Progressive groups promote Kim-Leno dual endorsement strategy

Forget the polls: The candidate who makes ranked-choice-voting work will be the next mayor

The chorus or progressive groups endorsing both Sup. Jane Kim and Mark Leno for mayor continues to get louder, with the Milk Club, the Sierra Club, the Tenants Union, and the Community Tenants Association holding a joint press event today to promote that ranked-choice-voting strategy.

The Milk Club, Sierra Club, Community Tenants Association, and Tenants Union all are promoting a dual-ensorement strategyalth

Deepa Varma, director of the Tenants Union, said that this race “is all about keeping people in the city. If we are going to stay here, we need a mayor who cares about tenants.”

She said that instead of splitting the vote, her organization had endorsed both Kim and Leno.

That was the theme of the event: For progressive groups, this race could have been seriously divisive with nasty fights over whether Kim or Leno should be the Number One pick. The dual-endorsement strategy has largely eliminated that.

Honey Mahogany, co-president of the Milk Club, said “I’m proud that San Francisco has such a diverse slate of candidates. It’s essential for we as progressives to work together.”

As far as the candidates go, however, Kim is a lot more enthusiastic about this strategy than Leno.

In a statement to the Bernal Heights Democratic Club, which asked if the candidates would endorse anyone else, Kim said: “I believe we should utilize rank choice voting and have committed to ranking Mark Leno in my ballot.”

Leno said he has not endorsed in the race and “is still considering options. There will be no consideration of any candidate who is not an early endorser of our Fair Campaign Pledge.”

That difference underlies the tricky politics of this race. In 2016, Leno supported Scott Wiener for state Senate – and Wiener and his allies launched vicious attacks on Kim. Wiener has endorsed Leno – and while Kim seems to be willing to let that be part of the past, Leno has a problem:

He needs the Wiener voters who were told to mistrust Kim. Will he alienate them if he now endorses her?

And if he doesn’t, will he alienate (or fail to inspire) the Kim voters to list him Number Two?

Is there any way these two candidates can work together the way their allies are?

That’s happening as pretty much everyone agrees this is a three-way race with none of the top candidates – Leno, Kim, and Sup. London Breed – pulling away. It’s almost certain that no candidate will get 50 percent of the vote in June; in fact, I’d be surprised if any candidate got 40 percent of the first-place votes. So this election will absolutely come down to second-place votes.

It will be the first truly dramatic test of the RCV strategy in a mayor’s race. And although some observers are saying Breed is pulling ahead, the narrow differences in the polls at this point mean nothing. Breed, Leno, and Kim will all be in the mix, all close, all in the top three.

So let’s just speculate here. Suppose Breed comes in first, by a few points, Leno comes in second, and Kim is third. The Kim second-place votes will decide the race: If enough are for Leno, he will pass Breed. If not, Breed will win. 

If Leno comes in first, Breed second, the same scenario applies.

If it’s Breed, Kim, Leno, then Leno’s second-place votes will elect the next mayor. 

It’s possible, but unlikely, that Leno and Kim will be first and second, and Breed third; in that case, her Number Two votes will decide the outcome. (I am assuming that Amy Farah Weiss and Angela Alioto will be far behind, and that Alioto’s second-place votes will be all over the place.)

You get the point. The candidate who figures out how to make RCV work will be the winner.


  1. How is Tim being unfair? His report looks fully objective to me. Please explain your reasoning to us.

  2. OK, I understand your point. I found this nifty diagram to illustrate the point.

    First of all, thanks for an interesting conversation.

    Any multiple-candidate voting system you can think of asks some questions but not others. Let’s take the 2011 elections as an example. A runoff system asks first, “Who do you prefer?”, and then, “Do you prefer Lee or Avalos?” The Chiu-Herrera-Dufty voter gets to say, “Who do you prefer”? “Other than Chiu, who do you prefer?” “Other than Chiu and Herrera, who do you prefer?”, which the runoff voter does not get to answer.

