I just learned about approval voting and it has my...approval.
Jokes aside, "upvoting" all the candidates I approve seems much more doable than ranking the candidates in order of merit, especially given that there are a lot of candidates running.
Nah, that's how the people who want to change the voting system kill time. People have their preferences, but they're fairly united in thinking most systems would be an improvement.
I like Approval voting for single positions that are supposed to represent a community rather than a faction: mayor, governor, county commissioner, etc.
Mixed member for legislative bodies and Ranked Choice for anything else.
Approval feels a bit too binary to me. I like the idea of scoring or ranking. Not sure how scoring doesn't get gamed by people giving 10's to the ones they like, ad 1's to the ones they don't, though.
IMO "gamed" is a bit of a strange choice of word. The whole point of score voting is to give each voter more power to express their true preference. If their preference is to assign top scores to several candidates, fantastic, thank you for your vote! If the next voter prefers to give more diversified scores, wonderful, thank you for voting! And if the next voter would prefer to rank order the candidates, you can do that too using score, so great, thank you for voting!
Yes, if you are highly strategic and acutely fear some popular candidate(s), you might give maximum scores to all candidates of which you approve to minimize loss risk. But again, great. That doesn't seem like "gaming" things to me.
With score, voters leave the ballot box feeling they were able to express their preferences and therefore have less regret about their votes. Incidentally, I feel in the long term, approval or score voting would do more to improve the satisfaction people have with voting thereby increasing participation than all of today's "get out the vote" drives have had.
Over time, the people who use middling scores might 'catch on' that doing that lessens their influence, so everyone starts gravitating towards the extremes?
It's not gaming the system, range voting is robust to such an expression of voter preferences. However, there is something to be said for simplifying the system if that ends up being the way that most people vote.
My first election reform activism was for IRV. It is mathematically better than approval voting.
However, my election integrity concerns came to outweigh the fairness considerations. Tabulating IRV is a pain, requires computers to be feasible, and is much harder to audit.
Happily, approval voting get us most of the fairness of IRV and greatly improves election integrity over both FPTP and IRV (for different reasons).
Surely one should be able to give a negative vote to somebody they didn't like?
That would also solve the issue with polarizing candidates, since by definition you now have to care that those who would never vote for you, would at least not vote against you.
I used to be an ordinal-method purist, but I realized that approval voting is still (much) better than FPTP mathematically, and is also a win in terms of usability and understandability versus most ordinal systems.
Usability of a ballot system is a big factor when it comes to voter engagement and turnout, so at this point, I would be very happy with approval-method ballots.
That's a bold assumption. I think that numbers have less cultural understanding needed. Letter grades are common, but not ubiquitous. Some schools just use % based "grades" and the 4.0 GPA system.
If you say "Each candidate can be given up to 5 points, or as few as 0 points. There are no limits to the number of points you can give out in total (there are X candidates, so you could give a max of [X*5] points)." you're being pretty succinct and direct in the explanation. You might need more poll workers the first few years (which will create all kinds of problems because areas attempting to minimize voter turn out can exploit that) but once it becomes common you hope that people are used to it and know how the system works
Yep, STAR voting is generally my preferred method, but I do love the simplicity of "in favor of" binary voting as well. Either way, there's loads of improvements over our current system.
The biggest thing is we need to eliminate the need to hold your nose and vote for the lesser of 2 evils. Obviously all republics involve some level of "this candidate is close enough to be my proxy" decision making, but I'm currently stuck voting for someone who was indicted while in office for political crimes because the alternative is worse.
Moving to either system helps break the 2 party system open...which is why nobody in establishment politics is making a big push to get it implemented.
Score and approval have inconsistent meaning for ballot markings, and are for that reasons bad mechanisms unless that is resolved (approval is ideal for choosing group activities where either voting approve is a commitment to participate if that option wins or disapprove is a binding opt-out.)
I had this idea several years ago, but didn't really know how to talk about it. Essentially, if voters can attribute a value (between 0 and 10, let's say) to every candidate they like, then we can see a better picture of who the country likes.
