In my town, we elect our mayor with ranked-choice voting. We don’t bother with caucuses or primaries. I think it is a good thing. It alllows the election results to more closely match what the people want. It prevents fiascos like what happened in 2000. I think this should be used for the presidential election, instead of wasting time and money on months of primaries.
What town is this?
If all the candidates were still in it at the time of the general election, they could conceivably have spent more money.
And I don’t see how it prevent the 2000 situation. That was just a close result.
Portland, Maine.
The idea with ranked choice (sometimes called instant runoff voting) is that the lowest vote-getter has their votes distributed to the voters’ second choice on the ballot, and then the next lowest and so on until you have a winner. In 2000, that would have been relevant because presumably most Nader voters would have ranked Gore higher than Bush and so when Nader was eliminated those votes would have gone to Gore, giving him a comfortable win.
Although of course the major logistical issue with the 2000 election was interpreting the intentions of voters on ambiguous ballots, and I can’t imagine a ballot that looks like this would have helped that situation.
Yes it would have solved the specific 2000 situation, but it wouldn’t solve the general situation of the last two candidates getting with .1% of each other. That would still happen just as frequently.
But it doesn’t have to look like that. It can look like this.
That might be even worse, since (AIUI) that’s expecting the voter to fill the thing out legibly and correctly. I’d imagine lots of potential for things like 1’s that look like 7’s or voters accidentally marking two 2nd choices, which wouldn’t matter most years but could make close recounts tricky.
Sure for the problem of very close elections it doesn’t do anything (and perhaps makes things worse by making more complex ballots and tabulation procedures.) But a lot of people view the issue of more extreme third parties pushing the election results in the opposite direction to be a systemic problem. Nader arguably costing Gore the election in 2000 is maybe the most prominent example, but there’s others (Bush and Perot in '92 for example.)
Although problem IRV fixes (that’s perhaps of some relevance to the current Republican primary race) is when you get a situation where a fringe candidate wins with a relatively low percentage of the vote because multiple mainstream candidates are splitting the vote too thin.
:smack:
Australia has managed just fine with preferential voting for decades now. We also allow you to just put a single 1 or x in your first choice the rest of the preferences are allocated according to the pre published order from that candidate / party.
The main advantage is you can vote for third parties without wasting your vote, eg you can vote green party first, democrats second, republican third. It makes third parties a lot more viable.
The system does depend on the voter being sufficiently numerate to indicate his or her preferences legibly. But other countries manage this with low rates of vote spoilage, and I struggle to think that it would be beyond many American voters.
And it certainly wouldn’t be “even worse” than the Portland ballot. A voter who struggles to number a list sequentially would have to be baffled by the matrix offered in that ballot. Surely far more people will accidentally punch two holes in the “third choice” column than will accidentally write the number ‘3’ twice?
If levels of support for two or more candidates are very close, preferential voting will do nothing to change this. But I don’t see that that can be characterised as a “problem” of very close elections. If in fact the electorate is more or less evenly divided between two candidates, shouldn’t the voting results reflect that?
The difference that preferential voting could make in this situation is that it tends to award victory goes to the candidate with the broader appeal. If the two front-runners are more-or-less evenly matched on the first count, and neither of them has secured a majority of the votes, victory will go not to whichever of them got more first preferences, but to whichever of them is more acceptable to the remaining voters who didn’t, in the first round, vote for either of them.
And the same feature tends to diminish unrepresentative results caused by vote-splitting. If there are (say) two centre-right candidates, who respectively get 35% and 25% of the vote, and a centre-left candidate who gets the remaining 40%, under the current system the centre-left candidate will be victorious even though a substantial majority of the voters support the centre-right, and would prefer another representative. Preferential voting tends to minimise this.
I agree filling out either of those isn’t rocket science, but with large amounts of people doing even a very simple task there’s always going to be a certain number of random mistakes, and increasing the complexity of that task is going to lead to more of those random mistakes. Even with the relatively simple ballot in Florida in 2000 there were enough of those random mistakes that for a while it seemed like they’d be decisive.
The situation I see the Australian handwritten ballot being worse than the Mainiac computer-read ballot is that in the case of a recount it’s easy to make consistent rules for things like overvotes but if you get a situation like Florida in 2000 where armies of lawyers appear to contest every ballot, the potential for bickering over ambiguous handwriting in endless.