    One can come up with other systems, all of which would ask different questions. For example, you can have as many runoffs as there are candidates but one, eliminating each one in turn. Surely that gives more voice to the voter than a single runoff, but it’s obviously unpractical. All these systems have a give and take, none is perfect. RCV, like runoffs, eliminates the huge problems of a simple plurality vote, but is logistically simpler.

    Here’s another scenario. Suppose you have three candidates, White, Gray and Black. A third of the voters love White, can’t stand Black, and can live with Gray. A third can’t stand White, love Black, and can live with Gray. A third like Gray and think White and Black are too extreme.
    In a runoff system, Gray might lose in the general, and after the runoffs two-thirds of the voters would be dissatisfied. In an RCV system, Gray would win, and most of the voters would be satisfied.

  3. I don’t see how exhausted ballots are a significant issue.

    I’m not sure that you understand what “exhausted ballots” are. A voter could dutifully and properly fill in all of their choices and still wind up with their ballot discarded.

    If someone goes to vote in an RCV election and they don’t list either of the top two candidates then their ballot is considered ‘exhausted’ and they have no say in the final runoff total. So in 2011 if someone voted Chiu-Herrera-Dufty then they are denied the opportunity to vote in the runoff.

    This happened to 55,525 voters in 2011. That is 28% of the people who voted. More than 1 of 4. In the real world they could have decided between Lee and Avalos and taken part in that decision. In the RCV world they were simply “exhausted ballots” and other people decided who would be mayor.

    28%. In a big time election with big names and lots of drama. And that’s something that you don’t see as significant?

  4. Her election was of course related to the RCV format, but not her being a whack job. Perata could have probably beaten her in a runoff — it’s all hypotheticals. He had a lot of detractors. Polarizing politicians don’t do well in RCV elections.

  5. Yeah. Actually it very well might have had something to do with RCV.

    If the voters had been made to focus on Quan for two months during a run-off they might have seen cause to reconsider their vote.

  6. The Quan victory was, as G_G would say, achingly close. It is considered the first time that an obvious 1-2 strategy was implemented in the Bay Area, and it worked.

    The reason that race is now used to defile RCV is because Quan turned out to be a whack job. Obviously that had nothing to do with the system.

  7. Some people’s only interest in Oakland politics is to invoke Quan’s name in a hapless attempt to make some kind of point about RCV. Never mind everyone else who’s been elected via RCV before and since.

  8. “No one tells me what to do” is her specific response to accusations of being controlled by the very people her opponents claim she is controlled by. Rewording it to mean something else is not a move with integrity.

  9. By “moderating my position” I meant that voting participation does not always go down in reruns, but it does so sometimes: technically, in fact, most of the time, though I won’t belabor the “most” part, since elections are not all the same, and so quantitative measures are not the most reliable.

    Using a much larger data set, that of national Federal election runoffs, one study finds consistently lower turnout for runoffs. This drop-off (on the average 35%) is much larger than in SF mayoral elections, perhaps because Federal elections tend to have a larger turnout to begin with.

    “Every RCV election will see a significant drop in participation because of exhausted ballots”: I personally have never missed having a fourth option on an RCV ballot, and often have left the third spot blank, for lack of acceptable options. I don’t see how exhausted ballots are a significant issue, as you say.

    The jumping through hoops argument goes both ways. Filling a whole new ballot (and going to the voting station) is more of a hoop jumping than marking another box or two on the same ballot, especially for the poor and busy.

  10. I responded to your original post because you were spreading a false notion propagated by RCV supporters. You stated factually that “the bottom line” is that more people participate under RCV. An RCV proponent took the podium at the debate in the library and spread the same falsehood. I appreciate you “moderating” your statement but the original RCV talking point that you were pushing just isn’t true, certainly not in this election.

    What you don’t take into effect is that every RCV election will see a significant drop in participation because of exhausted ballots. And turnout isn’t up so exactly where is this increased participation coming from?

    And the only way you can minimize the exhausted ballot problem is to try and get people to rank every candidate. In other words, ask the voters to jump through hoops and rank candidates that they don’t care about in order to make your highly flawed system slightly less flawed. The system is supposed to work for the voters, not the other way around.

