It's annoying that the election reform movement is so segmented when STAR (score than automatic runoff) voting is objectively superior to the more commonly proposed alternatives. First past the post voting (what we have now) sucks, but ranked-choice voting is significantly more complicated while only being marginally better.


RVC is how New York city got Eric Adams as mayor, much to the chagrin of the progressives. Maybe next time the hard decisions about who will run will be made in a smoke filled room somewhere.

Here in WA, we just got rid of the caucus system which was not something I ever felt inclined to want to deal with. Let's see how the vote by mail primary system shakes out with likely more voter participation before even more restructuring of the system.


Here is quick comparison of potential voting methods:

As I said above, I think STAR is pretty clearly the best option, but approval voting is also arguably better than ranked choice.


Too late, this woman already signed! It’s not that we don’t have diverse candidates or office holders (meet the female majority of City Council and our last mayor, police chief, department heads). It’s the spoiler factor. Our primary dogpile system leaves us with gems like Cary Moon, Mike Mallahan, and overlooks the talents of the likes of Jessyn Farrell or Colleen Echohawk. Seattle could stand to have more organized primaries lol


"Sardinas stressed the simplicity of Approval voting. She said it would be much easier to explain to her immigrant grandparents."

I have to admit, even after reading Hannah's post twice, I have no idea what distinguishes approval voting from ranked choice voting.


@2, this would only apply to City elections. Not federal elections; we never did caucus for Mayor or City Council seats.


"The League of Women Voters of Seattle-King County (LWVSKC) does not want you to sign Initiative 134, a ballot measure to bring Approval voting to Seattle elections."

People who want election reform do not want YOU signing for election reform!

"While signature gatherers for the measure have earned a reputation for advertising Approval voting as an election system similar to Ranked-choice voting (RCV)..."

That's because it is. Approval voting is a simpler form of RCV. (Maybe we want to try a simpler version before attempting a more complex model?)

"’s worth holding out for RCV when it comes to selecting a system that would lead to gender parity and diverse representation more broadly."

WIll RCV actually "...lead to gender parity and diverse representation more broadly"? How do we know? Is RCV actually so much better than Approval Voting that it's worth the wait? (And, again, why can't we implement Approval Voting en route to RCV? If we can change our voting system once, we can change it again, correct?)

"...Approval voting does not express preference for one candidate relative to another."

Yes, it does. A voter either votes for a candidate or not. The candidate(s) each voter approves are ranked higher than the non-approved candidates. (That's why it's called "Approval Voting".)

'In the same vein, Kelly said “popular” candidates could equate to candidates we are used to seeing in power: White men.'

Any argument which depends upon Seattle's voters behaving in a racist and/or sexist manner is an argument which will fail. For example, in Seattle's 2017 Mayoral Primary Election, the top four vote-getters were all women, and the top three of those women had never previously held elective office. Back in the '90s, there was a period when the nine City Council seats and the Mayor's office together held a grand total of zero white guys. There's simply no evidence of racism or sexism within Seattle's electorate. If voter sexism is the best argument for RCV over Approval Voting, then the arguments for RCV are very weak indeed.

"In a city whose council has exceeded gender parity in seat distribution," the argument for RCV is that it will "lead to gender parity" means it will have Seattle electing FEWER women. That's actually the logical result of those two statements both being true, and thus it's an argument against changing Seattle's voting system at all (!).


@5 In ranked choice, you rank your candidates in order of preference. Once votes are collected, candidates are eliminated in rounds until a winner is determined. Each voter's vote goes to the remaining candidate who they have ranked the highest, and each round the candidate receiving the lowest number of votes is eliminated. Eventually one candidate will garner over 50% of the votes, and they will be the winner.

In approval voting each voter simply states whether they approve or disapprove of a candidate, the voting process very similar to the current system except you can pick as many candidates as you like. The candidate with the highest approval wins.


The main problem I see with approval voting is that if you do have a preferred primary candidate (which for me is about 95 percent of the time), voting for any additional candidates (i.e., "approving" of them) potentially hurts your candidate's chance of winning, meaning the smart thing to do in most cases under an approval voting system would be to vote for ONLY your #1 choice -- thus negating AV's main selling point. With ranked-choice voting, you can't possibly harm your preferred candidate's chances by ranking the others as a hedge against a truly bad outcome.

