Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Instant runoff voting is better than plurality voting

By MindMesh in Op-Ed
Wed Nov 27, 2002 at 06:42:06 AM EST
Tags: Politics (all tags)
Politics

Instant runoff voting, a.k.a. preferential balloting has the power to radically change election systems around the world for the better.


In plurality elections, "first-past-the-post", winner take all systems, the candidate who gets the most votes wins even if he doesn't receive a majority.  In a 3 person race, candidate A. might receive 48% of the vote, candidate B. 47% and C. the remaining 5%.  A. wins even if all the people who voted for C would have preferred B over A.  This results in arguments against voting for minor parties such as the common "don't vote for C., it's a wasted vote" and many people heed those arguments.  How many people who would have preferred Nader over Bush or Gore voted for Bush or Gore because they didn't want to "waste" their vote?  The counter line to the wasted vote argument usually goes something like "vote your conscience" or "vote your hopes not your fears"  (I'm in the U.S. but these examples apply to all similar winner take all systems)

Another problem with our system is that if there are 2 major candidates with different views, one of them might hire a 3rd candidate to run with similar positions as his opponent in order to split the vote between them and therefore have more of a percentage for himself.  This is crooked, but effective.

With instant runoff voting, when a person votes, he ranks the candidates in order of preference, 1, 2, and 3 (etc. if more than 3 candidates.)  When the votes are tallied, if no person gets a majority of votes,  the person who got the least amount of 1 votes is eliminated and the 2nd choice on the ballots for that person are counted.  So in our example above, the 5% for candidate C. would be split between A. and B.  For example, if 80% of the people who voted for C. voted for B. as their 2nd choice and 20% of them voted for A.  then A would wind up with his original 48% + 20% of the remaining 5% of the vote or 49% total while B. would have his original 47% + 80% of the 5% for a total of 51%.  B would win because out of A and B the voters prefer B.

C can run his campaign without getting bashed from all sides as a "spoiler".  This also renders the dirty tactic of hiring extra candidates useless.  Voters get to express their full opinion on all candidates involved, not just 1.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
This voting method:
o 2 thumbs up 47%
o 2 thumbs down 10%
o 1 thumb up and 1 thumb down 32%
o other 9%

Votes: 116
Results | Other Polls

Related Links
o Also by MindMesh


Display: Sort:
Instant runoff voting is better than plurality voting | 137 comments (122 topical, 15 editorial, 0 hidden)
Who's this article for? (2.50 / 2) (#2)
by Lord Snott on Mon Nov 25, 2002 at 06:16:16 PM EST

Most European countries, as well as Australia and New Zealand, already have preferential voting. The US is lagging here, but your preaching to the wrong crowd.

Do you really think the mainstream US population will even consider this "threat" to their freedom? I can't imagine your larger political parties supporting the only means to displace them.

I've always had the impression the US masses valued their "superior" democracy above others. This isn't going to change, not in the current political climate ("You're either with us, or you're Evil!")

NB: I've just proof-read over this, and it's a troll! I wasn't even trying!

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

it's for the U.S. and any other place that doesn't (5.00 / 1) (#5)
by MindMesh on Mon Nov 25, 2002 at 06:18:56 PM EST

already have it.  Most people here haven't even heard of it :(

[ Parent ]
OK (3.00 / 2) (#30)
by CaptainSuperBoy on Mon Nov 25, 2002 at 08:16:28 PM EST

"The US won't adopt it tomorrow, so it's not even worth discussing."

Riiiiiight....

--
jimmysquid.com - I take pictures.
[ Parent ]

No, what I said was... (none / 0) (#45)
by Lord Snott on Mon Nov 25, 2002 at 11:10:21 PM EST

it's not going to change as long as Americans are... Americans.

There are a few things that define Americans, such as "freedom of choice", "anti-commie", big corporations, big cars and Baywatch.

I know that's stereotypical, but in practice, it's pretty close. Even with huge community support, I'd like to see you change the "right to bear arms". It's not gonna happen (not with the present culture).

Sure, over time, your cars are getting smaller, your big corporations get broken up (AT&T) or broken down (Enron), or start to slide in popularity (McDonalds), but you still have the PERCEPTION of who you are.

Here (in Australia), movie makers were constantly berated and put down if they made a movie that wasn't about "the bush" and the "Little Aussie Battler". Aussie movies were like "Breaker Morant", "Man from Snowy River", and "Galipoli". If you made a movie about life in a Melbourne office, you were un-Australian. But Australia is one of the most urbanised countries in the world! (Apart from the city-states like Singapore).

As a culture, we have a PERCEPTION of who we are. One of the US's defining features is democracy. I don't see that changing in the near future.

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

[ Parent ]

Preferential voting is rarer than that (none / 0) (#60)
by Peter Maxwell on Tue Nov 26, 2002 at 04:50:15 AM EST

Yes Australia uses a form of IRV but New Zealand and "Most European countries" don't. They use party list systems that are proportional but not preferential.

[ Parent ]
have you read... (4.50 / 2) (#4)
by 5pectre on Mon Nov 25, 2002 at 06:18:44 PM EST

these:

http://bcn.boulder.co.us/government/approvalvote/altvote.html
http://www.sciencenews.org/20021102/bob8.asp

i definately prefer approval voting.

"Let us kill the English, their concept of individual rights might undermine the power of our beloved tyrants!!" - Lisa Simpson [ -1.50 / -7.74]

no, approval voting is better (4.60 / 5) (#8)
by dr k on Mon Nov 25, 2002 at 06:20:51 PM EST

There is already a massive amount of fraud in the voting system, and this horribly convoluted "preferential" scheme doesn't help. Not only is there the risk that an n candiate election may need to have the ballots counted n-1 times, people are even more likely to misunderstand the ballot and/or misvote.

With approval voting, you vote yes or no for each candidate. If you have a tie, flip a coin.


Destroy all trusted users!

For now, yes. (4.00 / 4) (#19)
by RobotSlave on Mon Nov 25, 2002 at 07:16:38 PM EST

I agree that approval voting is much better than plurality voting, and better than IRV, too. IRV is pretty broken in the way it tallies votes.

With that said, voters in an approval system will still be faced with a problem of strategic voting in the event that a third party approaches parity with either of the two established parties. Depending on circumstance, this could result in an extended period of political instability.

My personal election method preference de jour is for Condorcet voting, but it's as "hard to explain" as IRV. I think this problem is overstated, though, and such statements often strike me as depressingly condescending in their implied presumption of mental abilities vastly superior to those of "the masses."

[ Parent ]

Approval Voting Does Not Suffer From That Problem (none / 0) (#123)
by EraseMe on Wed Nov 27, 2002 at 06:38:11 PM EST

voters in an approval system will still be faced with a problem of strategic voting in the event that a third party approaches parity with either of the two established parties.
How you figure? Say Nader, Gore, and Bush are all about even. I would like to approve of Nader/Gore. How does this help Bush in any way? Where is strategic voting required?

[ Parent ]
Strategic voting in the approval system (5.00 / 1) (#125)
by captshad on Wed Nov 27, 2002 at 08:05:24 PM EST

Let's say your preference is Green over Democrat and Democrat over Republican. Under approval voting, I should vote for Nader and Gore. But what if the Green Party slowly turns into a major party while the Republicans decline? Voting for both Democrat and Green makes sense only if the Republican has a chance of winning. Once the Republican slips in popularity, then you are faced with dilemma of, "When should I stop voting for the Democrat?" Wait too long, you might help Gore beat Nader. If you commit too soon to Nader, you might help Bush beat Gore.

[ Parent ]
I'm told the American Mathematical Society uses it (none / 0) (#89)
by bee on Tue Nov 26, 2002 at 03:51:47 PM EST

I'm told that the American Mathematical Society uses approval voting, since among other reasons under a certain set of voting assumptions which I totally forget, it can be proven to be the 'best' system. Of course my memory is so faulty that this may be no better than FOAF information, but there you go.

[ Parent ]
Interesting but ... (3.00 / 1) (#11)
by sonovel on Mon Nov 25, 2002 at 06:30:08 PM EST

Some people can't even seem to make one choice correctly.

They claim they are disenfrachised because even winner take all ballots are too complicated.

How will these people respond to an even more complicated ballot?

How many minutes after the first instant runoff election will the first lawsuit be filed?

Even more complicated? (none / 0) (#22)
by Lord Snott on Mon Nov 25, 2002 at 07:36:04 PM EST

"I want candidate C to win. My next choice would be A. I'm putting B last."

Wow!! How can we get our minds around that one!!

If making one choice correctly is beyond someone, do we really need to take their vote seriously in the first place?

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

[ Parent ]

yes. n/t (none / 0) (#38)
by devon on Mon Nov 25, 2002 at 09:12:00 PM EST



--
Call yourself a computer professional? Congratulations. You are responsible for the imminent collapse of civilization.
[ Parent ]
Yes (unfortunately) (4.00 / 1) (#54)
by Cloaked User on Tue Nov 26, 2002 at 03:05:22 AM EST

If making one choice correctly is beyond someone, do we really need to take their vote seriously in the first place?

Yes we do, as they still make that vote just like everyone else. You can't tell from the ballot paper whether the voter knows nothing about the issues and essentially just picked at random, or if they researched the matter thoroughly and are making an informed choice. One cross looks very much like another.

Clueless or clued up, it still counts.
--
"What the fuck do you mean 'Are you inspired to come to work'? Of course I'm not 'inspired'. It's a job for God's sake! The money's enough and the work's not so crap that I leave."
[ Parent ]

They don't need to understand the system (none / 0) (#69)
by p3d0 on Tue Nov 26, 2002 at 09:35:52 AM EST

All they need to understand is that they should rank the candidates in order of preference.

The more educated crowd could try strategic voting, and would have some advantage, but that is true of plurality votes too. (For instance, a Nader advocate in Florida might have voted strategically for Gore, even though he was not the most preferred candidate.)