  11. Tim your lack of a sense of fairness in reporting is stunning. As a founding contributor to 48 hills, I think in order to have the publication survive, you ought to think about the fact that obvious bias is very ugly for any publication’s potential future. Just my opinion.

  12. We discussed this a month ago, here.
    To elaborate: there were six mayoral election years with runoffs. These are the numbers (votes cast and percentage change):

    1975: 215,559 -> 200,804 (-7%). Moscone’s 1st term
    1979: 204,869 -> 196,341 (-4%). Feinstein’s 1st term, after filling in for Moscone
    1987: 193,742 -> 153,720 (-21%). Agnos’s 1st term
    1995: 216,735 -> 198,165 (-9%). Brown’s 1st term, defeating incumbent Jordan
    1999: 203,674 -> 228,247 (+12%). Brown’s 2nd term
    2003: 209,723 -> 253,872 (+21%). Newsom’s 1st term

    You could compare voter turnout instead; the change will be lower, because at each election some more register between the regular election and the runoff.

    I’ll moderate my statement and say that some but not all runoff elections have lower turnouts.

    Speaking for myself (and I like politics, God help me,) I found having another round of political fighting and advertising, leading to another election, quite exhausting. With RCV I like that when it’s over, it’s over.

  13. Obviously an attempt at humor, but not impossible. Election fraud does occur. I know of one case in Alabama. And there was a ballot box found floating in the Bay during Brown’s tenure.

  14. Don’t forget that Breed is possibly the most out-for-herself candidate: “No one tells me what to do!” And that includes her constituents. She will stab your back just for criticizing her (i.e. asking a difficult question). And like Geek_Girl says, we won’t have Breed for just a little while if she wins. The Conway machine is behind her and she’s a smooth operator.

    I’m not crazy about Leno, particularly the endorsements he’s given, but he did enough in his State Senator position to piss off the real estate industry, so he must be doing something right.

  15. Even reappropriating the 20% Zhoosh mentions over to the 1-2-3 candidates, Lee still would’ve won. DOE ran the numbers and both Zhoosh and I have linked them.

    It wasn’t a close race.

  16. No, not legally. Ron Conway hired some homeless people to sneak into City Hall and change the results at the Dept of Elections.

  17. The comments on the article on the housing in Forest Hills being dropped show that the YIMBYs are frauds. They are pushing gentrification. Wiener’s bill will protect the western part of town.

  18. I don’t think I’m important, but I don’t underestimate what I can accomplish.

    My Leno issue is nothing to do with me. I approach it from an effectiveness stand point, and blowing up the comments yesterday was part of the plan. The call from his campaign came in at about 8:30 p.m. last night (Friday night). Think about it.

    You’ll see a lot less of me after the June 5 election, but until then I’m in it to win it. In the meantime I am prepared to weather some personal attacks.

  19. Actually, in 2015, three virtual unknowns gave Ed Lee quite a fright. He prevailed, but they came achingly close to pushing it into further rounds, which would very likely have cost Lee his office.

  20. If Breed gets in, Brown-Newsom-Lee continues for ten more years. Pay to play continues for ten more years. Ron Conway gets ten more years to sack the City.

  21. Anyone who is surprised that Randy Shaw is shilling for Breed is naive. He was a major cheerleader for Lee. He depends on the moderates to keep his position as chief slumlord of “Care Not Cash.”

  22. Dude, you have a way over inflated impression of yourself and your importance. Get a grip, for your own good. And find your own link,

  23. I’m not in a position for him to stab me in the back. It’s him who is vulnerable, that’s why I got the call back.

  24. Leno’s always gracious. He’ll remember your name and talk your ear off.

    Then he’ll stab you in the back.

  25. I think they are somewhat friends, but Johnson might not want to associate with someone with so many skeletons in their closet, not to mention the skeletons running around in broad daylight.

  26. I guess you decided you were interested in Oakland politics after all. Don’t you want to thank me for finding the spread sheet you’re using?

