Guest Rant Oct 5, 2022 at 10:17 am

Support Ranked-Choice Voting This November

Ranked-Choice Voting vs Approval Voting boxing on the ballot this November MIKEY BURTON

Comments

2

"introducing a system that works in favor of bland, uninspiring candidates rather than in favor of the preferences of the majority of voters."

Bruce Harrell, Jenny Durkan, Ed Murray, and Greg Nickels come to mind.

5

"Voters in Seattle will have two democracy-reform initiatives to choose from this November: A scheme for ‘approval voting’ bankrolled by local tech executives and a small number of wealthy out-of-state donors, and the City Council’s alternative proposal for 'ranked-choice voting' (RCV), which is supported by a broad coalition of Seattle-based community groups, seven city council members, and road-tested in cities across the country."

Roughly half of the contributions to RCV were made by renewable energy mogul Robert Poore and his spouse.*

The authors must know this.

*https://web6.seattle.gov/ethics/elections/campaigns.aspx?cycle=2022&type=campaign&IDNum=837&leftmenu=expanded#PopUpContributors

6

“But even after months of campaigning, their effort has failed to garner any meaningful community support.”

Apparently, actually getting enough signatures to qualify for the ballot does not qualify as “meaningful community support,” whilst getting onto the ballot purely via Council fiat does, because whisky monkey banana telephone SHUT UP, THAT’S WHY — hey, look over there — a rich guy!

“You won’t find any endorsements from local community groups on their campaign website. That's because their campaign doesn’t have any.”

What?!? No screaming, self-appointed activists have endorsed Approval Voting?

“Rian Watt works for a homeless nonprofit
”

RCV will fix Seattle’s voting problems just as well as Seattle’s homeless nonprofits have ended homelessness!

“RCV, by contrast, is part of a broader package of democracy reform supported by a comprehensive coalition of community groups who’ve been on the ground doing the work for years.”

Their many years of hard work were handsomely rewarded when they delivered enough signatures to put RCV on the ballot! (And every other democracy reform they support will do equally as well,)

“There are real, community-based, multi-racial coalitions that have pushed to get tried-and-true reforms like ranked-choice voting on the table”

And their great success at collecting enough signatures to get onto the ballot has made Seattle’s homeless nonprofits deeply envious.

“Voters should follow City Council’s lead when November ballots arrive
”

Just as they did with last year’s November ballots!

7

Still remains to be seen whether RCV is a fairer process and if it is worth the effort and complexity - nobody likes the voting outcome of Seattle races but there has never been a gripe that it wasn’t transparent or above board. RCV is confusing and cumbersome to implement. Voters already struggling with checking one box.

9

It comes down to if you have a vote between wonderful guy A who is not expected to win, lizard Y, and lizard Z. Lizard Y is better than lizard Z, but both are worse than wonderful guy A.

Our current system (vote for one) tends to elect Z because most people prefer Y or Z, and the people who prefer Y or Z over A find Y kind of bland.

With approval voting you'd vote for A and Y. You'd only vote for just A if you think A is one of the top two in popularity. Y tends to win. As the article points out, it's because Y is a compromise candidate, kind of bland. Some people who prefer lizards Y or Z actually prefer A over the other lizard, so would vote for both A and Y or both A and Z, so A would get some votes from them, and sometimes A would win instead of Y.

With ranked choice, you'd rank Z last. The honest vote is A Y Z, but you could vote Y A Z. Ranked choice works by repeatedly throwing out the candidate with the least first-choice votes, and the Y A Z pattern prevents Y from being thrown out early. Voting A Y Z tends to make Z win because Y is thrown out early and more people prefer Z than A. Voting Y A Z tends to make Y win, but makes it impossible for A to win because people who preferred A had insincerely ranked Y higher. A was unlikely to win anyhow. So voters who prefer A tend to vote Y A Z instead of A Y Z. This has been seen in practice: Australia (which has used ranked choice forever) has a stable two-party system.

So both ranked choice and approval voting would usually make Y win, which is an improvement over our current system that makes Z win. But approval voting lets A win much more often than ranked choice would. If you don't like having a two-party system, approval voting is better.

10

"RCV goes further by giving all voters, including those historically most excluded from our democracy—young voters, working class voters, voters of color—a more robust set of tools to express their preferences through their ballots.'

Bullshit! This isn't just ranked choice voting, but it is instant-runoff ranked-choice-voting. What this means is that your rank is meaningless unless your first choice does not advance. In this case, since we have an open primary (where the top two candidates advance, regardless of party) it means that ranked-choice is meaningless in a three-person race.

In contrast, approval voting always matters in a three person race.

Consider a real-world example: the last race for Seattle City Attorney. Only three people were on the ballot. Thus ranked-choice-voting (with instant-runoff) would not have changed anything. The third place candidate (Holmes) came in third, and where his votes went was meaningless. You would have had the same two candidates running in the general election.