For instance if I want Candidate A to win, I'll give him a 10. But if I'm only half bought into candidate B, I can give them 5, and could give Candidate C nothing because I really don't like them. Now if the rest of the country doesn't like A, and the battle is between B and C, then I'd feel good that I could contribute to the goal.
The actual process is up for remix of course, but that was the idea.
Maybe you realize this, but you just described score voting. Check the links I posted above. Welcome to the fan club!
(If you're not familiar, approval voting is a special case of score voting where you are giving either a 0 or 1 to each candidate; in other words the range is binary. It's not quite as expressive for voters as score voting, but still a huge improvement over what we use now and pretty low on "voter regret" since it avoids having to vote strategically. And it's arguably easier to use and understand than score voting. I would be so happy with either score or approval.)
Dan Carlin (hardcore history - which is awesome btw, and common sense) mentioned a couple of ideas that sounded interesting. I’m sure he’s not the inventor of the concepts and I have no idea how they’d actually work, but sounded similar to what you’re talking about.
One would work where you have 1 vote per ballot item. If there were 24 ballot items, you get 24 votes. You can allocate your votes across the ballot however you saw fit. If you thought president and senate were more important than city council and an infrastructure bond; you could put 18 votes for president, 6 votes for senate, and nothing for local issues. No idea how this would be normalized but it’s a novel concept.
Another idea is to rank options in each issue, and if my first option loses; my votes go to the second option. Say I wanted to vote libertarian for president in the US but was afraid it was a protest vote that wouldn’t win so I voted Democrat because I didn’t like the GOP candidate. Under this system I could say I want the libertarian candidate first, democrat second, and not pick the gop candidate.
After the votes were counted and the libertarian lost, my vote would be reallocated to the democrat; and the votes would be re-tallied. This would encourage people to vote for a non binary political candidate since they knew their vote wouldn’t be “thrown away”.
> No idea how this would be normalized but it’s a novel concept.
interesting concept indeed, but i think the normalization function would be a serious vulnerability. most state and national level elections in the US tend to be pretty close, so the normalization function would pretty much end up deciding the winner. people who realize this would fight hard to get a biased function implemented, and you would be stuck trying to explain statistics to the general public. basically the gerrymandering situation, but even harder for the public to understand.
..One would work where you have 1 vote per ballot item.
That seems extremely problematic to have votes determined by ballot items. It's impossible to normalize it. Even if you just determined that "each voter nationwide gets X voting points" you'd have politicians that would try to dilute or concentrate their ballots in order to game things.
How would this work across different voting districts? It's highly improbable that all districts have the same number of ballot items. Seems that if it were to work the way as you describe, some districts would get more votes than others. I could see districts gaming this by putting up BS ballot initiates to stuff the vote.
The voting system sounds reasonable, except local, state and national elections are much more independent than you're describing. The Constitution specifically and purposefully separates these entities, and trying to sew the legislation together would be more of a nightmare than it's worth.
While I immensely agree with these approaches, it is unlikely to change with the current incumbents (as they have perfected strategies for the current system).
Hey David, I suggest you take a look into Range Voting (Approval Voting being a case of 2-Range Voting), Score Then Automatic Runoff (STAR), and 3-2-1 Voting (V321).
Additionally Reweighted Range Voting as a very good Proportional Representative method.
Nope for the US getting rid of 18th century style primaries move to OMOV One Member One Vote for selecting candidates.
This would freeze out the fringe candidates and also save a metric FT of $.
Having the national Dem and Rep party get a fracking grip on the conventions and bring the rules into the 21st century would help, I am sure Tony Blair or George Brown would be able to give insights into fighting militant.
Ranked Choice voting would generally be a good move in that direction.
Edit I'm also going to add that this isn't a great time to sit on the fence:
https://www.economist.com/leaders/2018/11/03/why-the-mid-ter...
https://www.usatoday.com/story/opinion/2018/11/04/election-r...