    You’re post is so true — if you took back the partisan part. Understanding RCV scenarios has nothing to do with being progressive or moderate.

    Ps. Do you have a link to back your 57% claim?

  27. You want it both ways. If it’s a litmus test for selfishness, then AFW should bow out now. But it’s not, and she has a right to run.

    Leno and Kim have both made mistakes. What’s worse: sponsoring the Twitter Tax break or endorsing Scott Wiener? Tough call that personally I’m not going to worry about. There is no perfect politician.

    Leno hasn’t said he will not accept a dual endorsement. He likely will. Personally I think he’s effective and could do an excellent job. I stopped by his campaign headquarters the other day and spoke with him after the forum Thursday night. He was extremely gracious and I did received a follow up call from his staff yesterday.

    Breed is an accomplished politician and a very smooth public speaker. Plus she’s cool. If she wins, it’s a clean win.

  28. Don’t get me wrong, I love RCV. I campaigned for it in 2002. I’m just not sure Leno is worth ranking. He has a history of stabbing progressives in the back. I don’t think he’s ever once endorsed a progressive in a race where a moderate had a decent chance of winning. If it’s Leno vs. Breed, I honestly don’t know if it might be better to endure Breed a little longer and regroup with a better candidate next time.

  29. 3 main reasons:

    (1) There doesn’t seem to be a certified voting system to handle more than 3 choices.
    (2) The ballot is confusing enough with three choices. It would be exponentially worse with eight. People would constantly be marking down two #4s or ranking the same candidate as #3 and #5.
    (3) People don’t have an eighth choice to give you. You’re asking them to create one in order to fit into your contrived system. What is your 5th favorite color? 4th favorite author? Well better make one up because you need it to fit into our system.

    I hope that we do find a way to ask people to rank eight candidates. I am for anything that will awaken people as to the silliness of RCV and speed its demise. On federal elections people will be able to simply put a mark next to the person they want to lead the free world. Then, down ballot, when they get to their local city alderman they will have to jump through hoops and spend more time ranking people they aren’t even familiar with. That will have to help end RCV.

  30. Good points. One thing that the progressives don’t realize is that these RCV strategies don’t work as well as they think. There are posts in this thread where people are just adding up (Leno+Kim)-Breed.

    In the Jean Quan Oakland case only 57% of Rebecca Kaplan’s votes when to Quan. In the 2015 SF Mayoral race about 20% of the “Anyone but Lee” candidates’ votes went to Lee anyway.

  31. And you are the exact reason the third slot is crucial to the validity of the RCV system. You are the wild card factor; with the third rank the algorithm is increased 10,000 fold in complexity.

  32. Weiss is the least noxious of the three, even though she knows nothing about government. It’s hard enough holding my nose and picking Kim second. But holding my nose and giving Leno any kind of a vote is just too much to ask. I think I’d pass out from the stench before I could do that.

  33. I think Breed will win.

    A lot of Kim voters won’t choose Leno for second place because he’s seen as too much a part of the establishment. They’ll take the opportunity to vote for another woman of color instead. That’s Breed.

    A lot of Leno voters won’t choose Kim as second place. Jane Kim is really disliked by many conservative and moderate San Franciscans, though I can’t figure why exactly. She seems like a perfectly nice person (I’m seriously considering putting her as my second place choice). When I ask friends or acquaintances who feel this way why, they either launch into a rant about the Tenderloin or they mumble something about bedbugs. My point is, Leno straddles the line between Moderate and Progressive, and his more moderate voters won’t want to make Kim their second choice. They’ll pick Breed for that honor.

    Finally, a LOT of people are going with Breed as their first choice. She’s got the zeitgeist, she’s got a hell of a life story, and there’s a bunch of techies living here now. They just don’t view Conway as the bogyman like most activists/artists do.

  34. RCV is more complex, sure, but the bottom line is more people participate fully in the democratic process with RCV than with runoffs.

    That is actually one of the many falsehoods that the RCV people try to sell. It might be true in some cases but it is demonstrably not true when it comes to the San Francisco Mayoral race.