Also, strategic voting is harder with instant-runoff and other systems, so there should be less of it; with plurality, everyone knows that if you don't vote for the Republicrats, you're "throwing your vote away".
--
Patrick Doyle
My comments do not reflect the opinions of my employer.
[ Parent ]

I'm curious about a few scenarios and issues (4.00 / 1) (#12)
by HidingMyName on Mon Nov 25, 2002 at 06:30:13 PM EST

The example given has a small number of parties (more than 2) and is a relatively simple case. What about situations where:
  1. Does the winner nead a majority? What if noone gets a majority, who wins?
  2. If you have a large number of candidates, do you vote on all of them or just the top few (say k of them of them). Voting for a large number of candidates could mean long and complicated voting.
  3. Often, people don't know the minority party candidates, forcing people to vote for them might be a problem. How would their order be picked by voters?
  4. You think Florida's voter fraud issues are big, how will they prevent and in the event that it happens, how can we respond to allegations of voter fraud (vote reordering).
  5. Suppose that a large but tyrannical minority has a plurality, while the more moderate/acceptable candidates split the remaining votes.
  6. Another problem is if tall the choices are unpalatable, should there be a fictitious "none of the above" candidate? If none of the above wins, then must new candidates be solicited?
And finally, in the U.S. this sort of voting might actually be unconstitutional. How would it survive a constitutional challenge, or what motivation could be made to force incumbents to allow a change in the voting process?

Some answers (5.00 / 2) (#17)
by Simon Kinahan on Mon Nov 25, 2002 at 06:58:48 PM EST

Does the winner nead a majority? What if noone gets a majority, who wins?

You keep eliminating candidates, one per round, until one has an outright majority of votes in that round. When a candidate is eliminated, each vote is redistributed to the next candidate in the voter's list.

If you have a large number of candidates, do you vote on all of them or just the top few (say k of them of them). Voting for a large number of candidates could mean long and complicated voting.

Usually it is up to the voter. You can choose a first choice only, or grade every candidate. If your vote runs out of choices, it is discarded.

Suppose that a large but tyrannical minority has a plurality, while the more moderate/acceptable candidates split the remaining votes.

That is exactly the point. The article explains this pretty badly at present. Presumably moderate voters will choose one of the other moderate parties as their second, third, and so on choices, even though they might be heavily divided over their first choice. The strongest moderate candidate would end up getting all the moderate votes in the last round, and end up beating the extremist.

And finally, in the U.S. this sort of voting might actually be unconstitutional.

Ummm .... why ?

Simon

If you disagree, post, don't moderate
[ Parent ]

Constitutional Rules on Voting (4.00 / 1) (#23)
by HidingMyName on Mon Nov 25, 2002 at 07:42:50 PM EST

I was speculating earlier, so I took a quick look. I'm pretty far from being a lawyer, so please bear with me. Looking at the constitution, there appear to be some voting rules are specified. In particular U.S. constitutional rules of interest seem to be:
  • The president is elected via an electoral college as per Article II Section 1 and the 8th Amendment The electors must vote for a President/Vice President pair. Runoffs are not permitted.
  • Selecting the electoral college does not have a federally mandated method it seems. However, electors are traditionally an all or nothing affair, where the winner gets all the states electors. If electors were picked proportionately, then the electors might more closely mimic the popular vote.
  • Choosing Senators/Congressmen seems to be left mostly up to the states in Article I Section 4 .
However, voting issues have been reviewed for constitutional reasons by the high court, so I'm not sure how much federal intervention is permitted (e.g. Supreme Court) or is customary.

[ Parent ]
Thats not the 8th amendment (none / 0) (#29)
by X3nocide on Mon Nov 25, 2002 at 08:02:58 PM EST

Thats the 12th. There's additional rules as well if you keep looking up the amendments, like term limits.

pwnguin.net
[ Parent ]
yeah, but... (4.00 / 1) (#44)
by rantweasel on Mon Nov 25, 2002 at 11:10:10 PM EST

There's no reason why the Constitution couldn't be amended to use a different voting scheme, replacing the electors altogether.

mathias

[ Parent ]

Doesn't instant runoff (4.00 / 2) (#15)
by karb on Mon Nov 25, 2002 at 06:47:15 PM EST

Also have the best voting paradox ever? (Humblest apologies for the very long quote)
Suppose, for example, that 35 percent of voters prefer A first, B second, and C third; 33 percent prefer B first, C second, and A third; and 32 percent prefer C first, A second, and B third. In an instant runoff, C will be eliminated, leaving A and B to face each other. A scoops up C's first-place votes, winning a resounding 67 percent to 33 percent victory over B. But suppose A makes such an inspiring speech that some voters who liked B best move A into first place, so now 37 percent rank the candidates as A-B-C, 31 percent as B-C-A, and 32 percent as C-A-B. Now, A faces C in the runoff, not B. The votes that ranked B first become votes for C, and C beats A, 63 percent to 37 percent.

--
Who is the geek who would risk his neck for his brother geek?
That's the major flaw of Instant Runoffs (5.00 / 2) (#16)
by siobibble on Mon Nov 25, 2002 at 06:55:01 PM EST

The problem with Instant Runoff is that it can create weird candidates out of the blue because of the way candidates are eliminated, as detailed here. A better system of elimination is the Condorcet method. Condorcet is the best because there doesn't seem to be any way of expoiting the system, but is friken' complicated when it comes to deciding who actually won.

[ Parent ]
I'm confused (none / 0) (#75)
by gauntlet on Tue Nov 26, 2002 at 12:21:01 PM EST

The original poster was pointing out a paradox that can only happen in a runoff if there is another vote, that is, if voters change their mind. The voters don't get to go back to the polls, their votes just get counted again, differently.

Furthermore, Condorcet voting DOES actually have that paradox. In condorcet voting you can end up with a "Smith Set" of "winners" such that a beat b, b beat c, and c beat a.

The real problem with IRV, as I see it, is its non-monoticity. That is to say that in rare circumstances, giving a candidate a higher preference can cause them to lose.

I dno't like IRV for that reason, and I don't like condorcet because you have to make compromises in how you decide to resolve smith sets.

For various reasons, not the least of which is proportionality, I prefer Single Transferrable Vote.

Into Canadian Politics?
[ Parent ]

The same paradox? (none / 0) (#113)
by Arkaein on Wed Nov 27, 2002 at 02:38:41 PM EST

I'm not sure what you mean by Condorcet voting having the same paradox as IRV. In IRV an additional vote favoring a candidate can cause that candidate to lose. This is obviously wrong.

Then you state that in Condorcet voting, the Smith set (kind of like a winners bracket) can contain circular victory loops. While this is certainly possible there is still an absolute winner (which may vary depending on the exact method used to pick the winner from a closely contested Smith set with circular victories, of which there are a few methods to choose from, IIRC) which I do not believe will ever lose if another vote is added which favors the current winner of the other members of the Smith set.

I'm not an expert on the Condorcet method, but I believe these two issues are completely separate, and that picking the winner in Condorcet is not nearly as big a problem as the IRV paradox.

----
The ultimate plays for Madden 2003-2006
[ Parent ]

ya but (4.00 / 2) (#25)
by nodsmasher on Mon Nov 25, 2002 at 07:50:33 PM EST

those paradox's while possible are not particularly likely

I challenge any one to come up with a plausible way that could work, its going to be hard because you aren't going to have people like a then b then c, but others like b then c. What's going to happens if people are going to like b then c then a.
paradox's might be weirder hear but less common then the problems that happen in our current system
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Most people don't realise just how funny cannibalism can actually be.
-Tatarigami
[ Parent ]
I am not convinced (none / 0) (#121)
by eightball on Wed Nov 27, 2002 at 05:41:13 PM EST

First, that is a very simplistic model. It is unlikely that there would only be three distributions of votes in a three way race. Everybody's second vote is likely to be split between the two other options (this may not happen as much with voters for extremist candidates, as there may only be one direction to move towards).

I am not going to stand on this comparison, but I think it shows how unrealistic the above model can be. Say A is right wing, B is centrist and C is left wing. For the first set of voters, we have right, center, then left: makes sense. For the second set, we have center, left then right: still makes sense. For the last set of voters, we have left, right, center: ????

I am also unsure that it is necessarily a wrong out come for a person to be elected without getting more '1' votes than anyone else. You could, for example, have a very popular candidate, but the other two offer promises that a super majority decide they would like. If enough of people give our popular candidate second place, they still might have more support than either of the other two opponents.

[ Parent ]

Ok, now I'm scared. (4.66 / 3) (#18)
by Lord Snott on Mon Nov 25, 2002 at 07:11:21 PM EST

I can't beleive the comments I'm reading here. People saying that what they've already got is too complicated? Are they serious?!?!

Maybe they'd like to go to a nice, simple dictatorship. A benevolent dictatorship is the most efficient government possible (maintaining the benevolence is the problem).

Yes, preferential voting is more complex, but the benefits are there.

"Unix is SO complicated. How can there be value in it. DOS is so much more simple."

You get what you pay for.

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

Power doesn't corrupt (4.50 / 2) (#28)
by X3nocide on Mon Nov 25, 2002 at 08:01:16 PM EST

But it does attract all the wrong people. The problem is this: the people who would probably serve best in office are the people who will never run. Whether this is because they're just as well off as Secretary of Defense (secretary of war cough) without the added racist gunfire in your direction or because they're too busy dealing with life's smaller sized issues, you will never find people concerned with the preservation of American Democracy and what it stands for campaigning and telling you why you should vote for them.

But how exactly do you go about stopping people from campaigning without restraining their right to free speech? Best answer I can come up with is shame, but that takes a lot of people or a lot smaller voting public.

pwnguin.net
[ Parent ]

As unpopular as this is... (5.00 / 1) (#46)
by Lord Snott on Mon Nov 25, 2002 at 11:33:33 PM EST

I would go for a smaller voting public.

I'm always getting accused of being an elitist (maybe I am), but I think the vote should be limited to those capable of making informed choice.