I don’t think it’s a fatal flaw with preferential voting or anything like that, it’s just that in that particular very unusual situation it wouldn’t have helped or might have made things worse.
Well, they do face this already in countries where the voters number candidates, and it hasn’t cause the sky to fall. You need a couple of solid criteria for determining the status of disputed ballots (e.g. “Do the marks on the ballot express a clear preference?”) plus a clear process for deciding those rules (the returning officer hears submissions from the candidate’s agents, and then decides whether he considers the voter’s preference to be clear). A court won’t overturn a returning officer’s finding on this point merely because the judge would have made a different finding; it’s the returning officer’s job to decide whether the voter’s preference is clear, not the judge’s. The court’s role is to determine whether the returning officer went about doing his job properly, which is a different question.
This doesn’t prevent court challenges where the election is close - as long as the stakes are high and the margin is close you’ll always have the possiblity of court challenges - but, in countries where they have this system, court challenges are pretty rare. (It helps that the the election is conducted not by party loyalists but by professional public servants, and if you want to minimize recourse to the courts maybe that is a feature of the US system that you would look at, before looking at ballot paper design.)
Plus, of course, we know from the Florida case that the existing system attracts court challenges when the outcome is close. I’m not seeing much reason to think that a preferential system would attract more court challenges. And, even if it would, is that really the point? The advantage of the preferential system is that it gives the voters more say, and produces an outcome more in line with the voters’ wishes. Yes, the count will take a bit longer, and may cost more to run, but so what? And, yes, there will be further delay and expense if there are more court challenges but, again, so what? Are these not prices worth paying in order to have a more effectively democratic voting system?
Do you not understand the difference between a solution that solves all future instances of a problem, and one that would have solved one specific instance of a problem, but will never do so again?
Do you not understand clearly contradicting yourself within 3 or so posts without even pretending to acknowledge your error? The words are pretty clear. You were talking about a specific situation in both posts. Hey, it happens. But the solution might be to be a little less strident in your assertions the next time.
More importantly though, you have not at all demonstrated that this solution would not work in other cases.
I actually think Trump would benefit from this.
I am definitely a proponent of some sort of system whereby the voter is not stuck balancing the chances of a candidate winning against his like or contempt of that candidate. I’d rather just see voters choose who they like best rather than, as it seems in many cases, choosing the least evil. For things like congress I’m a fan of something like STV, but for seats that must be exactly one person, like president, governor, mayor, etc. I like the idea of instant run-off.
The problem I see here is, as others pointed out, in the case of the scantron form, issues with possibly confusing voters. Hell, if the butterfly ballot was an issue in 2000, that image in the fifth post is an issue. In that case, I like the other one better for clarity, but then you have the issue of how clearly and legibly people can write and potential court challenges. So, I know there’s been issues with electronic voting in the past, but it seems like that is how making STV and instant runoff voting work. We can have a simply interface like the latter, but also have it clear in the voter’s intention by them pressing keys or buttons and confirming their votes.
Obviously, that brings up an issue with how do we implement a solution that won’t be messed with and we’re sure is fair, but as a computer scientist myself, that seems like a problem that we ought to be able to solve. We have the technology to create open source code and digitally sign it so that we know it’s not been altered since it’s gone through various sources of approval. Hell, that sort of technology helps to address other probablems with voting raised by both conservatives and liberals too. It helps with accessibility since it could be done remotely, from a mobile voting unit or even over the internet. It fixes concerns with voter fraud since, presumably, a solution would mean providing voters with some sort of card with an encryption key to digitally sign their vote. It also makes “recounts” simple and consistent since it not only is all digital, but voters can even “verify” their votes by receiving digitally signed copies of their votes as proof of their intentions, etc. Frankly, it baffles me that with all the technology we have we’re still using paper ballots in so many places.
No, I was not talking about a specific situation in both posts. I was talking about a general situation in the first post.
Because maths. Or do you believe that a rank choice voting system is less likely, for some reason, to produce a dead heat?
I agree 100% with the OP; this one of those things that seems like such an obviously good idea it’s a wonder we haven’t implemented it already.
You should try refining it further to Approval Voting.