    The last real runoff we had was 2003. In November 209,723 voters participated. In the December Newsom-Gonzalez run off 253,872 voters participated. That is not a decrease in participation, it is a 21% increase.

    And that was not an outlier. In 1999 203,674 voters participated in November and 228,274 participated in December. That was a 12% increase.

    So those were the last two regular run-offs and both times significantly MORE people participated in the real runoff. The DOE only has one other Mayoral election on line and yes, back in 1995 there was an 18% falloff back then.

    But meanwhile, lets look at the terrible record of RCV. In the only RCV Mayoral election (2011), a grand total of 197,242 voters came out. Only 141,617 votes were counted in the Lee-Avalos “run-off” (Lee beat Avalos 84,457 to 57,160). That is a whopping 28% falloff because of all the ignored…oh, excuse me…”exhausted” ballots.

    You could argue that RCV is the best bad solution for supervisor races where nobody cares about enough to vote in a run-off. But it is an insult to the voters of San Francisco that we use it to decide our mayor.

  35. What I don’t understand is why rcv is only 1-2-3 here. In Australia (since 1918?) if there are 8 candidates running (say for SF Mayor) then each voter can number 1-8 if they choose. Seems to be a fairer system from the voters point of view imo… and too bad if it gives candidates and their campaign managers a reason to take two Excedrin and have a good lay down. I should also mention that Down Under it’s a citizens’ civic responsibility to “show up to vote”. No show without a legit reason and you’re up for a $150 fine.

  36. They are vying for the support of the CPA and David Ho. Every little thing – such as who’s name comes first in a dual endorsement headline – now carries weight.

  37. Johnson at least she has government experience, which with good PR she could present as sufficient.
    I think many D6 conservatives will put Trauss and Johnson in the first two positions, but I have a hard time seeing a formal partnership.

  38. RCV is more complex, sure, but the bottom line is more people participate fully in the democratic process with RCV than with runoffs.

  39. Leno and Kim where campaigning together in Chinatown last week (I believe it was March 1), walking the length of Stockton with a lot of supporters following them. The supporters had signs and many of them had placards from both Leno and Kim.

  40. You think they’ll team up?

    My take is the rcv strategy system is really quite simple. Either you do your best to clear the field and endear yourself to both sides, or you partner up for a 1-2. A 1-2-3 scenario would take extreme conditions to ever work.

    Progressive voters can decide closer to the election which candidate they prefer, Kim or Leno. Breed can keep appealing to people on the cusps in hopes of busting the strategy.

    This election is different than D6 because you have three extremely experienced and qualified candidates. Trauss and Johnson will have to pull up a long ways for it to be this close.

    h. and I went last night. All the candidates did very well and there’s nothing to report. Breed left early, but she was right to do so as the “yes or no” questions were set up to single her out. Somehow she saw it coming.

  41. No surprise that London Breed the darling of developers and YIMBYs was stated as “ahead” in that “poll.”

  42. The pleasure is to play, makes no difference what you say

    I don’t share your greed, the only card I need

    The Ace of Spades

  43. Also, YIMBYs like Watson Ladd love, love, Love SPUR, and some are members. SPUR is in favor of the Central SOMA plan. Can’t have it both ways, laddie.

  44. Oh I know I can get signs, I don’t care that they haven’t given me one. I’m concerned the campaigns aren’t doing enough organizing.

  45. Ranked choice has 3 slots. Vote 1-2-3, not just 1-2. Weiss, Kim, Leno, or Kim, Leno, Weiss, or any combo of the three.

  46. Keep in mind hypocritical yimby party supported Prop O last year despite lack of office/housing balance. Maybe you should work on your own team before railing against others.

  47. Prop M limits aren’t the issue with the SOMA plan. The issue, as always, is not adding enough housing. 7,000 homes is not enough for the number of new jobs that will be produced.

  48. Given that Trauss and Johnson would likely use RCV to defeat Haney in November, then your scenario sounds like a good one.