I don't mean those who agree with me, I mean those who can understand the implications of what they are voting for. They UNDERSTAND what changes to trade practice will mean. They UNDERSTAND what the sale of public assets will gain/lose. They UNDERSTAND what we are doing in East Timor and why we're doing it.

We don't have to agree, but I have faith you understand the consequences, and you know why I chose as I did.

My biggest problem is find myself voting AGAINST parties and their policies. If the voters were more savvie on what's going on, the politicians would be more inclined to do what's best, and not what's popular.

I've thought long and hard, and I have no way of defining what makes someone a good voter (in my eyes). I'm too subjective, dammit. Maybe I should write an essay on it to get some feedback...

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

[ Parent ]

Don't you fear violence? (none / 0) (#74)
by bob6 on Tue Nov 26, 2002 at 12:04:20 PM EST

I agree that education and information make better votes but that's a reason to educate more people and not to let less people vote.
How would the administration decide if so and so is well informed and able to vote? Would you set up a voting test? Then who set the criteria for the test? One part of the population won't have any influence on political decisions additionnally to not being educated (and probably poor). Then how could we blame them for expressing themselves in a violent fashion?

Cheers.
[ Parent ]
Not fear, exactly.. (none / 0) (#98)
by Lord Snott on Tue Nov 26, 2002 at 09:17:00 PM EST

...but you've highlighted what I mean. People are going to be more likely to be violent than think. These are the people who vote for a war overseas (where it doesn't directly affect them), yet whinge and complain when that war comes home to roost.

I'd put money down that 90% of US civilians have no idea why Al Qaeda hates them. ("They hate our freedoms," yeah right.)

Ignorance breeds ignorance. You can't stop it without removing a huge chunk of the ignorance that already exists.

And no, like I said I have no test. How do you test for common sense? It ain't that common! I don't want to become the next Lennin, Hitler or Bush.

~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

[ Parent ]

Agreed, sort of (5.00 / 1) (#103)
by bob6 on Wed Nov 27, 2002 at 05:47:37 AM EST

People are going to be more likely to be violent than think. These are the people who vote for a war overseas (where it doesn't directly affect them), yet whinge and complain when that war comes home to roost.
This violence towards the outside world would be turned into internal violence if they are excluded from political decisions.
Maybe I didn't understand correctly your post but you proposed a reduction of the electoral mass, which is the exclusion of some people from political debates. Interestingly enough, this is not the only way to exclude people. For instance one can convince that voting is useless because all candidates are the same and the only issue is the survival of existing bureaucracies...
90% of US civilians have no idea why Al Qaeda hates them
Yet it' obvious for everyone else in the world. One of my diaries might be of interest for you.
Ignorance breeds ignorance. You can't stop it without removing a huge chunk of the ignorance that already exists.
Ignorance is not a "thing" to fight against with instant measures. Education is a long term plan and a constant "waste" of resources. Countries with highly an educated population pay a high price for it.

Cheers.
[ Parent ]
Funny as it is (none / 0) (#131)
by X3nocide on Fri Nov 29, 2002 at 12:48:09 PM EST

That doesn't appear to be the unpopular opinion.

pwnguin.net
[ Parent ]
Electorate size (4.00 / 1) (#52)
by sigwinch on Tue Nov 26, 2002 at 01:06:05 AM EST

Best answer I can come up with is shame, but that takes a lot of people or a lot smaller voting public.
That's my answer too. If the U.S. had the original ratio of Representatives in Congress, there'd be ~5,000 Representatives today. Every moderate-sized city would have its own Rep, and they'd know the Rep's repuation by word-of-mouth.

Likewise with U.S. Senators. I think the 17th Amendment is the worst amendment, even worse than Prohibition, because it provides that Senators are democratically-elected. It thereby ensures that personal reputation has little influence on the most powerful politicians the human race has. (And also ensures that advertising dollars, a.k.a. campaign donations, are concentrated into as few people as possible.)

--
I don't want the world, I just want your half.
[ Parent ]

You're right (4.00 / 1) (#122)
by curunir on Wed Nov 27, 2002 at 06:23:36 PM EST

I agree completely. The fact that the country has grown has made it unmanageable. We just don't have a change to vote for someone that we'd like to represent us. We only have choices determined by political parties with their own agendas, not ours. Our only option becomes to vote for the agenda that is the least objectionable.

I would favor a situation similar to that of the EU. Have each state be, in essence, its own country. The federal government would turn into a much weaker entity, responsible for things like a common currency and a unified military presence (individual states could still opt-out of military action if they felt like it). On the state level, there is much more chance to have your voice heard. Some states like California would probably have to figure out something so that their situation didn't turn out like the situation on the federal level (maybe divide into a North and South California)

[ Parent ]
Sortition (none / 0) (#56)
by Peter Maxwell on Tue Nov 26, 2002 at 04:24:14 AM EST

So don't reward campaigners with office! Sortition, or selection by lot, is one underappreciated alternative to elections.

One way sortition could be combined with electoral selection would be to randomly select a jury / electoral college which then deliberates and votes. This massively reduces the cost effectivness of purchasing TV time. By reducing the number of voters it also solves the problem of rational ignorance (my vote won't matter so why think too much about it?).

But anyone who deliberately tries to get himself elected to a public office [in Utopia] is permanently disqualified from holding one.

-- Thomas More, Utopia



[ Parent ]
Benevolent Dictatorship (4.00 / 2) (#33)
by The Solitaire on Mon Nov 25, 2002 at 08:25:50 PM EST

A benevolent dictatorship is the most efficient government possible (maintaining the benevolence is the problem).
There's one more requirement for a benevolent dictatorship. The dictator can't be a moron or insane. Oh yeah, and people (example 1, example 2) don't agree on what counts as benevolence.

I need a new sig.
[ Parent ]
you do? (none / 0) (#59)
by MrLarch on Tue Nov 26, 2002 at 04:40:43 AM EST

"Unix is SO complicated. How can there be value in it. DOS is so much more simple."

You get what you pay for.

So free does mean better quality! Damn, I knew I was right.

[ Parent ]