In contrast, it is quite likely that many of the people who voted for either NTK or Davison approved of Holmes. Maybe he wasn't the first choice, but he would have been the second. It is unlikely that very few NTK voters approved of Davison or vice-versa. Thus Holmes would have advanced. It is also quite likely that Holmes would have won the election in a head-to-head against either candidate.

It is also quite likely the same thing would have happened in the Seattle City Council District 9 race. Thomas was the consensus candidate, but she lost to the more extreme candidates in the primary, even though she probably could have beat either one in a head-to-head race.

The whole point of an advanced voting system is to come up with a consensus candidate. In both of these cases, instant-runoff would not have changed the outcome. There are ranked-choice-voting systems that do a better job of this, but those aren't being considered. Given the choice of instant-runoff or approval voting, approval voting is much better.

11

I should mention that both systems solve voting problems. But in our system (which has an open, non-partisan primary) approval-voting solves those problems more often. Instant-runoff can't solve the problem I mentioned -- a consensus candidate has a minority of the first choice votes. I gave two examples of progressive candidates that didn't advance because of our current system, and instant-runoff would not have changed the outcome.

But there are situations where instant-runoff would change the outcome. In our case, you need at least four candidates running. Some combination of candidates could split the vote, thus allowing other candidates to advance. The thing is, with approval voting, you are likely to get the same outcome. This is easier to understand with an example:

Let's say you have four candidates running. One Democrat, one Republican, and two Socialists. Let's say the vote goes like this: Democrat: 35%, Republican 25%, Socialist A: 22%, Socialist B: 18%. With our current system, the Democrat and Republican advance, which is not representative of voter intent. Instant-runoff fixes the problem. Socialist B gets all of Socialist A's votes, putting that candidate into the lead. At that point, the race is over, and it becomes a race between a Socialist and Democrat (what the people want).

But with approval voting, the same thing would happen. Many Democrats would also approve of the Socialist. Even if the don't, there are enough Socialists who approve of the other Socialist to put one of them ahead of the Republican. The only way the Republican advances is if some of the voters approve of them -- which seems unlikely.

Thus approval voting often solves the exact problem instant-runoff is designed to solve, while also solving a problem that instant-runoff can't.

12

I should mention that both systems solve voting problems. But in our system (which has an open, non-partisan primary) approval-voting solves those problems more often. Instant-runoff can't solve the problem I mentioned -- a consensus candidate has a minority of the first choice votes. I gave two examples of progressive candidates that didn't advance because of our current system, and instant-runoff would not have changed the outcome.

But there are situations where instant-runoff would change the outcome. In our case, you need at least four candidates running. Some combination of candidates could split the vote, thus allowing other candidates to advance. The thing is, with approval voting, you are likely to get the same outcome. This is easier to understand with an example:

Let's say you have four candidates running. One Democrat, one Republican, and two Socialists. Let's say the vote goes like this: Democrat: 35%, Republican 25%, Socialist A: 22%, Socialist B: 18%. With our current system, the Democrat and Republican advance, which is not representative of voter intent. Instant-runoff fixes the problem. Socialist B gets all of Socialist A's votes, putting that candidate into the lead. At that point, the race is over, and it becomes a race between a Socialist and Democrat (what the people want).

But with approval voting, the same thing would happen. Many Democrats would also approve of the Socialist. Even if the don't, there are enough Socialists who approve of the other Socialist to put one of them ahead of the Republican. The only way the Republican advances is if some of the voters approve of that candidate -- which seems unlikely.

Thus approval voting often solves the exact problem instant-runoff is designed to solve, while also solving a problem that instant-runoff can't.

13

@10 I'm not convinced that a sufficient number of NTK voters would have also voted for Holmes (and vice versa) under an AV system to avoid a Davison victory. It's possible, but it shouldn't be presumed and it seems more likely to me that Holmes would have done better with RCV, where voters for both of the progressive candidates (who made quite different rhetorical pitches, despite their essential similarities) could signal their preference for one over the other.

Nonetheless, given everything else we're dealing with, this entire debate strikes me as a rearranging of the proverbial deck chairs. While I agree with the author that either of these proposed systems would likely be an improvement over the status quo, parsing out which one is better just doesn't seem to be a good use of our time and energy right now. Perhaps the one to vote for is whichever one happens to be polling better when you cast your ballot.

15

@14: The same thing can happen with Approval Voting in a three-candidate primary. Abolitionists approve the abolitionist and Republican candidates, because an Abolitionist is expected to lose to a Democrat in the general election. Republicans do the same. General election then becomes Abolitionist vs. Republican, which is what happened last year anyway.

@10: Your presumption that Holmes would have made it through to the general election seems unsupported by fact. In addition to the theoretical scenario I just described, Holmes had implemented a homeless policy — allow camping everywhere! — that voters had already rejected in 2015, when the City Council had proposed a law to alllow camping citywide. The Council beat a quick retreat in the face of public outcry, then Holmes de facto imposed this rejected policy anyway, via the simple expedient of not enforcing laws against camping. The results have been nothing short of disastrous, and he properly lost his job over this.

