Instant Runoff vs. Approval Voting

Things that don't belong anywhere else. (Check first).

Moderators: Moderators General, Prelates, Magistrates

Instant Runoff or Approval Voting?

Instant Runoff Voting
10
45%
Approval Voting
12
55%
 
Total votes: 22

User avatar
Qaanol
The Cheshirest Catamount
Posts: 3058
Joined: Sat May 09, 2009 11:55 pm UTC

Instant Runoff vs. Approval Voting

Postby Qaanol » Tue May 10, 2011 2:43 am UTC

Which do you prefer? We will vote on them using a first past the post system.

Instant runoff voting has each voter rank as many candidates as they want in order from favorite to least favorite. Everyone’s first choice is counted as a vote, and the candidate with the fewest votes is eliminated. Voters who preferred that candidate now have their second choice counted instead. The candidate with the fewest votes is again eliminated, and votes for that candidate switch the next-favorite for each voter. This continues until all but two candidates have been eliminated, and the one with more votes wins. This works exactly as its name implies, where runoff elections are held after the weakest candidate is eliminated at each round.

Approval voting has each voter choose any number of candidates to vote for. The one with the most votes wins.

Instant runoff voting has the problem that if one candidate is everyone’s second choice, that candidate will lose in the first round even though everyone approves of that candidate, and even if that candidate would beat each other candidate in a head-to-head election. Also this method is rather complicated for voters, as Murphy’s law guarantees some people will confuse “low numbers are better” for “high numbers are better”.

Approval voting is simple for voters, as you just need to vote for any candidates you like.
wee free kings

User avatar
Aaeriele
Posts: 2126
Joined: Tue Feb 23, 2010 3:30 am UTC
Location: San Francisco, CA

Re: Instant Runoff vs. Approval Voting

Postby Aaeriele » Tue May 10, 2011 3:03 am UTC

*cough*

Forum search is your friend.
Vaniver wrote:Harvard is a hedge fund that runs the most prestigious dating agency in the world, and incidentally employs famous scientists to do research.

afuzzyduck wrote:ITS MEANT TO BE FLUTTERSHY BUT I JUST SEE AAERIELE! CURSE YOU FORA!

User avatar
Mat
Posts: 414
Joined: Fri Apr 21, 2006 8:19 pm UTC
Location: London

Re: Instant Runoff vs. Approval Voting

Postby Mat » Tue May 10, 2011 8:11 am UTC

Qaanol wrote:Approval voting is simple for voters, as you just need to vote for any candidates you like.

What if I don't like any of them?

User avatar
Von Haus
Posts: 757
Joined: Thu Jun 19, 2008 5:15 pm UTC
Location: Probably somewhere he shouldn't be...

Re: Instant Runoff vs. Approval Voting

Postby Von Haus » Tue May 10, 2011 8:23 am UTC

Qaanol wrote:Also this method is rather complicated for voters, as Murphy’s law guarantees some people will confuse “low numbers are better” for “high numbers are better”.

This sort of argument was one of the ones going around against AV (meaning instant run off voting in this instance) before Thursday's referendum. Basically that people are too stupid to understand a voting system that represents them more realistically. If you are unable to understand the instructions on the ballot paper then you probably haven't understood what the parties are actually standing for and are just doing what the most vocal newspaper tells you to do, in which case, maybe, just maybe, you shouldn't be voting in the first place. Politics is actually one those things where the simplest method isn't necessarily the best.
Mat wrote:What if I don't like any of them?

Revolution? :P
A bee sees Dee 'effigy' Hitch, I 'Jay' kill, amen OP, queue arresty UV double, you ex-wizard!

User avatar
Angua
Don't call her Delphine.
Posts: 5781
Joined: Tue Sep 16, 2008 12:42 pm UTC
Location: UK/[St. Kitts and] Nevis Occasionally, I migrate to the US for a bit

Re: Instant Runoff vs. Approval Voting

Postby Angua » Tue May 10, 2011 8:36 am UTC

Mat wrote:
Qaanol wrote:Approval voting is simple for voters, as you just need to vote for any candidates you like.

What if I don't like any of them?
That's what spoiling the ballot is for.
Crabtree's bludgeon: “no set of mutually inconsistent observations can exist for which some human intellect cannot conceive a coherent explanation, however complicated”
GNU Terry Pratchett

broken lader
Posts: 63
Joined: Mon Dec 13, 2010 4:11 am UTC

Re: Instant Runoff vs. Approval Voting

Postby broken lader » Tue May 10, 2011 10:24 pm UTC


User avatar
Qaanol
The Cheshirest Catamount
Posts: 3058
Joined: Sat May 09, 2009 11:55 pm UTC

Re: Instant Runoff vs. Approval Voting

Postby Qaanol » Wed May 11, 2011 3:15 am UTC

broken lader wrote:http://www.electology.org/approval-voting-vs-irv

Thanks for the link, that’s an interesting read. I’d need to know quite a bit more about the methodology and definition of the metrics utilized, but it seems to have reasonable results.

It appears (for obvious reasons) that Range Voting is even better than (its strict subset which is) Approval Voting in theory, but in practice having more options than “approve” and “don’t approve” strikes me as complicating things more than it helps.