I won't argue that AV wouldn't be an improvement over our current system (as the St. Louis experience possibly illustrates), but I do take issue with the implied promise of its advocates that AV will completely eliminate the need for thoughtful voters to engage in some degree of strategic triangulation. No system can do that. But ranked choice voting definitely comes closer. I would not sign this petition.


When did making a decision become so difficult?

Both Ranked Choice and Approval voting seem like solutions in search of a problem.

Vote for the candidate you think is most qualified for the job and move on. What is so difficult about the system we have now?


@10 You're not entirely off-base with that sentiment. But the current primary system, in which a candidate can win despite being not only opposed but deeply disliked by a large majority of voters, does tend to empower "spoilers" (i.e., candidates who have no serious chance of winning a general election and whose presence in the race only improves the chances of their ideological opponents). I don't think it's unreasonable for people to suggest ways to limit such candidates' influence, which is what this debate is essentially about.


Obviously women in lakes with swords allocating elected officials is the preferred method, although coconuts is a close runner up.



Wouldn’t it be simpler to make it more difficult to get on the ballot?
Require perspective candidates to gather signatures equal to 10% of the last election total vote.
Low enough to not be onerous but high enough to show voter support.


@9 CKathes --

Extensive game theoretical analysis and even computer simulation shows that approval voting does in fact perform better than RCV, with any mixture of strategic or honest voters. See Bayesian regret figures here.

For example, Pete Holmes was almost certainly preferred by a majority to both Davison and Thomas-Kennedy, but didn't make the general due to vote splitting. RCV would have done the same thing—the notorious center squeeze effect. Whereas approval voting almost certainly would have sent Holmes to the general.



Vote for the candidate you think is most qualified for the job and move on. What is so difficult about the system we have now?

A majority thought Pete Holmes was better than Thomas-Kennedy. And a majority thought he was better than Davison. Yet he didn't make the general, due to the "vote splitting" or "spoiler" effect.

Surely you've heard of people voting for the more electable "lesser evil" rather than the person they really think is best. This makes the perception of electability as influential as, or sometimes even more so than, actual merit and preferences.


@9 You nailed it in one. And it would be nice if we would all stop this deflection and be honest about the debate. And "women and minorities" is deflection here.

It's this simple.
Present System: Spoiler effect. Voters pick between lesser evils in the primary often. Real world, consensus Pete Holmes squeezed in primary, majority stucker between (to them) two clinkers.
Approval: Drift to boring moderate consensus candidates: Real world postulation. Majority chooses Holmes by supermajority, clinkers bomb out.
RCV: Only system where more extreme positions may get in by mathematical quirk. Postulation: Lefties only vote NTK, some moderates vote 2nd, wingnuts only vote Davison. NTK wins.

This is why the Stranger contingent loves them some RCV. It's really the only possible system with wider acceptance that could allow their sort of candidate to have any shot. OTOH, approval makes their favorites even less likely than top two which is why they hate it.

The rest is distraction and deflection.



Primary Results
NTK = 71,367
AD = 64,179
PH = 60,093

General Election
AD = 106,235
NTK = 85,077

I don’t know where you learned to count but 60,093 is not a majority in either election. Pete Holmes lost the primary because a majority of people voted for another candidate.
NTK lost the General because a majority of people voted for Ann Davison.


@10 -- If you don't understand a subject, I suggest you read up on it, then offer your opinion. You obviously don't understand this subject (clearly evident by your first comment) and then you keep blathering on while people try to explain it you. This is bad enough, except you are now making contradictory statements when you clearly don't have a fucking clue. Just to be clear here -- it isn't that simple. You may have trouble understanding the more complex issues, and be too lazy or disinterested in learning them. But it isn't that hard to figure out the basics from readily available sources like the following: (covers the subject from a wide perspective) (a common form or ranked choice voting)

As it turns out, there is an excellent description of exactly the situation that occurred with the Seattle City Attorney race: (notice the chart). Basically "A" was NTK, "B" was Holmes, and "C" was "AD". Davidson was elected, even though Holmes would have beat her in a head-to-head race, just as he would have beat NTK. This is a flaw within the system, as it prevented the clear consensus candidate from representing the will of the voters.



This is a benefit, not a flaw. It's mathematically proven that no voting method can simultaneously satisfy HSC and FBC.

Suppose you prefer the "Green". With approval voting, you might tactically vote for the "Democrat", but then you'd also vote for the "Green". With RCV on the other hand, your best strategy is to tactical bury the "Green".

