San Francisco to repeal ranked-choice voting?

I am passing along the following action alert, dated 15 June, from Californians for Electoral Reform:

The San Francisco Board of Supervisors will, as early as June 26th, consider placing on the November ballot a charter amendment that will repeal RCV for all city-wide elections (Mayor, Sheriff, District Attorney, City Attorney, Treasurer, Assessor-Recorder, and Public Defender), replacing them with a two-round runoff. The first election will be held in September; if no candidate receives 65% or more of the vote (yes, it establishes a 65% winning threshold), the top two go on to a November runoff, seven or eight weeks after the first election. (RCV is kept for the Board of Supervisor elections.)

We need you to contact your Supervisor and tell them why you are opposed to it and ask them to vote against putting it on the ballot. We especially need people who live in District 5, Supervisor Olague’s district, to lobby her, as she may be the swing vote on this issue.

If you don’t know what district you live in, this link should help you.

Some talking points can be found at

Additional things to note: San Francisco isn’t used to voting in September; turnout will be abysmally low.

Eight weeks isn’t enough time to certify the first election and get runoff ballots to military and overseas voters. (In some years, such as 2015, there will only be seven weeks between the elections.)

Every year in which we vote for Governor there will be *three* elections: the June statewide primary, the September San Francisco election, and the November general/City runoff election. Talk about voter fatigue, what about poll-worker fatigue!

Please note that the Board of Supervisors will NOT be taking public testimony at the meeting(s) where they consider this issue, as they took public testimony in yesterday’s Rules Committee hearing. Contacting your Supervisor personally is now the best way to make your voice heard. (In addition, you can write the Board, but please contact your Supervisor.)

–Steve Chessin
President, Californians for Electoral Reform

20 thoughts on “San Francisco to repeal ranked-choice voting?

  1. Whatever one’s opinion about IRV vs. TRS, one has to wonder why eight weeks between rounds would be considered not “enough time to certify the first election and get runoff ballots to military and overseas voters”.

    Somehow, the French manage with one week for legislative elections and two for presidential. Several countries in Latin America and elsewhere that use two-round systems for president, and allow overseas voting, manage with a month between rounds.

    For more, see the discussion at the thread on the French elections, especially comments #8 (rici) and #10 (Alan).

    I concur with Alan’s observation that “Somehow US election administrators have persuaded themselves and everyone else that elections are a lot harder than they are.”

  2. The problem is that we are assuming that the only options are IRV vs. TTR65. If the question is “how do we get a better voting system” then we are ignoring the ~50 or so systems that exist out there.

    Score voting (a.k.a. range voting) would have the benefit of both and the detriments of neither.

    Like TTR it is easier to understand than IRV, you can count the ballots in the precincts, and there are no rounds of voting. Like IRV there is no need for a runoff, which saves money, and you have only one voting group.

    Score voting goes a step further and allows you to always vote for your favorite candidate without penalty to them in every circumstance, requires smaller ballot space than IRV (reducing printing costs), reduces ballot spoilage more so than TTR or IRV, and there are no spoilers.

    San Francisco has the spirit to recognize there are better ways to do things. Too bad it wasn’t smart enough to choose the best way.

  3. While I don’t want to launch the annual range voting debate, I think I have to say for most of us this is a reasoned position, not an assumption reached in ignorance.

  4. Regarding ignorance, I concur with Alan. I do not agree that range voting gets rid of tactical voting. A Paul supporter might have an actual preference of Paul 100%, Romney 70%, Obama 0%. If Paul has any chance of winning, then her incentive is to give Romney 0% in stated preferences, so that Romney doesn’t defeat Paul. If Paul looks like running third, though, she would want to give Romney a 100% rating so he can block Obama. Ultimately range voting would devolve into approval voting, which in turn would become a very crude form of AV, where Nader supporters give a second tick to Gore to keep Bush from winning, but Gore voters have no incentive to give a second tick to Nader.

  5. @Tom Round,

    > Gore voters have no incentive to give a second tick to Nader

    Wrong. Say you believe Gore=10, Nader=7, Bush=0. If Bush has a greater than 30% chance of winning, then your expected value MUST be lower than 7, which means your best tactic is to approve Gore AND Nader.