I am somewhat surprised at the initial results of this poll here. I find Approval Voting extremely simple for both voting and tallying. You walk into the voting both, look at the list of candidates, and check off each one you’re okay with. No fussing around with ranking them in your head, no deciding how high of score each one gets, just a straightforward yes or no for each candidate. On the other end of things, there’s no danger of a spoiler candidate splitting votes away from someone who would otherwise win.

The way I see it, using approval voting most people would vote for the “major party” candidate they favor (if any) plus all the minor party candidates they favor more than that major party candidate. That way the major party candidates are a fallback sort of worst-case scenario, and you might possibly end up with an even better outcome if enough people are likeminded. In IRV, there’s still the “fear factor” where if you vote for your favorite minor party candidate first, that takes away a vote from the major party candidate in the first round, and however many subsequent rounds until your preferred candidates are gone, by which point it might be too late.
wee free kings

User avatar
Kulantan
Posts: 999
Joined: Mon May 04, 2009 9:24 pm UTC
Location: Somewhere witty

Re: Instant Runoff vs. Approval Voting

Postby Kulantan » Wed May 11, 2011 3:36 am UTC

Qaanol wrote:The way I see it, using approval voting most people would vote for the “major party” candidate they favor (if any) plus all the minor party candidates they favor more than that major party candidate. That way the major party candidates are a fallback sort of worst-case scenario, and you might possibly end up with an even better outcome if enough people are likeminded. In IRV, there’s still the “fear factor” where if you vote for your favorite minor party candidate first, that takes away a vote from the major party candidate in the first round, and however many subsequent rounds until your preferred candidates are gone, by which point it might be too late.

This is 100% not a problem. The only circumstance under which taking a vote away from your preferred major party candidate would cause a problem is when it would mean them getting knocked out in the first round. At this point they are totally not a major party with a viable chance of winning anyway. Basically, fuck it, if Labour got less primary votes than UKIP they were going to loose anyway.
TEAM SHIVAHN
Pretty much the best team ever

phlip wrote:(Scholars believe it is lost to time exactly which search engine Columbus preferred... though they are reasonably sure that he was an avid user of Apple Maps.)

Blog.

endolith
Posts: 229
Joined: Tue Jan 01, 2008 2:14 am UTC
Location: New York, NY
Contact:

Re: Instant Runoff vs. Approval Voting

Postby endolith » Wed Jan 03, 2018 1:19 am UTC

IRV is crap and the only reason people advocate it is because they haven't heard of any of the much better alternatives:

Image

Image

What arguments are there against ranked-choice voting?

STAR/SRV vs IRV

DavidSh
Posts: 133
Joined: Thu Feb 25, 2016 6:09 pm UTC

Re: Instant Runoff vs. Approval Voting

Postby DavidSh » Thu Jan 04, 2018 3:15 pm UTC

Intellectually, I like a method where
(1) Everybody ranks the options from best to worst, allowing for declaring subsets of options as equivalent, as: A, B, {C, D, E}, F, meaning A is preferred to anything else, B is preferred to anything besides A, no preference among C, D, and E, and finally, F is least preferred.
(2) If a candidate would win every head-to-head contest, it is the winner.
(3) I'm not sure about tie-breaking, or about resolving cycles. Do cycles actually show up much when people do rank-order voting?

However, this would be difficult to explain to voters, and require more work to implement than just counting how many buttons were pushed, levers turned, or holes were punched, for each candidate.

In voting for, for example, books, I have sometimes wanted to say "I really like book A. I don't have any preferences between books B, C, and D. I detest book E." Thus my desire for including subsets of equivalent options.

I think that first-past-the-post looks like this where everybody votes two sets: {a preferred singleton} and {everybody else}. Approval voting looks like this where everybody votes two more general sets: {approved options} {disapproved options}.

endolith
Posts: 229
Joined: Tue Jan 01, 2008 2:14 am UTC
Location: New York, NY
Contact:

Re: Instant Runoff vs. Approval Voting

Postby endolith » Thu Jan 04, 2018 6:10 pm UTC

DavidSh wrote:(2) If a candidate would win every head-to-head contest, it is the winner.


That's Condorcet's criterion, which a dozen different voting methods meet (IRV doesn't)

DavidSh wrote:allowing for declaring subsets of options as equivalent, as: A, B, {C, D, E}, F


Some Condorcet methods allow this, such as Schulze. Many ranked methods don't, though.

Do cycles actually show up much when people do rank-order voting?


Yes. It doesn't happen if voters are 1-dimensional (mindlessly left-right political spectrum), but does if there are 2 or more preference dimensions that aren't strongly correlated with each other (Nolan chart's "economic freedom" vs "personal freedom", for instance).

https://en.wikipedia.org/wiki/Condorcet_paradox

How often do Condorcet Cycles occur? says "In computer simulations of random elections, top-cycles happen ... quite commonly with 20 candidates (68% of the time)."

However, this would be difficult to explain to voters, and require more work to implement than just counting how many buttons were pushed, levers turned, or holes were punched, for each candidate.


Yeah, try explaining Schulze's computation method to the average Joe: https://en.wikipedia.org/wiki/Schulze_m ... omputation

But ranked voting methods are fundamentally flawed anyway, in that they assign equal weight to every preference. Rated voting methods collect more information from voters, since they can express both indifference and strength of preference.


Return to “General”

Who is online

Users browsing this forum: No registered users and 28 guests