's the point of convincing people that we should pass stricter emissions legislation if the only people who appear to be "electable" are the ones that won't do it, because they've taken so much money from big business (in order to appear to be electable) that they can't focus on that?

http://rangevoting.org/LivesSaved.html
http://rangevoting.org/WorldProblems.html

andrew_e:

Your point about strategic voting has been duly considered by election methods experts. If every voter gives out only 10's and 0's, then you effectively have Approval Voting, which is still an excellent voting method. In fact, computer simulations show that Range Voting is about as representative with 100% strategic voters as Instant Runoff Voting (the single-winner form of the STV method you mention) is with 100% honest voters. (Notice that IRV is one of the worst, right after plurality.)

http://rangevoting.org/UniqBest.html
http://rangevoting.org/StratHonMix.html

Of course the reality is that there will be a mixture of both honest and strategic voters in any voting method. In the IRV system you propose, a situation can commonly arise in which a bloc of voters who prefer e.g. Green>Democrat>Republican can cause their second choice to win instead of their third by strategically "betraying" their sincere favorite in order to top-rank the Democrat. Since the odds that this move will help them are greater than the odds that the third party candidate will win, this move is almost always strategically advisable, and so IRV does not help to break two-party domination, and so you still get two choices, and so IRV is effectively not much better than plurality in practice. Here is some deeper mathematical analysis of that, as well as some empirical proof:

http://rangevoting.org/TarrIrv.html
http://rangevoting.org/AusAboveTheLine07.html
http://rangevoting.org/AusIRV.html
http://rangevoting.org/TTRvIRVstats.html

One common point of confusion is between single-winner and multi-winner voting methods. Range Voting, as well as IRV, are both single-winner methods, for things like mayor or president. STV is called IRV when used to elect 1 candidate, but it can also be used to elect multiple candidates; and in that case it is actually a pretty decent voting method, because it causes a degree of party proportionality. Still, proportional methods like Reweighted Range Voting and Asset Voting are better and simpler.
http://rangevoting.org/PropRep.html

But as Range Voting is a single-winner method, it is only fair to compare it to the single-winner use of STV, called IRV. You mention Ireland, but you fail to note that since IRV come into use in Ireland to elect the President, in 1938, the Fianna Fail party has won it all but once, causing not even a duopoly but closer to a MONopoly:

http://rangevoting.org/Ireland2002.html
http://rangevoting.org/Ireland1990.html

Also it increases ballot spoilage rates whereas Range Voting experimentally *decreases* spoiled ballots:
http://rangevoting.org/SPRates.html

Also IRV cannot be conducted on ordinary voting machines, like Range Voting can, and so it incentivizes the adoption of (fraud-prone) electronic voting machines.

Also, IRV cannot be totaled in precincts, and so it requires central tabulation, which is a nightmare for election integrity, as it's an open invitation to centralized fraud.

With regard to strategic voting, which was your main point, here are some additional informative links:

http://rangevoting.org/HonStrat.html
http://rangevoting.org/Honesty.html
http://rangevoting.org/RVstrat1.html
http://rangevoting.org/RVstrat2.html
http://rangevoting.org/RVstrat3.html
http://rangevoting.org/RVstrat4.html
http://rangevoting.org/RVstrat5.html
http://rangevoting.org/RVstrat6.html
http://rangevoting.org/DH3.html
http://rangevoting.org/ShExpRes.html

Here's the review you wrote for Salon, with corrections by voting experts:

http://rangevoting.org/FarhadMS.html

Has range voting been used in any political elections?

andrew_e:

See this page:
http://rangevoting.org/OlympHist.html

Approval Voting, the simplest form of Range Voting, was used (apparently quite successfully) for over 500 years in ancient Venice. It took Napolean's conquering rampage to snuff it out. Also the "Shout" method of voting in ancient Sparta was essentially Range Voting.

Furthermore, Approval Voting is used in several organizations with thousands of members -- more than the populations of many whole cities.

See http://rangevoting.org/Approval.html

animenin:

i'm not convinced that voter turnout is a very significant issue. if you consider that those who turn out are at least somewhat representative of the electorate as a whole, and actually cared enough about the elections to show up, it may even be better to keep less interested people at home.

however, studies of voter turnout suggest that range voting would have a "side-effect" of increasing turnout. that is because it would allow for a greater number of viable contenders (not just 2), and one major factor in turnout is the impression within a voter that at least one viable contender is greatly different in quality from another. so say you were a ron paul supporter who hated mccain and obama so much you wouldn't turn out for the general election, but with range voting paul could easily run as an independent and conceivably do well (since there's not the vote-splitting problem). then all those ardent paulites would have a reason to show up and make their voices heard.
http://rangevoting.org/NonVoters.html

of course, voter psychology is very bizarre -- the biggest thing affecting turnout may be weather. and the first candidate on the list has on average a 2.33% better result. some countries actually randomize the order of candidates on different ballots, so this effect is minimized.

Range Voting 10 - well, there's hope for sexier 'dates, um, someone under 71yo -
though there's nothing wrong with 71 - if you're having an old party!

putting RV into service 01 - no sexy 'dates -
too bad, it's a good idea

Well, personally I think it sounds like a good system. It's a bit unconventional though, or maybe just more modern; for that reason, I don't think it would be ratified, not for a while anyway. And it also depends on the popularity of the book and who finds out about the idea.