    Or say you think Gore=10, Nader=1, Bush=0. Then it will be tactical to approve Nader IF Bush has over a 90% chance of winning.

    > Ultimately range voting would devolve into approval voting

    Ludicrous. There’s ample evidence that a significant number of people will use intermediate scores, at least partially. Which, by the way, can be a better tactic than exaggeration, for people who are sufficiently unskilled with math. *Cough*

    And no one is saying that Score Voting gets rid of tactical behavior. But it allows you to safely support candidates you prefer to the frontrunners, including your sincere favorite. So if I think Green=10, Democrat=7, GOP=4, I might tactically vote G=10, D=10, GOP=0. But I don’t have to throw Green under the bus.

    With IRV, you’d want to rank D>G>R.

    I don’t expect most voters to understand any of that complex game theory, but most of them will just do this anyway, because they intuitively believe that IRV works like Borda. E.g. here in San Francisco, NO ONE understands how IRV works. Even smart software engineers I work with at tech startups — who claim to vote — can’t explain it. See this conversation for instance:

    And I’ve called the offices of Australian political parties, where Green Party representatives have told me that one of the most common questions they get is essentially, “Why should I waste my vote on the Greens?” I.e. they think it makes more tactical sense to rank Labor in first place. That they are correct is just pure coincidence.

    Further, tactical voting is basically irrelevant, because Score Voting performs better with 100% TACTICAL voting than IRV does with 100% HONEST voting.

    Clay Shentrup
    The Center for Election Science

    • I would like to propose that those who comment here avoid calling the arguments of other commenters “ludicrous” and such, and also to avoid questioning others’ understanding of elementary math.

      Such approaches are not productive. If one’s arguments really are superior to those of others, it will all come out in the discussion and reasoned debate.

  6. I should add that, while I think IRV is a terrible system, I agree that this proposal to go to a runoff with a 65% requirement would be even worse. But it’s just a tiny difference to quibble over when Score Voting and Approval Voting (and Condorcet and Borda) blow BOTH OF THEM out of the water in Bayesian Regret calculations. It’s focusing on the splinter instead of the bullet hole.

  7. @Alan,

    > I think I have to say for most of us this is a reasoned position, not an assumption reached in ignorance.

    I don’t know about that. Shugart’s own promotion of IRV has contained MASSIVE misunderstandings of elementary school mathematics. Similar to the one made by Tom Round above, when he asserted that Gore supporters would have no reason to approve Nader. In six years of hearing every pro-IRV argument in the book, this has been the overwhelming trend.

    Then you have groups like FairVote, which just blatantly lie and fabricate things.

    • Funny to read “Shugart’s own promotion of IRV” when I have been criticized by other readers of this blog for claiming IRV (AV) not clearly better than FPTP!


  8. MSS,

    It’s relevant to the point. It’s more understandable if someone makes an error related to differential equations. But if you make an error on on some trivial expected value calculation, leading to argue that cardinal voting will degenerate to Plurality Voting, that’s a flagrant kind of irresponsible behavior (because this issue is especially important to human welfare).

    Also, if something is ludicrous, I think that’s a valuable observation. If it is backed up by facts, what is wrong with pointing out that something ACTUALLY IS ludicrous?

    In any case, I admit that I don’t have my initial wiki response to your post many years ago. I lost it when I deleted an old Google Group where it was hosted. I could be mistaken to remember it as “promotion of IRV”. It may have just been “criticism of Approval Voting, in the context of the merits of IRV vs. cardinal voting.” I guess I shouldn’t have brought it up.

  9. I actually propose a twist in the rules of Instant Runoff Voting.

    – If a candidate has a majority of the first round votes, and also doesn’t have the most last place votes, that candidate is the winner

    – If there’s no case, whether being that the candidate with the majority of the first choices also was the candidate with the most last place choices, or that no candidate got a majority of the first choices, there’s a runoff

    – Last place choices are counted, the candidate with the most last place votes is eliminated and those votes move up a place

    – The previous process happens until either a candidate gets a majority of the second choices or until there’s one candidate left

    First and second choices in any race are always choices for candidates you like, last place choices are always choices for candidates you don’t like.

    And, there’s an idea that I like: that the top half of all choices will be for candidates you like, meaning that those choices will eliminate the candidates with the least choices here, while the bottom half of all choices will be for candidates you don’t like, meaning that those choices will eliminate the candidates with the most choices here.

