Post-Election Post-Mortum: D3 and Voter Contradictions

An awful lot has already been written about the 2015 midterm elections in San Francisco even though there wasn’t actually a lot going on elections-wise. I’m finally getting around to writing my thoughts because 1) the act of moving to a new apartment took up every inch of mental and physical real estate that I had; 2) I went on vacation, and 3) the actual final election results weren’t published until November 19. Prior to that they were still counting all our ballots, which means that all of that journalism that was happening immediately after Nov 4 was based on incomplete numbers. Ultimately that didn’t change any of the results, but it does make some of that post-election analysis less meaningful.

There weren’t any national or state-level elections on the ballot. Nevertheless, after all the ballots were counted, 45% of San Franciscans cast their ballots for a handful of city-wide issues and candidates (of which the most high-profile was the reelection campaign of Mayor Ed Lee).

The issue at the heart of this election was housing. Local Measures A, D, F, I, J and K dealt directly or indirectly with the cost of housing. Several poorly funded candidates banded together to take down Mayor Ed Lee, whose policies, they felt, reflected close ties to moneyed interests. Similarly,  the District 3 Supervisorial election, where former progressive Supervisor Aaron Peskin took down the Mayor’s appointee Julie Christensen, was dubbed by many on the far left (the “progressives”) as a referendum on the more moderate Mayor.

The results are in: the Mayor won, but not by much, which is pretty bad for considering he faced only token opposition. Ed Lee’s reelection was worse than that of both his predecessors (Gavin Newsom, who, like Lee, faced little opposition; and Willie Brown, who faced a real challenger in Tom Ammiano). They were re-elected with 73 and 57 percent of the votes, respectively.

I’ll admit that this next comparison isn’t really fair, but I’m going to make it anyway. In San Francisco, we have ranked-choice voting. This means I get vote for my top three choices for mayor, in order of preference. Once all the votes are cast, the ballots cast for most losing candidates are transferred to the candidates those voters’ marked their second choice. This happens until either the ballots are exhausted or one of the candidates gets over 50% of the vote (and wins!. In Ed Lee’s first election he got almost 31% of the vote in the first round, and then won with almost 60% of the vote once ranked-choice kicked in. In his reelection, the ranked-choice votes didn’t matter because he was the first choice of over 50% of voters, but when you compare his winning total of 59% percent in 2011 to his winning total of 56% in 2015, it kinda looks like he did worse. Now, if you do include the ranked-choice voting (not including it is what makes it an unfair comparison) his 2015 total shoots up to 67%. That being said, in 2011 he faced at least six other serious candidates, so the contest, in general, was much tougher back in 2011. Anyway, I don’t totally know exactly how to splice the mayoral results other than to say that Lee was both extremely vulnerable but also his reelection was inevitable. Maybe both are true.

Inconsistent Results from D3 Voters?

Lee’s appointee to the Board of Supervisors for District 3, Julie Christensen, wasn’t quite as lucky. Aaron Peskin reclaimed his old seat with about 52% of the vote.

However, all of the Mayor’s ballot initiatives did really well. Props A (Housing Bond) and D (development at Mission Rock) soared. The mayor opposed the controversial and progressive-backed Props I (Mission District Market Rate Housing Development Moratorium) and F (stricter AirBnb regulations), which went down in flames. So why did San Francisco voters validate the Mayor’s legislative agenda, but then vote against him and his legislative representative in such large numbers? Or, conversely, how did Aaron Peskin get elected given his support for Props F and I. Aren’t voters sending mixes signals?

Did voters associate Ed Lee, Julie Christensen and Aaron Peskin with their positions on Propositions F and I and vote accordingly? Or were they able to separate these candidates from their positions on these few issues?

Here’s a map of the District 3 election results by precinct. Note: each precinct is coded with three letters; the first letter (P or C) indicates who won (Peskin or Christensen), the second letter (Y or N) indicates whether Prop F passed (Yes or No), and the third letter (Y or N) indicates whether Prop I passed (Yes or No). The combination of possible results is indicated with a color. Peskin wins are in shades of red/pink; Christensen in blue. Ties are purple (There was only one, and they voted no on both Props F and I). Some precincts reported combined results.