And to refute an earlier idea that someone had, even if someone were to vote a 10 on their candidate and 0 for all others, that's fine. I doubt a vast majority would. For instance, in this particular race I know voters who are at a loss as whether to vote for Hilary Clinton or Barrack Obama. And I imagine there were (or still are) voters who are debating between Republican candidates, or a Green party member.

I think it's at least worth a try, perhaps in a big city local election, or a state willing to try it. If it doesn't work any better, there's always the old system or the transferable vote system that one person was talking about.

zweed:

this isn't some kind of revolutionary concept.
http://rangevoting.org/OlympHist.html

I don't know if I agree only because just like with hot or not, most people vote in extremes anyway. If they like a candidate, they'll choose 10. If they aren't interested in voting anyone else into office, they won't have to rate them regardless. Obviously, if you dislike a candidate, people will pencil in a bubble and a -100 number next to it and fill it in.

I think this would work in the primary season where there are dozens of candidates and secondary votes matter more. As for general elections, I think that most people wouldn't bother rating someone they're not even going to vote for. And if we're voting in 10s and 1s, we may as well just keep it as fill in the blank for yes, leave it blank for no.

VSiskos:

As I mentioned above, at some length, evidence says that you are wrong that "most people will vote in extremes anyway". e.g.

http://rangevoting.org/HonStrat.html
http://rangevoting.org/Honesty.html
http://rangevoting.org/RVstrat6.html

And even if there is a great deal of strategic exaggeration, this is less harmful to Range Voting than it is to most other voting methods, according to extensive computer modeling of elections.

http://rangevoting.org/UniqBest.html
http://rangevoting.org/StratHonMix.html
http://rangevoting.org/ShExpRes.html

You make the classic "rookie" mistake of assuming that everyone will vote only 0's and 10's, ignoring the vast research that has been done by experts in this field saying otherwise. I'd suggest you browse through RangeVoting.org and even consider reading this book to see how much reality differs from your intuitive assumptions. It's a really fascinating and important subject.

Well, there are studies that say you're wrong. So let's please not go down this right/wrong path. I was speaking anecdotally as I obviously didn't offer proof, just thinking off the top of my head. You're obviously passionate about range voting, so considering I haven't dealt with this issue, debating with you would be close to impossible since I'm sure you already have all your points bookmarked, read and memorized.

But just doing a quick google of the subject brought up a larger debate that's not so one sided as you'd like it to be. Maybe not for the reasons I mentioned, but it's still far from decided.

In the end, I make no opinion of it because I don't know much about it. But based on my experience with range voting, and my friends who have played with sites like hot or not, my anecdotal statement from before hasn't changed.

I wish you luck, and I can at least agree with you that any voting system that allows for more people to be heard is the best way to go whether that ends up being what we have now, range voting or otherwise.

VSiskos:

You say there are "studies" that say I'm wrong. Instead of challenging you to cite those "studies", I will lower the bar and ask you to cite just one. I'll bet that you cannot.

Instead you cite anecdotes which you assume apply more universally than they really do. I also made the same argument when I was new to Range Voting and had only my intuition to go on. If all (or nearly all) Range Voting users were going to strategically exaggerate their scores, then we might as well use Approval Voting (just like our current system except that you can vote for more than one candidate). But there is massive evidence that this won't happen; and even if a small number of voters use intermediate values, social utility (net satisfaction) increases, so Range Voting is better than Approval Voting.

You also ignore (probably because you do not have any analogous anecdotal experience with them) that other methods are *more* subject to strategic exaggeration, like when Nader supporters vote for Gore so they don't "waste" their vote. Or the burial problem that comes up with ranked methods like Condorcet, Borda, and IRV, e.g.
http://rangevoting.org/TarrIrv.html
http://rangevoting.org/DH3.html

So when you say that your anecdotal statement hasn't changed, it's a little frustrating, since you didn't even know the relative strategy resistance of alternative methods, nor anything about the prevalence of of this behavior outside your selective memory of your selective set of friends.

Furthermore, the fact that you have found Range Voting detractors online says nothing about whether the situation is one-sided. You will find, for instance, supporters of Instant Runoff Voting who criticize Range Voting for violating majority rule. But they ignore that...

1. IRV *also* can go against majority rule, electing candidate Y even though X is preferred to Y by a majority of voters *and* got more first-place votes than Y, e.g.
http://rangevoting.org/CoreSupp.html
(But they will lie and say this is "unrealistic", ignoring numerous real world examples.)

2. That the axiom that X must be better than Y if X is preferred to Y by a majority of voters is logically disproved:
http://groups.google.com/group/socorg/web/majority-myth

So you become the one millionth person to

1. Hear about Range Voting.
2. Decide that you know more about it than experts who've spent decades studying voting methods, since *obviously* your friends prove that 100% of voters would exaggerate with Range Voting.
3. Naively assert that Range Voting has this flaw that it's proponents must have naively missed.

So, not to bash you, but I'd spend a lot more time on this subject before making these assertions to experts.

I'm not going to engage you any further since you obviously didn't read what I wrote above. I don't disagree with you, I simply stated my assumption based on my experience while ALSO stating I don't have an opinion of range voting either way exactly BECAUSE I don't know much about it.

If you want to teach people about range voting, I suggest a better tone and understanding as opposed to bashing people for not even disagreeing with you in the first place.

Again, I wish you luck with everything, and I'll probably see you around the site :D

okay, i didn't mean to bash you. it just seemed like you were saying that

1. there are studies that contradict my claim that range voting will inspire a great amount of use of intermediate values, not 100% exaggeration.

2. there is some legitimate controversy about the issue such that it is not "one-sided".