    For example, in a 5-way race, the top 3 choices will be for candidates you like and the bottom 2 for candidates you don’t like.

  10. As a member of the Australian Greens I am not unfamiliar with being asked about ‘wasted votes’. On the other hand I am also not unfamiliar with electors going off quite happy in the knowledge that if their vote cannot elect a Green it will default to their next preference.

    I am entirely unfamiliar with the idea that the Australian Greens support range voting or approval voting because they actually support STV in multimember districts, although in the specific variant known as Hare-Clark.

    I’d include a link to Robson Rotation as well but WordPress would come hysterical.

  11. Some purists, usually South Australians, prefer Hare-Spence for Catherine Spence, who called the system Efefctive Voting, rather than the Tasmanian Andrew Inglis Clark.

    Do not google Hare-Spence.

    You would end up reading:

    There was a Grand Dame of Australia
    Who proved the block system a failure!
    She taught creatures in coats
    What to do with their votes,
    The Effective Grand Dame of Australia!

  12. Yesterday, after several weeks of delays and intense lobbying, the anti-IRV members of the San Francisco Board of Supervisors dropped their partial repeal efforts for the time being. They are likely to be back next year but, for now, IRV remains intact in San Francisco.

  13. Thought of a way to introduce a modified 2-round system.

    First round: voters get to vote for the candidates they’d like to see qualify for the second round.

    Second round: voters get to vote for one candidate.

    Now, this could work with either IRV or plain old x-voting. If IRV is used, the candidates that would qualify for the final round would have to reach the quota.

    Let’s say you have 6 candidates and either the top 3 candidates or all candidates that reach at least 17% of the vote qualify for the final round.

    However, under ordinary x-voting, the first round would be like approval voting, while the final round would be plurality. The first round in all cases would be like a primary, while the second round would be the runoff.

  14. Derek, that’s essentially what the French two-round system is for legislative elections, unless you’re saying the first round should allow voters to cast multiple votes, followed by a second round with a single vote.

    In French National Assembly elections, a candidate wins in the first round if (s)he gets an absolute majority of the vote and turnout is above 25% of registered voters.

    If either condition is not fulfilled, a second round is held between the top two candidates, plus any other candidate who received the votes of at least 12.5% of registered voters. In the second round, the candidate with the most votes wins.

    It’s also similar to the non-partisan blanket primary system (also called a “jungle primary”) used in Louisiana, California, and Washington. The first round is held between all candidates for the election, potentially including multiple candidates affiliated with a party. Even if one candidate wins a majority in the first round, the top two candidates (even if from the same party) advance to a runoff, where the candidate with the most votes wins. The nonpartisan blanket primary system is also used in special elections for partisan offices in Texas, but there is not a second round if a candidate wins a majority in the first round.

    The IRV version with a quota for the second count also exists–it’s called the contingent vote. Voters rank the candidates in the order of their preference, and if no candidate has a majority, all but the top two candidates are eliminated, and the ballots of the excluded candidates are distributed to their highest remaining preference. The candidate with the most votes at the second count wins. It used to be used to elect the Queensland Legislative Assembly, but they have since switched to optional preferential voting.

    There are two prominent variants used in elections today, both of which differ from the traditional contingent vote by limiting the number of preferences a voter may list. In presidential elections in Sri Lanka, a voter may rank their top three choices. In direct mayoral elections in England, most prominently the Mayor of London, a variant known as the supplementary vote is used, in which voters may indicate their top two preferences only.

  15. Precisely Chris! My proposal would allow for the first round to be a primary, where voters would vote for multiple candidates, and the second round would be the runoff, where voters would vote for only one candidate.

    But I don’t like the top 2 idea of the average 2-Round Runoffs. Unless it’s a race between 3 and 4 candidates, fine. However, in races with more than 4, the runoff should have at least half of the total candidates.

    Let’s say Louisiana used my proposal. If there’s a 8-man race, each voter would be able to vote for the 4 candidates they want to see in the runoff. Then in the runoff, held between the top 4 candidates, voters would get to vote for only one candidate.

  16. @ Derek, 19, How would having a four candidate runoff improve things? Such an election could still end up with a scenario where nearly 75% of voters prefer one of the three losing candidates.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.