Julie Christensen supporters were the most ideologically consistent. She opposed Props F and I and she only won in precincts that voted against those Props (with one exception). Aaron Peskin also had a lot of ideologically consistent voters. He won all the precincts that supported Props F and I. However, he also won a lot of precincts where voters were against Props F and I. Additionally, he won quite a few mixed precincts where Prop I failed, but Prop F won (Julie Christensen only carried one mixed precinct).

Even though Peskin won more precincts, it’s important to remember that not all precincts are created equal. Winning a greater number of precincts (which Peskin did handily, 21 out of 30) doesn’t necessarily mean winning the election. Depth of support is also very important. If you overwhelmingly win a few precincts and hold your opponent to a tie everywhere else, you could still win. Julie Christensen didn’t do that either. In fact, Aaron Peskin won 8 precincts by over 100 votes, Christensen had only two (It’s crazy to think of 100 votes as being significant especially considering that this campaign cost the candidates and their PACs about $1 million each. As small as 100 votes sounds, fewer than 20,000 voted for either Peskin or Christensen in this race, so 100 votes can make a big difference).

The table below is sorted in order of precincts that most favored Julie Christensen to most favored of Aaron Peskin (the column called “Peskin Vote Lead”). Precinct 7308 (Telegraph Hill) was a tie, which is why the Peskin vote lead is zero. More red=more Peskin, more blue=more Christensen.

D3 table

Aaron Peskin won because he had breadth (many different precincts) and depth (several landslide precincts) to his support despite the fact that he was a progressive candidate who was in favor of two losing progressive causes.

Why might that be?: Even though Aaron Peskin associated himself with two losing propositions, it might be the case that those specific initiatives didn’t matter very much to District 3 voters. Prop I, which sought to ban market-rate housing development in one neighborhood, really only effected the Mission District . Even though Prop I was a big progressive cause, it didn’t actually impact voters outside of District 9. As a result, D3 voters could opine on this policy without actually having to live with the consequences. Prop F on the other hand almost passed in D3, and voting against it didn’t necessarily mean you didn’t want stricter regulations on AirBnb, it just meant that maybe you were uncomfortable with Prop F specifically, which was criticized for being too long and complicated.

Aaron Peskin, on the other hand, had several other things going for him. He’s very closely associated tenants rights and eviction reform. He was also an important figure in defeating the 8 Washington development, which was hated by both rich NIMBYs and affordable-housing advocates. He spends a lot of time on cases of individual constituents. He was instrumental in negotiating with developers on behalf of the flower vendors after they faced displacement from the Flower Mart. Christensen’s only real other association was with the mayor.

If you cared about defeating Props F and I, you probably also voted for Julie Christensen. If you cared about supporting Props F and I, you probably voted for Aaron Peskin, but if you didn’t really care all that much about them, then you also probably voted for Aaron Peskin; and you probably did so because he had done other things you liked.

Treading Carefully

Despite the success of his campaign, Aaron Peskin may need to watch out. District 3 is considered a “swing” district, which in San Francisco means that it sometimes votes for moderate (or neoliberal) candidates or causes, and it sometimes votes for progressive candidates or causes. Ed Lee and Julie Christensen are considered moderates; Peskin, and Props F and I are considered progressive. Let’s telescope out to look all the candidates and ballot propositions and their relative success in District 3.

d3 results in order of success

You can see that at the top here are the Propositions that passed overwhelmingly in D3 as they did citywide. These are Props G, A, K, D, C, and B, which all got more than 60% of the vote in D3. I haven’t yet poured through the results of every precinct, but I’m willing to bet there are very few where these Props didn’t pass.

After that we have Mayor Ed Lee, Prop J, and Vicki Hennesy (who beat Ross Mirkarimi for Sherriff) all receiving about 60% of the vote.

Then Aaron Peskin scoots over the finish line at 52%

Next we have Props F (48%), Prop I (44%) and Julie Christensen (43%), all losing in D3, but by less than 10%.