Free beer for all!!! n/t (1.00 / 1) (#99)
by Lord Snott on Tue Nov 26, 2002 at 09:18:42 PM EST


~~~~~~~~~~~~~~~~~~~~~~~~
This sig in violation of U.S. trademark
registration number 2,347,676.
Bummer :-(

[ Parent ]
It IS too complicated. (4.00 / 1) (#107)
by jmzero on Wed Nov 27, 2002 at 11:37:05 AM EST

It requires an explanation.  It means you have to do more stuff on your ballot.  And it doesn't necessarily solve the most grevious problems (for example, you'd likely still not see a Green party representative even if they had 10% of the vote country wide).

Adding a few wildcard seats fixes "the problem" better, doesn't require a change to the ballot, is easy to keep track of (no extra exposure to fraud), and is something people are familiar with (think: baseball playoffs).  
.
"Let's not stir that bag of worms." - my lovely wife
[ Parent ]

Yeah, and NOTA also (4.50 / 2) (#20)
by dirvish on Mon Nov 25, 2002 at 07:22:57 PM EST

I agree with this article. Instant run-off voting would have avoided the mess in Florida two years ago. Another good idea for voting changes is a NOTA option.

Technical Certification Blog, Anti Spam Blog
Works for Australia (4.50 / 2) (#21)
by daniels on Mon Nov 25, 2002 at 07:31:52 PM EST

We use preferential voting in Australia and have never had problems like Florida. It works, and works well.
--
somewhere in space, this may all be happening right now
If I edit this article, will I lose the comments (none / 0) (#24)
by MindMesh on Mon Nov 25, 2002 at 07:49:56 PM EST

that are already posted?

[ Parent ]
no (none / 0) (#26)
by nodsmasher on Mon Nov 25, 2002 at 07:52:09 PM EST


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Most people don't realise just how funny cannibalism can actually be.
-Tatarigami
[ Parent ]
I don't think so. (none / 0) (#27)
by dirvish on Mon Nov 25, 2002 at 07:52:30 PM EST



Technical Certification Blog, Anti Spam Blog
[ Parent ]
Yes and no... (none / 0) (#31)
by goonie on Mon Nov 25, 2002 at 08:19:58 PM EST

IRV is a better system than first-past-the-post, no doubt, but I'd take issue with your implied claim that IRV automatically prevents a Florida-like bustup.

There's nothing stopping an IRV election from being really, really close. In fact, this happens, both at an individual seat level and at a parliamentary level, from time to time. A recent example of this was in the state election in Victoria in 1999, where a few seats were decided by a dozen or so votes (out of electorates of about 20,000) after preference distribution. Not only that, the overall election resulted in a hung parliament, where three independants held the balance of power and left Victoria in limbo for several weeks while they negotiated a deal with the Liberal-National coalition, then Labor when the Liberal deal fell through. Now, that was a state election, not a federal one, but basically the same system applies.

The difference between Florida and Victoria are threefold. Firstly, "spoiler" candidates don't spoil as the second (and subsequent) preferences of the people who voted for them are taken into account. Hence, Green candidates can run without ultimately denying Labor (who they are more closely ideologically aligned with) government. Secondly, we use good old pencil and paper. Thirdly, the organisation that counts the votes is well-seperated from the political parties themselves and is left largely alone by them.

Only one of those things has anything to do with the actual voting system used.

As an complement to IRV, particularly for the US presidential election (and probably governorships), might I suggest that barring the replacement of the US electoral college with direct election, individual states consider going from the "winner-take-all" system to allocating electoral college representation in proportion to votes won?

[ Parent ]

Some states do, sort of (5.00 / 1) (#36)
by leviramsey on Mon Nov 25, 2002 at 08:48:43 PM EST

Nebraska and Maine both allocate electoral votes on the following basis. Keep in mind that the number of electoral votes each state is equal to its number of representatives in Congress plus its two Senators (with each representative representing a defined geographic district approximately equal in population): 2 electoral votes are at large, that is based on statewide plurality voting. The remainder (1 per Congressional district) are based on how a particular district voted.

The major practical problem with this is that, since it is up to the state Legislatures to switch to this system, it would be most likely to occur in states of uniform political persuasion, geographically speaking, which are not the states where it would do the most good. For instance, New York would likely be very split (Long Island and most of Upstate tend to vote Republican but New York City is nearly monolithically Democrat; with Westchester and the rest of "Upper Downstate" being more of a battleground), but it's doubtful that the Democrats (who thanks to the city tend to control the Assembly) would approve a move that would erode any electoral votes that are perceived to belong to the Democrats.

Going to Congressional district allocation has other fringe benefits, though whether they are benefits is arguable:

  • It may tend to encourage racial/ethnic minority candidacies, especially if the current trend of creating majority minority districts continues
  • It would discourage vote fraud by requiring a substantially more geographically distributed effort, as each precinct would only have any effect on 1 electoral vote. Irregularities in one area (say Palm Beach), could not possibly affect the tallies in other districts.
  • It would preserve a fundamental rationale for the Electoral College: preventing a tyranny by one region over another
  • It would allow for more impact by smaller parties, as one enclave of a few people who hold a view can more effectively campaign over a smaller area.


[ Parent ]
How will that work? (none / 0) (#40)
by NoBeardPete on Mon Nov 25, 2002 at 09:57:50 PM EST

As an complement to IRV, particularly for the US presidential election (and probably governorships), might I suggest that barring the replacement of the US electoral college with direct election, individual states consider going from the "winner-take-all" system to allocating electoral college representation in proportion to votes won?

I don't see how this would work, actually. How would a state do both 1) use instant run-off voting, and 2) allocate electoral collegians in proportion to the votes won? The two seem mutually incompatible.


Arrr, it be the infamous pirate, No Beard Pete!
[ Parent ]

Deciding the last college vote (none / 0) (#42)
by goonie on Mon Nov 25, 2002 at 10:46:40 PM EST

It would only come into play in deciding who gets the last member of each electoral college in each state.

However, why not make the electoral college larger? Then you could allocate people more closely resembling the population proportions.

[ Parent ]

That's the point (5.00 / 1) (#50)
by NoBeardPete on Mon Nov 25, 2002 at 11:55:05 PM EST

However, why not make the electoral college larger? Then you could allocate people more closely resembling the population proportions.

The electoral college is designed to _not_ reflect the population proportions. That's why the number of electors each state recieves is the sum of its senators and its congressmen.

This was considered a bargain between the large and small states. The large states, of course, wanted representation to be proportional to their size, on general democratic principle. The small states, predictably, wanted representation to be the same for each state, on the theory that the USA was being formed out of a federation of equally sovereign states. While the idea that the USA is a loose federation of sovereign states has long since lost any currency, it is unlikely that this bargain could be successfuly renegotiated. The smaller states still have enough clout to make it very difficult to change the terms, and would fight with everything they had. The larger states don't have a compelling enough interest to push something through that would only marginally increase how well represented they are.


Arrr, it be the infamous pirate, No Beard Pete!
[ Parent ]

Because you've never had so close a race. (none / 0) (#51)
by Demiurge on Tue Nov 26, 2002 at 12:48:19 AM EST

Had Bush or Gore gotten a 60% majority in Florida, then there never would have been a furor over recounts. It's because the margin was down to a few hundred votes(and under one hundred by some estimations) that all the problems plaguing the Florida electoral system were revealed.

[ Parent ]
The Problem (none / 0) (#104)
by yooden on Wed Nov 27, 2002 at 07:12:16 AM EST

It's because the margin was down to a few hundred votes(and under one hundred by some estimations) that all the problems plaguing the Florida electoral system were revealed.

The problems seems to be that you still must rely on estimates instead of just counting and be over with.

[ Parent ]
wrong (2.33 / 3) (#32)
by j1mmy on Mon Nov 25, 2002 at 08:20:44 PM EST

The biggest problem in most election systems are the candidates. Until we have people worth voting for, how we go about voting is largely irrelevant.

... and introduce new many ones. (4.50 / 2) (#35)
by xriso on Mon Nov 25, 2002 at 08:40:04 PM EST

You shouldn't just talk about the pros of this method, but also address some of the downsides. It's better to have a discussion than just advocacy.

As you have said, the good things are that people don't have to worry about wasting their vote. Also, it reduces the effect of vote-splits: (two similar parties are equally supported, and together they are a majority, but the other unified parties win first-past-the-post)

As noted in other comments, it can have some peculiar effects. Another problem is that it quite directly reduces the power of "common" parties like Dem/Rep in the USA, or Liberals/Canadian Alliance in Canada. Everybody would put these kinds of things as a second choice to their own little niche party. There would be lots of these custom parties since aspiring politicians have such a great chance at winning by making one.

  • 10 parties A to J. A is a middle-of-the-road party. Everybody has one of the B through J that they really like, but they frankly can't stand the other non-A parties.
  • Everyone votes [B-J] first, A second.
  • A loses for lack of support.
  • One of [B-J] wins, simply because they are the largest minority. It's first past the post all over again.

--
*** Quits: xriso:#kuro5hin (Forever)
You sure? (none / 0) (#97)
by qbwiz on Tue Nov 26, 2002 at 08:15:47 PM EST

Instant runoff voting(IRV) is designed to prevent this. Parties [B-J] would be knocked out one-by-one as their votes would be a minority of the total. Since everyone who voted for them would have A as their second choice, A would keep on getting more and more votes, until they had a majority. Result: A is voted in.

IRV would decrease the power of the Dem/Rep, but through a different method. Since your vote for B couldn't hurt A like in the plurality system we have today, people could vote for B. It probably wouldn't win at first, but over time, it could become a party that could be voted into office.

[ Parent ]

Yes IRV is silly (none / 0) (#102)
by Peter Maxwell on Wed Nov 27, 2002 at 04:02:39 AM EST

Given that "Everyone votes [B-J] first, A second." A will be the first candidate to be eliminated by IRV even though he/she/it is probably the Condorcet winner.

OTOH, IRV doesn't really lead to the creation of lots of small primary-vote winning parties like that. It certainly hasn't done so in Australia.

[ Parent ]

Am I just stupid... (none / 0) (#111)
by Kintanon on Wed Nov 27, 2002 at 02:04:48 PM EST

I thought the idea was that, after totalling all of the votes, the lowest percentage of 1st choice votes was eliminated, then the 2nd choice votes of those people were applied, all the way up until some party had 51% of the votes. Wouldn't that result in party A winning because none of B-J could possibly end up with 51% ?

Kintanon

[ Parent ]

Almost correct (none / 0) (#117)
by dez on Wed Nov 27, 2002 at 05:01:04 PM EST

After totalling all the votes, A would have the lowest percentage of 1st choice votes, and thus would be eliminated.

[ Parent ]
Ahhh... (none / 0) (#119)
by Kintanon on Wed Nov 27, 2002 at 05:19:23 PM EST

So you have to have some decent percentage of 1st choice votes to be elligible for second choice... Ok.