A four-term incumbent losing the primary is an extremely rare result, and tweaking the voting system would not have altered voters’ desire to punish an office-holder who had persistently refused to do his job.

16

Hopefully in November Seattle voters will choose to reject the currently used plurality (winner-take-all) voting and vote to adopt one of these two reforms. Both are much better than plurality voting - not exactly a high bar.

One can, I suppose, debate the pros and cons of ranked-choice voting vs approval voting as this article does quite well. But there can be no debate that one of these reforms has a long well-documented track record, and the other one does not.

Earlier this year when the "Seattle Approves" folks were collecting signatures to put their initiative on November's ballot, I wondered whether perhaps a few potential signers may have thought they were supporting ranked-choice voting. Did they understand the difference between ranked-choice voting and approval voting? Did they even know there was a difference? Surely the "Seattle Approves" signature gatherers tried to make this difference clear.

17

@9, I was on the fence about this until I read this. Thank you for explaining the meta like that. My only criticism is that it's a little hard to follow -- maybe some diagrams would help. My bet is that if we go to instant-runoff RCV, we'll get one or two elections where people vote honestly, before it all settles into the state you describe. Maybe it'll happen immediately if the PACs all figure out the strategy and tell their constituents what to do right away.

The problem with any of these initiatives is that if you've looked into them at all, you've probably come across the Condorcet method, and know the flaws of each of the alternatives, so they all look bad. And the imperfections of each method is a perfect place for politicians to attribute malice. "RCV supporters are just a bunch of grandstanding activists who don't understand game theory", "Approval Voting supporters just want to see more status quo candidates and want socialists off the ballot for good," "Condorcet voting is just a bunch of ivory tower pinheads trying to solve a problem that doesn't exist," etc.

Yes, the Seattle Approves campaign is headed by a white urbanist who really doesn't like Kshama Sawant, but it's not done out of a desire to co-opt the organizing efforts of the left. Remember, Seattle Approves was there first -- the City Council put RCV on the ballot in response to it. Both campaigns come out of a sincere desire to fix a problem, and both campaigns reflect the flaws of their champions.

The term "both sides" has become toxic these days because of false controversies between normal folks and literal Nazis, but this is definitely a case where both sides have a solution that is better than the status quo, and I really hope we can all get behind one of them.

18

1A for Approval voting is the clearly superior and simpler option here.

1B is a novel form of ranked choice voting called " bottoms up", that has never been used in a history of the United States.

Bottoms up is extremely bad for this purpose because it gives all the power to primary voters, who skew older, wealthier, whiter, etc. Tends to advance the strongest overall candidate against the strongest opposition candidate, who will by definition be weak in the general election, leading to an uncompetitive election in which the consensus choice of the primary voters wins by a large margin. A show trial.

Whereas approval voting will advance the two most popular candidates overall, meaning they will be fairly similar, making it plausible for the underdog to win the general by appealing to a sufficiently broad swath of the general electorate.

The rank choice voting proposal superficially appears more democratic because it provides the appearance of a diversity of choices. But that's an illusion given that the opposition candidate will virtually always be a heavy underdog with almost no chance to win.

We saw the effect of approval voting in the 2021 mayoral race in St Louis, in which the two most progressive candidates, also both the two women, advanced to the general. While that may intuitively seem unfair to moderate to conservative voters, it means those voters actually had a stronger voice because the slightly more moderate of those two progressives was viable in the general election. Even though the more progressive won, the slightly more moderate underdog wasn't far behind. Both candidates had to fight for every vote.

With the rank choice voting bottoms up process, the runner-up would have likely been an establishment Democrat or even a Republican, who would have been crushed by a mile. You might like that as a progressive, but it could easily go the other way. A competitive general election is a good thing that makes elections more beholden to voters than special interests and money.

19

If millionaires and billionaires are for it, I'm against it.

20

"RCV goes further by giving all voters, including those historically most excluded from our democracy—young voters, working class voters, voters of color—a more robust set of tools to express their preferences through their ballots."

But does it? It can give voters, all voters, not just the aforementioned, the warm and fuzzy feeling that one of their choices won. Even if that person was way down on their list. Approval voting does something similar from the voters point of view. "Hey! Someone I checked won. I feel great." But RCV is more effective at undermining candidates' cults of personality and encouraging coalitions. Particularly for parties whose platform is made up of more diverse issues. Like the GOP. One can be pro-2A, tough on crime, low taxes, less regulation and find a number of suitable candidates. Prioritize them as you like. Sorry about those other issues (abortion perhaps). But at least I got someone who supports much (but perhaps not all) of the issues that I like. My Second Amendment candidate doesn't have to fight the low tax one.

RCV (and to an extent AV) will kill off the "trust the party" campaigns. And the "trust Donald" or "trust Hillary" ones as well.


Please wait...

Comments are closed.

Commenting on this item is available only to members of the site. You can sign in here or create an account here.


Add a comment
Preview

By posting this comment, you are agreeing to our Terms of Use.