  49. Keep in mind that despite Kim’s support, she is not planning to override Prop M limits on office space development. And she may well decide to withdraw her support after hearing from her constituents. This remains to be seen.

  50. I’m not hedging my bet. I’m saying that I want Breed but it would be in the best long term interests of the city if we could have real elections and not simulations. This might be our Jean Quan moment.

    So if Breed and Kim are the two strongest candidates perhaps we just ask the voters which of the two they prefer. And not based on some throwaway second preference. Let them cast a vote in a real runoff where they actually know who is running against who and have been able to think about the consequences of both. Which they won’t be able to do in this simulated “election”.

  51. you’re right about the anonymity of the poll – but the source wasn’t any campaign. it was a developer’s poll mostly about development issues.

  52. forget Beyond Chron and its obvious bias. a very recent independent poll has Breed at only 28%, Kim at 24%, and Leno only at 17%.

    That’s good advice, thanks. Nobody should rely on an unidentifiable poll from an obviously biased source.

  53. forget Beyond Chron and its obvious bias. a very recent independent poll has Breed at only 28%, Kim at 24%, and Leno only at 17%. so that’s 41% combined for the progs and 28% for the mods, with the 31% remainder scattered or still undecided.

    that’s a big lead for the progs. Kim has picked up momentum with her always-obvious energy, but Leno?

    Breed has her mod base, and the relatively small numbers of African American voters still left in SF, but what is her plan to win 23% of that remaining 31%? that’s a heavy lift. “move left”? the proposed affordable housing business tax she supports would normally be such a move, but adding the poison pill to it to kill Kim’s universal childcare business tax measure is a big turn-off for many progs. that was a mistake which undercuts its value. we’ll have to see how much.

    if Kim/Leno stick successfully to a dual endorsement approach, in theory they just need to pick up 10% of that 31% – certainly likely. but of course the dual endorsement strategy won’t be 100% voted that way, so we’ll have to see.

    if Leno refuses to support Kim as second choice, he really hurts himself. it looks petty, and to many prog voters supporting dual endorsements, that he thinks his own winning is more important than the future course of the city in the alternative where he comes in #3. that flunks the ‘character test’ for leadership. if he’s taking Weiner’s spiteful advice about this, he’s a fool.

  54. Who cares about polls. Where is the organizing? I see very few Kim or Leno signs around the neighborhood. My partner has been trying to get Kim signs for our mission district apartment. Someone said they’d deliver some but then radio silence from their campaign office.

    Whoever knocks on the most doors and gets the most signs up and talks to the most people will win this race. Get going, campaigns!!!!

  55. Meanwhile Jane Kim is planning to support the Central SOMA plan. What’s that going to do to displacement?

  56. Hedging your bet I see.

    Breed winning round one by 11 points is a decent prediction, but to say it would be an aha moment for rcv elections isn’t accurate.

    The 11 points would be more about how this election is stacked. If Herrera wouldn’t have been kicked to the curb, it would be different.

    Standard elections are often decided by who has the most commercials. RCV’s effectiveness has many variables. This election won’t provide direct evidence whether it is or is not better than a standard run-off.

    The only card you need is the Ace of Spades. Double or quit, double stake or split.

  57. Zhoosh wants Breed to win. But the BEST thing that could happen is if Breed wins round one by 11 points and then is sent packing because of ranked choice voting, without even getting a chance to go one on one against the candidate she just trounced. Then we will finally be able to get rid of this silly voting simulation game.

    This Tim Redmond article should be required reading. By all means, let’s decide who will run the city based on their ranked choice voting strategy.

  58. Who could’ve ever seen this coming.

    But seriously, tonight I toast Iris Canada. Forever a Fillmore legend.

  59. By and large, it will be Breed’s conservatives vs. Leno and Kim’s liberals. Where it gets tricky are the crossovers, where identity politics and politics don’t match: will Castro conservatives vote Leno first and Breed second, or vice versa? Who will Sunset conservatives pick for first choice? Those are tricky unknowns.

Comments are closed.