Hrmmm. I don't see a good way around that off hand... Let me think.

Damn, I don't see a good way to resolve that issue.
I shall ponder it more and post my thoughts after some more deliberation.

Kintanon

[ Parent ]

This has already been considered. (none / 0) (#137)
by vectro on Wed Dec 18, 2002 at 07:25:12 PM EST

And really, the best system out there is Condorcet. It states that if you have a single candidate that beats each other candidate in pairwise races (that is, if q is the number of ballots where candidate A was ranked higher than candidate x (for any x), and p is the number of ballots where candidate x was ranked higher than candidate A, and q-p>0, then candidate A wins).

It is actually possible for there to be no Condorcet winner sometimes. This means that A beat B pairwise, B beat C, and C beat A. The problem is that this reflects real confusion in the electorate. There are a number of possible solutions to the problem but none of them are good; it's somewhat of an unsolveable problem.

“The problem with that definition is just that it's bullshit.” -- localroger
[ Parent ]

Problem (4.00 / 1) (#41)
by jman11 on Mon Nov 25, 2002 at 10:38:45 PM EST

One of the bigtest problems with this system of voting is forcing a voter to chose. I us this in respect to the possibility of a voter being forced to choose one party over another.

If for example there are 5 parties and you really hate A & B (both of which are really popular), then after ordering CDE in your desired order you are forced to place either B or A in 4th. Thereby giving them your vote. This does assume a predominatley 2 party system.

Thereby you enforce a 2 party system and in fact reduce the alternatives. Before I get slapped down for offering an unrealistic alternative you should be aware that I have lived in a country which has what we call a "prefferential" system and encountered exactly this issue.

Of course if a person is allowed to put two people at the same number then we have a different issue. Of course in Australia someone got into serious legal trouble when he published "how not to vote" cards, detailing this method of choiuce. Of course I have forgotten his name (damn).

As I understand it: (none / 0) (#43)
by MindMesh on Mon Nov 25, 2002 at 10:53:10 PM EST

"One of the bigtest problems with this system of voting is forcing a voter to chose. I us this in respect to the possibility of a voter being forced to choose one party over another."

As I understand the system,  you are not forced to choose.  If you find them equally despicable, just rank the others and don't give either of them your preference.  

[ Parent ]

re: as I understand it (none / 0) (#49)
by jman11 on Mon Nov 25, 2002 at 11:48:31 PM EST

This is an invalid vote and will not be counted at all (including your top choice), this is the dilmena. Vote for no one at all or one of the people you don't want to. My self I always chose to vote invalid.

By the way I am expat Australian, so I did vote under this system. If it were implemented under the alternate system, where equal voting is possible I woud not have a major problem with it.

A possibility should be to vote for equal 4th and then have a 6th or something similar. Allowing you to not vote for either, but to say anyone but these people. One Nation made this a reasonable choice. I believe one of the reasons against allowing equal voting is the complexity this can raise. I don't believe this is valid, but it is an arguement used.

[ Parent ]
It is possible to rate candidates together (none / 0) (#76)
by gauntlet on Tue Nov 26, 2002 at 12:27:31 PM EST

In IRV, if you rate a number of candidates together, you can give a portion of a vote to each when counting that ranking. No reason it can't work.

Into Canadian Politics?
[ Parent ]

equal condidates (none / 0) (#100)
by jman11 on Tue Nov 26, 2002 at 10:03:04 PM EST

I'm not saying it isn't possible, merely in a system with prefferential (or IRV) voting that it was not implemented like this. The system I reference here is the Australian one, I'm assuming there are quite a few holier-than-thous coming from compatriots. These Australians are probably not considering that this can reduce the deomcratic nature of a parliament.

It also makes sense in a 2 party system that this kind of voting is enforced. Both parties dominate and want to keep it this way. The result being that both parties remain the dominant ones, with secondary parties failing due to the enforcement of this choice. In the context of adding this to the American system this will inevitably be a consideration. Both parties will want to force a voter to chose one or the other. This wil reduce the impact of 3rd parties. It must be said though that in the USoA this will not make a huge difference with 3rd parties having no influence to begin with

[ Parent ]
It depends. (none / 0) (#136)
by vectro on Wed Dec 18, 2002 at 07:21:21 PM EST

With some systems, each ballot must represent a total ordering of the candidates. But with Condorcet, this is not the case - you are absolutely permitted to leave some blank or rank multiple candidates the same. In IRV, you would be allowed to leave some blank but may not rank two candidates the same; if all your ranked candidates lost the race then your vote would be uncounted.

“The problem with that definition is just that it's bullshit.” -- localroger
[ Parent ]
Re: Better (4.66 / 3) (#48)
by thecabinet on Mon Nov 25, 2002 at 11:39:27 PM EST

One of my problems with instant runoff voting is that it treats that it implies that I disproportionately support the third party candidate (presuming that person lies somewhere between the other two). The fact that I like A, B, C in that order, does not mean I like B. Simply that I find his ideas slightly less distasteful than C's.

I'd like to see some resolution to this problem. Unfortunately, the only schemes I can think of equate to FPTP.

I'm sleepy now.

There is no solution (none / 0) (#77)
by gauntlet on Tue Nov 26, 2002 at 12:30:14 PM EST

basically because there is no way to objectively quantify how much a candidate is "liked" by a voter. You can try and give them percentage rankings, but it ends up being that people have no reason to give anything but 100% support to the candidates they like.

That problem exists in IRV, but it also exists in single approval (first-past-the-post) elections, so the problem isn't getting any worse, and we can't use that as an excuse to not change methods.

Into Canadian Politics?
[ Parent ]

I thought... (none / 0) (#120)
by WhiteBandit on Wed Nov 27, 2002 at 05:40:51 PM EST

Heh, I thought in IRV though that you don't *have* to put in a vote for any candidate. You simply rank the candidates you like, in order. For the last election, we had Gore, Bush and Nader as the "major" contenders. Like many (though I question that as of late), I really really dislike Mr. Bush. So on an IRV ballot my vote would be: 1.) Nader 2.) Gore Bush wouldn't get my vote. AFAIK, it is perfectly legal to do. Now if there was a tie and no candidate received a majority(I believe it was 48% Bush/Gore and like 3% Nader), there would be an "instant runoff." So the candidate with the lowest amount of votes would be dropped. My Nader vote would be thrown away, and now my Gore vote would count towards his total tally. And no Bush vote at all. So I believe that if you don't like a candidate, you simply don't put them down on the ballot. Also, this article could have included an obligatory link: Instant Runoff Voting.

[ Parent ]
Problem is... (5.00 / 3) (#55)
by enterfornone on Tue Nov 26, 2002 at 03:46:06 AM EST

...it still generally results in a dictatorship of the majority. It's just that the majority is decided in a different and perhaps slightly fairer way.

A better system is proportional representation, where your party gets a number of seats based on the percentage of people that vote for you. This means that if you get 1% of the vote you get 1% of the seats - first past the post and preferential systems you would normally end up with nothing.

Wouldn't work for the US presidential elections of course, but IMO having a single person with that much power is a bad idea anyway.

--
efn 26/m/syd
Will sponsor new accounts for porn.

u.s. (none / 0) (#57)
by MrLarch on Tue Nov 26, 2002 at 04:34:49 AM EST

The president is a figurehead anyway. Institute it in Congress and let our collective minds be blown.

[ Parent ]
Proportional representation (4.00 / 1) (#58)
by Craevenwulfe on Tue Nov 26, 2002 at 04:40:22 AM EST

I remember back when Scottish independance(or devolution) was being discussed, proportional representation was a hot topic in Scotland (if not the whole UK) but has since died for no apparent reason.

Does anyone know of people still campaigning for PR?

[ Parent ]
Yes, they are (5.00 / 2) (#71)
by Simon Kinahan on Tue Nov 26, 2002 at 09:47:42 AM EST

The Scottish Parliament and Welsh Assembly are elected by a semi-proportional system, in which some members are selected from national party lists, and some are selected on a per-constituency basis. I'm not quite sure precisely how it works.

The Northern Ireland Assembly, when it exists, is elected by Single Transferably Vote, which is also a proportional system, and is (AFAICT) the same as the instant-runoff system described in the article.

The Jenkins report suggested using Alternative Vote (which is like STV, but you only get to choose 2 candidates) for elections to the House of Commons, but the government has burried it in peat for five years to see how it rots. They still claim they're going to introduce elections to the House of Lords. If they do, that will probably be by a PR system.

Simon

If you disagree, post, don't moderate
[ Parent ]

Single Transferrable Vote is not IRV (5.00 / 3) (#78)
by gauntlet on Tue Nov 26, 2002 at 12:38:11 PM EST

Single Transferrable Vote is a proportional system that works as follows:

People rank their preferences on the ballot. Let's say there are 10,000 voters, and 5 seats available. Each seat will require 2000 votes. First preferences are counted to determine all the candidates that acquired 2000 first-place votes. Those people are given seats. The votes are then counted again. Votes for people that have already been given seats are recounted as a proportion of a vote for the second-place candidate, and anyone receiving 2000 votes gets a seat.

The process is repeated until all seats are filled.

I'm glossing over some technical details, here, but that's the gist of it.

It is, as you can see, totally different from IRV, which is designed to elect a single candidate with a majority. STV is designed to elect mutliple candidates with pluralities.

STV is my first choice for a voting system, because it is semi-proportional, and doesn't use party lists, which I feel could elect representatives that don't represent anyone.

Into Canadian Politics?
[ Parent ]

Oh, I see ... (4.00 / 1) (#82)
by Simon Kinahan on Tue Nov 26, 2002 at 01:47:57 PM EST

... but in that case isn't IRV just a degenerate case to STV, where the number of candidates being selected is one ?

Simon

If you disagree, post, don't moderate
[ Parent ]
STV & IRV (4.00 / 1) (#92)
by Peter Maxwell on Tue Nov 26, 2002 at 05:30:18 PM EST

... isn't IRV just a degenerate case to STV ...

Yes

[ Parent ]

No. (none / 0) (#109)
by nowan on Wed Nov 27, 2002 at 12:32:20 PM EST

Proportional elections may seem nice if you're and "underrepresented" party, but they're still a very bad idea.

First of all, it tends to give an inordinate amount of power to small groups of people. Look at the situation in Israel if you want evidence. You still tend to end up with a couple of major interests (which may fall along party lines, or possibly along coalition lines) and a few radical elements with both major groups court in order to get a majority.

Secondly, it encourages factionalism. By causing people to identify with a group or platform rather than a position on an issue, debate turns into an us vs. them war of politicking. Interestingly enough the founding fathers in the US felt that this was the biggest danger that the (then) new republic faced. James Madison (one of the members of the constitutional congress and an author of the Federalist Papers) felt so strongly about this that he thought having any sort of party system was a bad idea. Of course, when it became clear that a party system was inevitable he went on to help establish the party system we (in the US) have.