Everything below that was defeated handily.

The interesting thing here is that Ed Lee actually over performed in District 3. He received about 55% citywide, but got about 60% of the vote here. This points us to another contradiction from the voters in D3. They like the mayor here more than the rest of the city, but were deeply uninterested in electing the woman he appointed to represent them at the Board of Supervisors. That means a lot of people voted for the Mayor and against Julie Christensen. Where was this phenomenon most present? Where were voters most likely to vote for the Mayor and against Julie Christensen? Take a look at this table sorted by precincts where the Lee share of the vote most exceeded the Christensen share of the vote.

C percent minus L percent


The top three precincts (7322, 7329, and 7338) where Lee had the shortest coattails are all in (or contain parts of) Chinatown. Number four (7313) on the list is, at the very least, Chinatown-adjacent (if not actually in a more expansive definition of “Chinatown”). As is number six (7319).

A lot has been written about Aaron Peskin’s success in Chinatown. How Peskin’s win may have ridden on his ability to win over a neighborhood that really loves the mayor.

All of which seems true. Christensen won a few of the Chinatown precincts, but only by a slim margin. Meanwhile, Ed Lee stormed through Chinatown like Godzilla in Tokyo. Some take this to mean that Ed Lee’s support is really pretty shallow, which could be true. On the other hand it may mean that Aaron Peskin needs to tread carefully. He won Chinatown despite the fact that he wasn’t supported by the popular Mayor and was on the wrong side of both Props F and I. He has to face another election in 2016 (his election was only to fill the remainder of Julie Christensen’s appointment). Should he become embroiled in issues D3 residents oppose, he may draw some more well-funded opposition.

It’s obviously early in his term, but the progressives were crowing about finally being able to pass legislation. Yet we haven’t seen them reintroduce (or announce that they’ll reintroduce) any of the recently failed legislation, like stricter AirBnb regulations or a development moratorium for the Mission. Progressives said that both were critical to saving the “soul” of San Francisco.

I even tweeted to David Campos and Aaron Peskin to ask if another Mission Development Moratorium bill was in the works. I did not receive a response.


Part of the reason for that reluctance to move on these issues is that they’re pretty unpopular with most voters, including those in D3. Nowhere was that more true than in Chinatown. Should Peskin actually go head to head with the mayor he may see that support erode. I think Aaron Peskin is going to do his best not to find out if that’s true.

One note about Chinatown voting: the Chinatown voters were more critical for Ed Lee and Julie Christensen in D3 than for Aaron Peskin. Votes from Chinatown made up about 11% of Peskin’s support, whereas it was 12% and almost 16% for Christensen and Lee respectively. Cleaving some of those PNN precincts (of which Chinatown was a big part) away from Peskin will be primary task for any would be challenger.

With an eye to 2016, fending off possible challengers will be important for Peskin. The progressives have their work cut out for them. They currently have a one seat majority on the Board of Supervisors, but have a lot of big elections ahead of them, and having to re-defend D3 would force the progressive political establishment in the city to spread it’s resources (money, volunteers, etc.) more thinly. In Districts 9, 11, and 1 progressives Supervisors David Campos, John Avalos, and Eric Mar will be term-limited from seeking reelection. In district 7, progressive Norman Yee faces reelection in one of the city’s most conservative districts. Progessive Supervisor Jane Kim (D6) faces off against moderate Supervisor Scott Weiner (D8) for a seat in the state senate. Whoever wins, the moderate mayor will get to appoint their successor. They only have one pick-up opportunity, in District 5, where the more moderate London Breed will be seeking reelection in a progressive district. Despite their slim majority on the board, the progressives are playing defense in 2016, and that game starts now.


2 thoughts on “Post-Election Post-Mortum: D3 and Voter Contradictions

  1. Pingback: Can 2015 Foreshadow 2016? | Lord of the Fails
  2. Pingback: 2016 Election: Can the Primary Results Predict the Next Six Days? | Lord of the Fails

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s