Ninth Circuit Upholds San Francisco’s Variety of Instant Runoff Voting

On May 20, the Ninth Circuit upheld San Francisco’s particular type of Instant Runoff Voting. The case is Dudum v Arntz, 10-17198. The opinion is 33 pages. The plaintiffs, San Francisco voters, had filed this lawsuit in 2010. The lawsuit did not challenge the constitutionality of Instant Runoff Voting in general. Rather, it challenged San Francisco’s version, which limits each voter to ranking only three candidates. Thanks to Rick Hasen for the link.

The attorneys for the voters who challenged the San Francisco system are the same attorneys who represent former California Lieutenant Governor Abel Maldonado in the lawsuits that challenge certain aspects of California’s top-two system. Maldonado intervened in those lawsuits, Field v Bowen and Chamness v Bowen, to defend the parts of California’s top-two system that do not treat all candidates equally relative to party labels, and which permit write-in space on run-off ballots but do not allow those write-ins to be counted. It is somewhat ironic that the same attorneys who attacked San Francisco’s version of IRV on the grounds that it fails to “count” the votes of all voters, are defending the top-two law that specifically says some votes can not be counted.


Ninth Circuit Upholds San Francisco’s Variety of Instant Runoff Voting — 6 Comments

  1. It is only the attorneys for the intervenors who have acknowledged that there is a reasonable interpretation of SB 6 that would let candidates use the party preference that they indicated on their voter registration affidavit (signed to certify that it is truthful and correct, subject to perjury charges).

    The plaintiffs’ attorney has offered an interpretation of the law that is so strained and contorted that it would not be unreasonable to consider it a frivolous lawsuit.

    In the 41 special elections prior to the Open Primary (going back to 1995), there were a total of 4 candidates not affiliated with a qualified party. In the 4 special elections since the Open Primary has been instituted, there have been 5.

  2. It is interesting that the appeals court was aware of 2010 supervisors election (included in their footnotes) but did not mention the District 10 election, which was a massive failure of Restricted IRV.

    In fact, there were more voters in the conventional runoff in December 2000, when voters had to make a second trip to the polls, than there votes cast in the final round of the “instant runoff” in 2010, because so many votes were discarded because of limitations in the ballot and voter confusion.

    20% of voters who chose to vote in 2010, cast an irregular ballot of some sort, which is a strong indicator of confusion or frustration with the system, and less than half voted for either of the two finalists, even though they had three chances to do so.

    With respect to Footnote 8, the 2006 election for District 6 is the extreme outlier, clearly not the norm.

  3. Concerning District 10, it would have been a “massive failure” under the original two-round runoff system. The two highest vote-getters in the first round of tabulation accounted for less than 24% of the vote (and fewer votes than accumulated by the ultimate winner); 76% of the voters wanted someone other than those two. The person who did win in District 10 would not even have made the runoff in San Francisco’s original system.

    The right answer to the problem of exhausted ballots is to get equipment that allows for more rankings. The San Francisco Voting Systems Task Force is supposed to be making recommendations concerning San Francisco’s “next generation” voting system, yet their report (so far) does not mention extending the number of rankings. The task force can be contacted at

  4. P.R. legislative bodies
    App.V. executive/judicial offices

    — pending major public education about head to head Condorcet math.

  5. IRV does NOT have EQUAL treatment for ALL 2nd, 3rd, etc. place votes

    — i.e. one more blatant violation of the EQUAL protection clause in 14th Amdt, Sec. 1

    — regardless of math challenged MORONS doing the case and especially the judges in the case.

  6. #3 There were 500 1st-rank overvotes. My understanding is that they were more likely to occur in Hunters Point and Bayview than other areas of the district. If this had been a conventional runoff, they would have been examined to determine whether Tony Kelly or Malia Cohen made the runoff.

    And of course there would not have been as many overvotes in a conventional system (for example only 0.27% in the governor’s race). It is quite likely that Malia Cohen would have made the runoff.

    It is less likely that there would have been 23 candidates in a conventional runoff system. Under restricted IRV, it is a possible election strategy to support extra candidates that might appeal to supporters of your main rivals because of the confusion it would cause.

    So under a conventional runoff system Lynette Sweet and Malia Cohen would have debated, and the voters would have chosen who was to represent them.

Leave a Reply

Your email address will not be published. Required fields are marked *