It's very important to tread cautiously in a thing as important as voting mechanisms -- there can be a lot of hidden consequences. So while I'm very interested in something like instant runnoff (or even condorcet, perhaps) I would be completely against implementing it on a nationwide scale until it'd been tested out thoroughly by a few states first.

[ Parent ]

Giving points (3.00 / 1) (#63)
by Barly on Tue Nov 26, 2002 at 07:34:38 AM EST

What if, instead of having runoffs, the candidates received points based on the order in which they were voted for.

If there are 3 candidates then each first place vote would get 3 points, second place votes would get 2 points, etc. This would alleviate the paradox that karb mentions below. This also solves the problem of getting people to vote for every candidate...they don't have to. If you like one candidate and hate all of the others then only one person gets points from you.

It's early (for me) and I just came up with this but the only major problem I've thought of concerns write in candidates. Maybe, if there are 3 people on the ballot, the first place vote would get 4 points. All of the write ins are counted together. If they actually combined to receive the most points then the vote counters would have to go back and tally the votes for each write in. I doubt this would ever really be a problem but it would still need to be worked out.



I don't think this works (5.00 / 2) (#64)
by Lacero on Tue Nov 26, 2002 at 07:59:07 AM EST

This is the same as giving scores to each candidate. This system encourages tactical voting, if your prefered candidate is at an equal footing as another it is in your interest to give the other candidate the lowest score possible, even if you like him second best. While I'm here, heres a link to the last time this was discussed. The media are listening: Let's promote Approval Voting

[ Parent ]
Good point n/t (none / 0) (#66)
by Barly on Tue Nov 26, 2002 at 08:54:01 AM EST



[ Parent ]
Look up Borda Counts (nt) (none / 0) (#79)
by gauntlet on Tue Nov 26, 2002 at 12:38:48 PM EST


Into Canadian Politics?
[ Parent ]

How about this? (3.66 / 6) (#65)
by kitten on Tue Nov 26, 2002 at 08:40:31 AM EST

One vote = one vote. Period.

This is the Information Age, isn't it? We should be able to accurately take a vote in the 21st century without resorting to all kinds of gimmicks and tricks.

This would help solve another problem with the current system. Say you're a Democrat living in an overwhelmingly Republican state (or vice versa). You may as well not even go to the poll - if all the electoral votes are going to be given to someone you didn't want, why bother?

But with a 1:1, direct vote, a Republican in a primarily Democrat area could vote and his vote would actually count for something - the same as everybody else's. Novel concept.

We'd also avoid travesties like Bush being in office, when in fact more people voted for Gore than Bush.
mirrorshades radio - darkwave, synthpop, industrial, futurepop.
It's only a travesty if it wasn't intended (4.66 / 3) (#67)
by Vygramul on Tue Nov 26, 2002 at 09:15:56 AM EST

The framers SPECIFICALLY set up the system so that this result might occur. In many ways, it's a good thing in that it creates an incentive for politicians to pay attention to the needs of those in less populous areas.

It's also important to note that MOST people did not vote for Gore. Frankly, with a vote separation of less than 1%, it's ridiculous to make this out as some sort of sin against God. It's happened some half-dozen other times in US history with much wider margins and it's never been a controversy before.


If Brute Force isn't working, you're not using enough.
[ Parent ]

Hold on a minute. (4.22 / 9) (#72)
by kitten on Tue Nov 26, 2002 at 10:39:29 AM EST

The framers SPECIFICALLY set up the system so that this result might occur.

The framers set the system up because they didn't trust the people. Period.

The actual electoral voters - the ones that make up the electoral college - are not bound by law to actually vote for the person they say they will. When you cast a vote, you aren't casting a vote for George Bush - you're casting a vote for Mary Smith, electoral, who says she'll vote for Bush. But she doesn't have to.

In many ways, it's a good thing in that it creates an incentive for politicians to pay attention to the needs of those in less populous areas.

What? When's the last time anyone made Alaska or Hawaii or some low-population state a priority? Politicians focus on the population centers - California, New York, Texas, Georgia, etc.

Fact is, if you're an X living in a Y-dominated state, there's zero point for you to even bother voting. Your vote means nothing, swallowed in the antiquated electoral college system spawned by well-meaning founding fathers of over two hundred twenty years ago, during a time when people were uneducated and information was difficult to move around. At the time, it made sense. But as the nation has evolved, the manner in which we elect our leaders has stood still.

It's also important to note that MOST people did not vote for Gore. Frankly, with a vote separation of less than 1%, it's ridiculous to make this out as some sort of sin against God.

I didn't say "most". I said that more people voted for Gore than Bush.
BUSH 50,456,169
GORE 50,996,116
A ridiculously small margin, yes. But to me it's still an issue - more people wanted Gore than Bush, yet Bush wins. The race came down to 100-something votes in a backwards county in Florida, ignoring the half million people who supported Gore over Bush.

It's happened some half-dozen other times in US history with much wider margins and it's never been a controversy before.

And I say it's wrong, and it's time to make it a controversy. Appeals to history or tradition are useless. It's like arguing that slavery is okay because it's "happened hundreds of times before in much larger numbers". At some point, someone has to step up and say, This is stupid, wrong, and needs to be changed.
mirrorshades radio - darkwave, synthpop, industrial, futurepop.
[ Parent ]
Electoral college isn't going anywhere. (4.00 / 2) (#88)
by Kyle on Tue Nov 26, 2002 at 02:51:38 PM EST

To remove the electoral college system requires an amendment to the Constitution. Amendments require 2/3 of the states to ratify them (IIRC).

The electoral college gives more power to small states than they would have otherwise, and they don't want to lose it. Therefore, they won't ratify the amendment to eliminate the electoral college.

It would take a huge outcry from the populace to overcome this. I'm not saying that eliminating the electoral college is a bad idea, just that it's a hard battle.

What I think would be easier is to change the way states do their voting. If we can get Florida (for instance) to use some other voting method, other states might take notice of the (hopefully positive) effects, and change their ways as well.

Easier still would be to start even more locally. Get a county to do it, or a city.

[ Parent ]

Iowa and New Hampshire get a lot of campaigning (5.00 / 3) (#91)
by Vygramul on Tue Nov 26, 2002 at 05:02:35 PM EST

OK, so that's disingenuous. But it's not as if smaller states are completely ignored. They would be under a direct election.

The actual electoral voters - the ones that make up the electoral college - are not bound by law to actually vote for the person they say they will. When you cast a vote, you aren't casting a vote for George Bush - you're casting a vote for Mary Smith, electoral, who says she'll vote for Bush. But she doesn't have to.

When's the last time an elector didn't vote as the people wanted?

ignoring the half million people who supported Gore over Bush

It's misleading to bring up the number 500,000 for the margin of defeat, especially when that margin is less than 1%. That's within the error of COUNTING the darn ballots. Heck, hold the election on Wednesday instead of Tuesday, and you could have had the exact opposite happen. The difference is meaningless yet the sheer number 500,000 evokes an emotional outrage that is not in proportion to the problem.

Changing the system is a valid goal. It's perhaps not the best system, and the concerns that spawned it are largely outdated. The Framers did, however, provide us with a mechanism for changing it, so as people invest less of their self-identification with the state and more with the nation, eventually it will change. (This has happened over the course of civilization. In the 1800's, an American travelling overseas did not identify themselves as "American" when asked where they were from, but rather as "Virginian" or "Marylander." In the more distant past people would answer what city they were from.)


If Brute Force isn't working, you're not using enough.
[ Parent ]

So tell me... (none / 0) (#110)
by John Miles on Wed Nov 27, 2002 at 01:04:38 PM EST

Why should New York and California get to act as supreme dictators to the rest of the country? Without the Electoral College, that's exactly what would happen: a classical tyranny of the majority, where the 'majority' is really just two large, isolated clumps of like-thinking Democrats.

The real problem isn't that the founders didn't trust the people. It's that the founders never intended the Federal government to have anything close to the kind of power to micromanage all of our lives that it has today. If they hadn't saddled us with the quaint historical artifact that is the Electoral College, the US's representative democracy really would boil down to two wolves and a sheep discussing dinner plans.

For so long as men do as they are told, there will be war.
[ Parent ]

Because... (4.00 / 2) (#114)
by Arkaein on Wed Nov 27, 2002 at 02:43:42 PM EST

right now the people in the least populous states have votes that count more than those in the most populous (due to automatic state electoral votes from senators). How is this fair? Are Californians not as important as Rhode Islanders?

Right now we have a tyranny of the states with low populations. Of course this does not sound as catchy, which is probably why it is not parotted as often.

----
The ultimate plays for Madden 2003-2006
[ Parent ]

I wouldn't call it a 'tyranny'... (none / 0) (#118)
by John Miles on Wed Nov 27, 2002 at 05:15:22 PM EST

Right now, the only way the "flyover" states can override the will of the NY and CA powerhouses is by voting as a cohesive bloc... something that doesn't usually happen to the extent it did in 2000. In the last Presidential contest, one or two states would have been sufficient to strike a decisive balance in favor of Gore. Hell, if Gore could have mustered the charisma and/or chutzpah to win his own freaking state, that would have been enough.

On the other hand, the contest for NY and CA's electoral votes wasn't even close. I guess I'd put it this way: you're right, Californians are just as important as Rhode Islanders, but they're not as important as Rhode Islanders, Montanans, Oklahomans, Texans, and just about everyone else in the country put together. Without the electoral college, though, that's exactly how it would be. An even greater injustice would prevail over the one we have now.

For so long as men do as they are told, there will be war.
[ Parent ]

100% Agreed (4.00 / 1) (#116)
by FourDegreez on Wed Nov 27, 2002 at 03:33:24 PM EST

I disagree with the very premise that smaller states need padded influence so that they aren't overlooked. This amounts to a system of weighted votes. Bob's vote carries more weight than Jane's because Bob decided to live in a rural state and Jane lives in a more urban one. This is wrong. Rural voters shouldn't have influence out of proportion with their numbers. And I contest the notion that the electoral college makes politicians pay more attention to these states. Even with the padded electoral votes, these states are still the most easily-ignored by politicians. Nothing is gained by this system, and democracy suffers.

[ Parent ]
Not quite... (none / 0) (#124)
by kerinsky on Wed Nov 27, 2002 at 06:49:19 PM EST

The actual electoral voters - the ones that make up the electoral college - are not bound by law to actually vote for the person they say they will. When you cast a vote, you aren't casting a vote for George Bush - you're casting a vote for Mary Smith, electoral, who says she'll vote for Bush. But she doesn't have to.