For example, Thomas-Kennedy voters would have been better to vote for Holmes in order to get their 2nd choice (Holmes) rather than their 3rd choice (Davison). RCV would have had the same problem. Approval voting would almost surely have elected the true majority winner, Holmes.


@18: You've assumed Holmes was the "clear consensus candidate," but as your own source notes, that needs to be determined empirically. Here is an argument against your assumption.

In the City Attorney's race, the incumbent Holmes was the status quo candidate: he had declined to enforce laws against camping. NTK and Davison represented two alternatives to this status quo. NTK represented an enhancement of it: no enforcement of most other laws, as well as not enforcing laws against camping. Davison represented a reversal of the status quo: enforce laws, including laws against camping.

In the primary election, the status quo (Holmes) lost exactly because most voters considered it to be unacceptable. (Feature, not flaw.) It was unacceptable to the voters who wanted less enforcement (NTK), and it was unacceptable to the voters who wanted more enforcement (Davison). Once the primary election had eliminated the status quo, voters in the general election faced a very clear choice, and chose enforcement.


@20: Holmes almost certainly was the consensus candidate. The vast majority of Thomas-Kennedy's progressive voters would have preferred Holmes to the Republican Davison. And of course the vast majority of Davison's voters would have preferred Holmes to prison and police abolitionist Thomas-Kennedy. No one seriously disputes this.


Here's Tishaura Jones, the first black female mayor (and only second female mayor ever) of St Louis, endorsing approval voting.

Here's Cori Bush endorsing approval voting.

When Fargo held the first approval voting election in history, one incumbent won, and the other (male) incumbent was ousted by a woman.

I'm not sure what the KCLWV is thinking here.


@18 What evidence, other than your own opinion, do you have that Pete would have beaten Ann in a head to head race? take NTK out of the equation and the entire dynamics of the race changes.

there was an election, your side lost, deal with it, we all have. Or are you one of those dolts who agrees with #45 that he actually won the 2020 election?


@20 Biden won by 7 million votes. Out of the 62 lawsuits filed challenging the presidential election, 61 failed.


@24 Actually Joe Biden won by 74 Votes, 306 to 232.


@21: "Holmes almost certainly was the consensus candidate."

Almost 70% of the voters in the primary chose someone else. Holmes was a "consensus" candidate, yes, but the voters' consensus said he should not continue in office.

"The vast majority of Thomas-Kennedy's progressive voters would have preferred Holmes to the Republican Davison."

That assumes those voters would have voted for a candidate other than NTK. She was an extremist, fringe candidate, running on a radical platform. What if some, most, or all of "her" voters would not have voted for anyone else? Why would a status quo incumbent get their votes? Your argument depends entirely upon an assumption you have not validated.

@18 & @21: How narrow can that central slice of the electorate become before your 'consensus' assumption loses validity? Assume NTK and Davison had each received 49% of the primary vote. Would you still claim that Holmes, an incumbent who'd received a whopping 2% of the vote, was still the "consensus candidate"? Because I don't see anything in your logic which limits how small that central slice of the spectrum can become. Please feel free to educate me on this point.



Almost 70% of the voters in the primary chose someone else.

This doesn't dispute anything I said. This is only a measure of first place support. We cannot know anything with absolute certainty, but there's high probability here.

And regardless of the specific example, the vote splitting effect does happen fairly often. You see it in both simulations and exit poll experiments where you ask people who they would have voted for have their favorite not run. Bottom line is that restricting people to only one vote is purely harmful.


@25: Yes everyone knows how the electoral college works. Thanks for admitting he won though.


@27: "...there's high probability here."

No, there is merely your assumption of a high probability. Again, you must provide support for your assumption, namely, that voters who chose the radical, extreme, first-time candidate (NTK) would also have supported the status quo, multi-term incumbent (Holmes). Absent that assumption, your argument for a different outcome from a different voting scheme simply fails.

"And regardless of the specific example,"

But in this specific example, a multi-term incumbent lost the primary election. That's an extraordinarily unusual event, so your claim that a change in voting method would have completely changed the outcome requires extraordinary support. Furthermore, in the resulting general election, Seattle chose a Republican for the first time in decades. Saying that a change in voting method would have produced a status quo election really does require a lot of evidence, which you have yet to provide.

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

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