This isn't quite true, the electors aren't bound by the US constitution to vote in a particular manner however article 2 section 1 states, "Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors, equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress: but no Senator or Representative, or Person holding an Office of Trust or Profit under the United States, shall be appointed an Elector."

So it's up to each state, some don't require the electors to vote a certain way but others do. Actually to be totally accurate (and totally pedantic) the electors don't have to do anything but they may be required to take oaths to vote for their candidate ahead of time and presumably there could be sanctions if such an oath outlined penalties for breach of contract or some such.

Some states also split their electors so if party X gets 40% of the vote they get ~40% of the electors. Obviously this makes a state less powerful, so it's unlikely you'll see California doing this anytime soon. Electors are often, perhaps always, chosen by the political parties so it's viewed as a pretty safe system. There have only been 3 times before 2000 that electors defected so if it ain't broke don't fix it and all that jazz. Of course last Presidential election it was reported that some Democrats were calling upon Republican electors to do the right thing and defect so that the will of the people could be expressed.

Of course the fourteenth amenment does have a little more to say on the issue, "... when the right to vote at any election for the choice of electors for President and Vice-President of the United States, Representatives in Congress, the Executive and Judicial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States, or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced in the proportion which the number of such male citizens shall bear to the whole number of male citizens twenty-one years of age in such State."

Alas IANACL so I can't say whether this would be interpereted by the courts as meaning that there must be a vote for the electors or merely that IF there is a vote there will be consequences for denying any non-criminal male citizens over twenty-one but I'd lean strongly towards the latter.

Of course the electoral college is fairly high up on my list of things that should be changed in the federal government. Their are more pressing issues to me but this is a simple idea that actually has a snowballs chance in hell of coming about in my lifetime so I'll support it heartily.

-=-
Aconclusionissimplytheplacewhereyougottiredofthinking.
[ Parent ]

Stay on topic (none / 0) (#134)
by Josh A on Fri Nov 29, 2002 at 06:09:36 PM EST

This is from your original post: We should be able to accurately take a vote in the 21st century without resorting to all kinds of gimmicks and tricks.

And now you say A ridiculously small margin, yes. But to me it's still an issue - more people wanted Gore than Bush, yet Bush wins.

Yet this article specifically used Bush/Gore/Nader as an example for why we need a a better voting system such as IRV. Quite frankly the majority winner, be it Bush or Gore, in our voting system is meaningless. You can't know how many people would have rated Nader #1 if they could have voted Gore #2 and thus not felt they were wasting their vote.

This is a gimmick, a trick? Why? Because you would have preferred Gore to win, no matter what it takes? The author of this article presented what seems like at least a somewhat better system than what we use now, a few flaws notwithstanding, and yet you'd rather talk about the electoral vote? Why? Because, as far as I can tell, it didn't help your candidate win?

That just seems way off topic to me. I don't see how jettisoning the electoral college helps alleviate the problems pointed out in the article and some of the comments.

---
Thank God for Canada, if only because they annoy the Republicans so much. – Blarney


[ Parent ]
Just for fun (5.00 / 1) (#68)
by lb008d on Tue Nov 26, 2002 at 09:22:48 AM EST

This is an overview of some interesting studies an old prof of mine did regarding the computational feasibility of a couple alternative voting systems.

"Kuro5hin: politics and pretension, from the $3,000 leather recliners on the hill overlooking the trenches."DarkZero

Negative Voting (4.00 / 2) (#70)
by ShiteNick on Tue Nov 26, 2002 at 09:42:20 AM EST

Dream: Some kind of easy way to register negative as well as positive votes - not just for a candidate in his/her totality - but for each proposal / action / bill.

What I'd like is a system in which people (directly) vote for proposals, bills and the like. The populace is educated and the internet provides the reach. Why can't we have distributed judgements? I mean politicans should:
    Suggest new ideas/schemes/bills/laws
    Educate / Canvas for these schemes and ideas
    Monitor the implementation/execution of the schemes.

This way people would vote for/against each bill (and be provided with all required material for making a decision) instead of the current system in which you have to vote for someone who you believe will implement schemes and ideas that they have proposed.

That would get rid of a lot of the messes, I'd think. And we're all educated enough to do it.

The whole idea of "elected representatives" does not work all that well. Representatives may represent some ideas but not all!



that would be a disaster (5.00 / 2) (#93)
by ogre on Tue Nov 26, 2002 at 06:25:54 PM EST

People wouldn't vote on most of the issues for the very good reason that they have lives and don't have the time to keep up on ever law someone somewhere wants passed. So they decide if they don't have time to learn the facts and make an informed decision they should make no decision at all. But this means that if you can get a large enough minority who care enough to vote on it, you can get anything at all passed.

Specialization is an essential factor of civilization. You have specialists make your shoes, cure your diseases, fix your cars and make your laws. You can't expect everyone to do everthing, it isn't efficient and doesn't lead to quality work.

Everybody relax, I'm here.
[ Parent ]

Good point. (5.00 / 1) (#94)
by ShiteNick on Tue Nov 26, 2002 at 07:20:49 PM EST

There had to be a bleeding flaw, hadn't there?


[ Parent ]
Thats why we live in a republic (none / 0) (#128)
by auraslip on Thu Nov 28, 2002 at 02:20:56 AM EST

So rich white men(sorry for the typical racism) who have more time to study the issue can decide what's best for us.

And not a democracy
124
[ Parent ]

Thoughts.. (5.00 / 1) (#73)
by jmzero on Tue Nov 26, 2002 at 10:39:59 AM EST

I think the best system for the states right now would be the one they have - with one change.  5 or 10 representatives should be allocated to parties based on "unrepresented" country-wide support.  Thus, the party whose percentage of representation lags behind their percentage of support by the largest amount gets a seat (and then this process is repeated until all of the wildcard seats are allocated).  Thus, you might see the Greens or Libs getting a few representatives (potentially very powerful when the house is balanced).  You'd possibly also see a few extra Democrats or Republicans.

Unlike the setup the article suggests, this setup would actually see the Greens/Libs/whatevers get representation right away, and would certainly dispel the notion of "wasted votes" in any jurisdiction.  Even if you knew your area was going Democrat, your Republican/Green/Whatever vote would still possibly mean something.

.
"Let's not stir that bag of worms." - my lovely wife

land mines are better than nuclear bombs (5.00 / 7) (#80)
by gauntlet on Tue Nov 26, 2002 at 12:54:23 PM EST

Because they don't cause as much damange. They still cause damage arbitrarily, causing atrocious injury to people, but whatever.

Instant Runoff Voting is non monotonic. In rare circumstances, placing a candidate higher in your list will actually cause them to lose. Instant Runoff Voting is subject to voter manipluation, but only as much so as the current system, so I can't really complain about that.

Condorcet voting, on the other hand, is monotonic, and almost insucceptible to voter manipulation. The trade-off there is that sometimes you don't know who wins. :)

Let me be more clear. In condorcet voting, which requires people to use the exact same ballot as Instant Runoff and just counts the votes differently, you can end up with a circle of "winners" such that candidate A defeated candidate B, candidate B defeated candidate C, and candidate C defeated candidate A.

You'd hardly believe it was possible, but it is. It accurately represents a real ambiguity in the opinions of the voters. The fact that these sorts of ambiguities CAN exist is the very reason that most other voting systems are considered ineffective. They don't reflect things like this.

In an ideal world, you would keep having elections until the community lost its ambiguity, and picked a condorcet winner. In the real world, you have to pick a method to solve these ambiguities, and each of the methods for solving them have tradeoffs.

There is also a question as to whether or not having condorcet winners elected is actually a good thing. On average, people will tend toward the centre. That's why it's called the centre. Which is more effective, a parliament comprised of 100 representatives aligned generally with the centre, or a parliament comprised of 100 representatives, representing the actual proportional distribution of the individuals they represent?

I don't know the answer to that question. But if we decide to change voting methods in a single-winner contest, I'm voting for condorcet over Instant Runoff any day. (See how my opinion, in real life, is a ranking? Neat, hey?)

And that's all without opening the can of proportaional representation worms.

Into Canadian Politics?

Voter manipulation in the real world? (5.00 / 1) (#126)
by Repton on Wed Nov 27, 2002 at 08:36:36 PM EST

I've seen examples of various voting systems showing how you may need to vote for someone you dislike in order to have your preferred candidate win, or similar tricks.

The thing these examples have in common is that they all feature only a handful of voters.

In the real world, you typically have millions of voters. So to what extent is it possible for one person to manipulate the results? ...given that your single vote (or preference list, or whatever) has much less weight (comparitively), and that you have no way of knowing exactly how everyone else will vote.

?

--
Repton.
They say that only an experienced wizard can do the tengu shuffle..
[ Parent ]

We can't even get it to work on K5... (4.00 / 1) (#81)
by dipierro on Tue Nov 26, 2002 at 01:20:31 PM EST

Do you really think anyone's going to do it for the whole country?

Anyway, here's an example of a vote which probably would have turned out different using instant runoff. You'll notice that most people voted against creating a science section, yet adding a science section still won the plurality. But hey, y'all voted me down when I pointed it out to rusty, so I guess y'all will vote me down again.



It's all in the technique (none / 0) (#83)
by inadeepsleep on Tue Nov 26, 2002 at 01:49:25 PM EST

Your first mistake was thinking that k5 is a democracy. It's run by Rusty. You second mistake was thinking that that poll had some kind of scientific validity. Note that there are a couple of opposition questions to break up the opposition vote. And the way the idea was worded before the poll affected the results. So, he gave himself a good chance of it succeeding, and if it did not, there's always the tried and true method of all failed ballot measures: Put it up again and again until it finally passes.

Not that I know whether Rusty was really strongly in favor of the change, but any talk about k5 being a community run site is pretty much just talk.

Whether and how much this applies to any particular government is left as an exercise for the reader ;)




[ Parent ]
Interesting (none / 0) (#84)
by dipierro on Tue Nov 26, 2002 at 02:21:59 PM EST

Your first mistake was thinking that k5 is a democracy. It's run by Rusty.

Well, no, I never thought K5 was a democracy. Of course, neither is the United States.

You second mistake was thinking that that poll had some kind of scientific validity.

Never thought that either.



[ Parent ]
Ok, but... (none / 0) (#85)
by inadeepsleep on Tue Nov 26, 2002 at 02:29:19 PM EST

But when you say "We can't even get it to work on K5" you're at least saying that you think it should be a democracy or is purported to be one or some aspects are or something somehow about democracy.

And I apologize for saying you made several mistakes. It was merely meant to be a rhetorical device, I assure you :)


[ Parent ]

No problem... (none / 0) (#87)
by dipierro on Tue Nov 26, 2002 at 02:32:21 PM EST

Rusty said "imrdkl suggested it, and you voted on it, so in accordance with the majority vote, I've added a section for Science." I consider that purporting that K5 is a democracy in some aspects.

[ Parent ]
Simpler still... (3.00 / 1) (#86)
by poopi on Tue Nov 26, 2002 at 02:29:23 PM EST

...vote for the person you DO NOT want to win. The person with the least votes is the the most agreable to the majority. Sure this is a cynical type of democracy, but at least it keeps the rif raf out of governament.

-----

"It's always nice to see USA set the edgy standards. First for freedom, then for the police state." - chimera

Keen.. (4.00 / 1) (#96)
by Kwil on Tue Nov 26, 2002 at 07:59:01 PM EST

..so by throwing my hat into the ring at the last minute and then just keeping as low a profile as possible, I win!

After all, since you can't please everybody, anybody who puts forth an idea is going to get some people who don't like it, and thus vote him out. Meanwhile, the person who puts forth nothing hasn't offended anybody - so since everybody has someone else they dislike more, the silent sam wins the election in the end.

Of course, nobody says you have to stay silent after you've won.

That Jesus Christ guy is getting some terrible lag... it took him 3 days to respawn! -NJ CoolBreeze


[ Parent ]
Ah voting...... (4.00 / 1) (#90)
by ipex on Tue Nov 26, 2002 at 04:58:48 PM EST

Great topic- terrible article

There are even more fundamental problems with elections- ONE person wins!

Not to mention election fraud. It is actually very diffcult to vote in america- during the last election my voting location was behind locked doors, literally, it in a house which was situated in a private gated community where you had to explain to the security gaurd you were just trying to vote and some address. That is bullshit but is probably very typical. Then there are the issues about double registration.

We can even go theoretical and discuss how majority voting (even between runoffs) is NOT RATIONAL! Yes the definition of rational is that 1 we can decide between items and 2 tranversality, if A>B and B>C then A>C. However in majority voting you can run into situations where this does not hold- yes its true.....

then there are voting systems. Given all the wonderful technology which we entrust to secure every financial transaction we make (which I trust a lot more than election), its a wonder that we use voting technology from ancient times

Arrow's Paradox and other problems (4.75 / 4) (#105)
by mysta on Wed Nov 27, 2002 at 07:24:47 AM EST

There's also a pretty fundamental problem with preferential voting that wasn't mentioned: Arrow's Paradox. Under some pretty reasonable conditions it shows that there cannot exist a social choice that accurately reflects the choices of the individuals.

Preferential voting isn't free from crooked dealings either. In Australia we use a preferential system that allows you to either state your preference for all the candidates individually (below the line voting) or by just ordering your preferences for the parties (above the line voting). The parties have huge "how to vote" campaigns during the lead up to an election in which they try to convince you how you should place your preferences. Each party's preferred ordering is the result of a lot of backroom dealings.

To make things even more interesting, it is compulsory to vote here if you are over 18.
---
Are we not drawn onward, we few, drawn onward to new era?
[ Parent ]

Re: Arrow's Paradox and other problems (none / 0) (#130)
by Fizyx on Fri Nov 29, 2002 at 11:20:06 AM EST

Thanks for mentioning Arrow. Any discussion on this topic must include his work. To understand why, let me paraphrase it, "EVERY voting system can be unfair under certain scenarios." It's really wild stuff, but at least in designing a new system you should be aware of the unfairnesses you choose.

Setting aside the problem on agreeing on what to change to change it to (vote on it? with what system? vote on that too?), the real problem is in getting it implemented:

How many people who would have preferred Nader over Bush or Gore voted for Bush or Gore.
Don't think the powers-that-be aren't aware of that possibility. Why do you think we have the system we do!

[ Parent ]
sounds like nader (2.00 / 4) (#95)
by minus273 on Tue Nov 26, 2002 at 07:26:57 PM EST

" Another problem with our system is that if there are 2 major candidates with different views, one of them might hire a 3rd candidate to run with similar positions as his opponent in order to split the vote between them and therefore have more of a percentage for himself.  This is crooked, but effective."

hahah


Great topic, but.... (4.00 / 1) (#101)
by NFW on Wed Nov 27, 2002 at 12:28:52 AM EST

While I love to see discussion of voting systems (mostly because I strongly believe that the USian system is fxxked), this article doesn't do the topic justice. IRV is cool and all, but there are a lot of things better than plurality voting, and there are a lot of pages out there already that do a better job descriving IRV and it's up and downs.

I beg you to do an MLP article that links to a handful of web pages describing a handful of voting systems, and summarizing their strengths and weaknesses. I'd +1FP something like that for sure. It would probably get some great discussions rolling, too.


--
Got birds?


Try this on for size (2.00 / 1) (#106)
by phybre187 on Wed Nov 27, 2002 at 08:25:32 AM EST

The person with the least votes wins the election, with a minimum of 1 vote necessary, of course. No electoral college involved. And everyone gets as many votes as they want. In the event of a tie, all the tying candidates are announced, and it comes to a CNN poll, and whoever scores lowest wins. Either people would be prompted to get up and anti-vote right off, or they'd be prompted to change the atrocious system afterwards. Either way, people are getting of their ass and paying attention to government policies. I don't know how well this would work for legislation votes, though.

It wouldn't work at all, though, unless people learned how to count ballots. I'm talking to Broward county.

Correct! (none / 0) (#108)
by poopi on Wed Nov 27, 2002 at 11:50:05 AM EST

Oh sure it ain't perfect. But at least you won't have to hear "Well, I didn't vote for him!" ever again!

-----

"It's always nice to see USA set the edgy standards. First for freedom, then for the police state." - chimera

In other news... (2.50 / 2) (#112)
by The Great Wakka on Wed Nov 27, 2002 at 02:29:04 PM EST

Sky found to be blue; Earth round; software buggy.

How on Earth did you ever find out that plurality voting doesn't work?

Here's another system (none / 0) (#115)
by joemorse on Wed Nov 27, 2002 at 03:04:56 PM EST

Yeah, IRV is great. I support it whole heartedly, and can't wait to see it used in San Francisco. But then, I think I have a better idea.

Instead of elections, let's have a single national legislature with 1000 representatives. Each representative would be selected every 3 years by random lottery, from the pool of citizens. After the end of 3 years, said representative would face a statewide referendum, a simple yes or no vote. If the vote is yes, the rep gets another 3 years. If no, another lottery is held. Representatives would be paid high salaries, and those who refuse to serve would pay 100% income tax for the duration of the 3 year term.

The best part about this system is that it has a greater statistical chance of creating a government that is actually representative than our current system does.

NOTE: Those who haven't figured out by now that I'm kidding should consult their psychiatrists.



Now let's you just drop them pants!
       -Don Job, from Deliverance
ahh yes, Chaocracy (none / 0) (#133)
by Josh A on Fri Nov 29, 2002 at 05:39:57 PM EST

Peter J. Carroll has suggested randomly choosing our legislators as well. But, at a regular interval, 1/2 of them are replaced, and replacements chosen randomly again. No voting necessary. They are also to be paid well enough to make corruption as unlikely as possible.

---
Thank God for Canada, if only because they annoy the Republicans so much. – Blarney


[ Parent ]
Condorcet criterion (none / 0) (#127)
by dhilvert on Wed Nov 27, 2002 at 11:30:21 PM EST

Regarding the comment, "B would win because out of A and B the voters prefer B":

The Condorcet criterion for an election method is that if a candidate would be preferred in a one-on-one election over each of the other candidates, then this candidate will be the winner of the election. IRV does not satisfy the Condorcet criterion.

A more complete discussion of the criteria which various election methods satisfy is available at the URL below; the authors of this site apparently do not look favorably upon IRV, but it should be noted that IRV is not the only method in which candidates are given rankings:

http://electionmethods.org/evaluation.html

I don't take that site too seriously (none / 0) (#132)
by eightball on Fri Nov 29, 2002 at 03:15:57 PM EST

Considering that two of their criteria are only applicable to Condorcet elections. Like saying Democrats are worse because they are not Republicans.

I believe they also have a simplistic view of the strategies that could be employed with condorcet. It may take more thinking ahead to have a strategic vote in condorcet, but that does not mean it is not possible. In the same way you could arrange your votes in a simple "order your preferences" vote, you could arrange your condorcet votes to mirror that insincere voting pattern.

I believe in a modified Borda system. I would get rid of the requirement to order every candidate. To replace this, any candidates not ordered would get a score of one less than the lowest voted candidate. I would even consider a "none of the above vote" and the ability to have "ties". As in sports scores, if there is a two way tie, the next lower is decremented 2 points, not just one. I believe this system can be used to simulate a plurality, approval, cardinal and borda votes depending on how the person wants to vote.

Even stranger, I would consider ordering the points up instead of down. The person with the least number of points wins. What you get from this is the ability to have ballots with a different number of candidates and still have it work out properly. Your first choice will always equal the same no matter how many candidates there were.. I not sure how much sense this would make and am not sure it would introduce more problems that I haven't seen yet.

[ Parent ]

But not the best... (none / 0) (#129)
by pla on Thu Nov 28, 2002 at 03:45:48 AM EST

IRV does better at expressing the will of the people than plurality voting.

But *ranked* voting (sorry, don't know a catchphrase for it) performs even better.

Consider an example similar to your own, where IRV fails: 3 candidates, 34% of people choose A as their first choise, 34% chose B, and 32% chose C. Now, in IRV, C gets tossed. BUT!!! What if a significant majority of the A and B first-choicers would have picked C as their second choice? In that situation, C *clearly* "should" win, but IRV would exclude C as the very first elimination.

And if you think that seems like quite a stretch, consider the US in the situation of a hard-core democrat, a hard-core republican, and a middle-of-the-road 3rd party running - Many people would hesitate to vote their own party (people seem to dislike extremists), and of those who do, as their second choice they would want the person least antagonistic to their own party's views...

As a nice bonus, ranked voting doesn't even differ from IRV in the format, just the post-processing. Instead of eliminating the person with the fewest "1" votes, the candidate with the overall lowest (or highest, depending on how you assign points) score wins.


Borda Count is the least flawed (none / 0) (#135)
by narfle the garthog on Wed Dec 04, 2002 at 12:12:04 AM EST

Chaos, but in voting and apportionments?

Instant runoff voting is better than plurality voting | 137 comments (122 topical, 15 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!