Keith Bell stands outside the Mission Police station. Photo by Lola M. Chavez, June 2017

Members of the San Francisco Police Commission say they’re unsettled by recent police department data suggesting dramatic and persistent racial bias in officers’ daily interactions with people of color. 

These numbers are disturbing to me,” Commissioner Cindy Elias said Wednesday night during an oft-heated, hour-long exchange with Chief Bill Scott regarding the statistics and what they ultimately revealed about the department’s progress on reform. 

Elias was referencing reports on incidents in which officers stopped a person on foot or in their car throughout 2019 and the first three months of 2020. The SFPD’s Racial and Identity Profiling Board report showed that, in 2019, people of color composed 65 percent of these stops, while white people accounted for 35 percent. 

Moreover, Elias pointed out, in the fourth quarter of 2019 people of color composed 65.6 percent of stops and 72.1 percent of those searched. Only 34.4 percent of those stopped were white, and only  28.7 percent of those were searched. 

“Those numbers are alarming … we are searching people of color more frequently than anyone else,” Elias said. “I asked in 2019 why this was happening and what the solutions are — and I still haven’t heard.” 

“I want to know what SFPD is doing about it,” Elias continued. “We know what these numbers are. … The question needs to be, ‘What are we doing about it and how do we get it to stop?’” 

Police Commissioner Cindy Elias questions Capt. Steven Ford on the SFPD’s disproportionate use of force on black men. March 6, 2019.

Scott said he agreed that the disparities remain a persistent problem. But he argued that the department was actually making progress. He pointed specifically to the so-called “yield rate” of searches, which shows how often officers find something illegal when they stop and search someone. A high yield rate means that officers are more accurate when stopping and searching people. A lower yield rate arises when officers stop and search people, including people of color, based on bad or biased hunches. 

He said that when the United States Department of Justice reviewed the SFPD in 2016, the yield rate during searches of African Americans was 10 percent, meaning only one out of every 10 African Americans searched was carrying any contraband. In the first three months of 2020, Scott noted, that number was 29 percent. “That factor has improved,” Scott said. 

The police are, indeed, searching people less frequently: between late 2018 and early 2020, police have performed 47 percent fewer searches, according to its numbers. The same period saw a 42 percent drop in stops. 

The chief also noted that the department is now offering better training to officers on when and when not to stop people, and when to better employ “reasonable suspicion” and “probable cause” — legal justifications for stopping and searching people. “The training hopefully will improve those indicators and reduce some of these disparities,” Scott said. 

He also said the department’s new and revised “Bias-Free Policing” and “Investigative Detentions” policies, both aimed at curbing bias in everyday police encounters, should help. (Both have yet to be implemented.) 

But Elias countered that the department could be doing more, noting that the SFPD still relies too heavily on “reasonable suspicion” and “probable cause.” It’s searching too many people simply because they may be on parole or probation, she said. These kinds of searches rely heavily on an officer’s subjectivity, and, perhaps, their internal biases. 

Mission Local has been covering the police reform efforts consistently since 2016.  If you want to keep us at it — and have not already done so — please support us now.

In the first three months of 2020, in fact, people of color are being searched more frequently than whites on these subjective legal bases — and officers are often wrong. Between January and March of 2020, officers found illegal contraband only 34 percent of the time when searching Black people and Latinxs, respectively, based on those searches.

“We were failing,” Elias said. “This is a failing grade because we are searching people of color and not finding anything.” 

Elias was, at times, unrelenting in her questioning of Scott and why the numbers were the way they were. The chief continued to point toward the department’s improvements. 

“Shootings have gone from an average of six a year to the last two and a half years to less than two,” Scott said, growing visibly flustered. “How can we say that’s not significant progress?”  

Read:Taking too long’: What an SFPD policy for Deaf people says about police reform

He said use of force is down 49 percent since 2016, and pointed to the fact that, overall, that number has declined with African Americans. “What I’m saying is there has been concrete progress made,” Scott said. 

Elias did not disagree, but she pointed out that being satisfied with incremental progress is not good enough, especially as it remains at the expense of people of color. 

“My point is: How, as a department, do we get officers to a point where they aren’t scared when they see a person of color?” Elias said. “Because what these numbers … are telling me is that we’re using weapons against people of color more than we are against white people. We are stopping them more than white people. We are searching them more than white people.”  

Commissioner John Hamasaki agreed. “How do you feel if you’re out in your own neighborhood, going to the store and you get run up on by gang task force or a dope team searches you?” he said. “It doesn’t feel like your neighborhood anymore — it doesn’t feel like you’re part of the community anymore because your rights as a citizen have been taken away.”

Read more about efforts to solve bias within the SFPD:

Don’t book ’em, Danno: Police move to limit release of mugshots to prevent racial profiling


SFPD wants to better respond to racially motivated calls to police. It’s not so easy.

Follow Us

Julian grew up in the East Bay and moved to San Francisco in 2014. Before joining Mission Local, he wrote for the East Bay Express, the SF Bay Guardian, and the San Francisco Business Times.

Join the Conversation


  1. A note for the journalist to help give you more credibility. Here’s a note quote from the article.

    “Between January and March of 2020, officers found illegal contraband only 34 percent of the time on Black people and Latinos, respectively, based on those searches — meaning they found nothing 66 percent of the time. ”

    Providing one side of the story without the other leads people to think the author is leaving out information to prove their point. I’m not saying you did this on purpose, but it appears this way which creates doubt from folks looking for ways to pick apart the data.

    Was the amount of contraband found on White and Asian populations more or less than 34 percent? IF a lot more, then your statement would make sense. If the same or less, then it means the police are pretty consistent. Leaving out this information gives the reader a sense that the author doesn’t want to provide it. That leads folks to think that if the data was left out, it must prove the author wrong.

    I realize I’m only one person, but if I’m thinking this way, I’m certain others do as well. Must be the lawyer in me. Leave no doubt and keep up the hard work!

    votes. Sign in to vote
  2. This article seems to depict the problem in “Black and White”?? Are those the only two groups that police interact with? Article mentions ‘Latinos” once and also POC, but never Asians — I guess Asians of all strips don’t commit crime (Mr’s Hui and Wong notwithstanding). But that omission just makes the picture almost pointless.

    I know statistic paint a very nebulous picture. I’m curious why the parole or probation is a factor in stops – how does a cop even know someone is on parole/probation, unless they actually recognize the person?

    It would be interesting to know the ‘why’ of why these people were stopped. “jaywalking”. “looked sketchy” “sporting paraphenalia” “tail light out”. Are all these stops totally subjective (DWB)?

    And one other number I’d like to see — how many stopped were actually residents of SF!

    votes. Sign in to vote
  3. I’ve attended several court hearings in SF and know that as a condition of probation/parole, in lieu of incarceration, people elect a warrantless search condition with or without probable cause. Perhaps people should elect incarceration as opposed to being set free into society with such imposing conditions of release.

    votes. Sign in to vote
  4. I would also like to compare Police stops and searches Vs crime rates in the City.
    Let’s say women are committing the vas majority of crime in a city but the police are still stopping and searching men. That would be a problem.
    But if Asian women are rarely committing crime and the police are rarely stopping them that would make sense.

    votes. Sign in to vote
  5. I think that we can conclude that after several terms on the Police Commission, progressive darling Petra de Jesus has proven that this police commission, this department and this union are incapable of being parties towards meaningful police reform.

    If de Jesus is the best they can do, remember when Campos warmed a seat on the Commission and was elevated to the Board of Supervisors after making no waves, or when Ammiano appointed Theresa Sparks in a deal cut with Toklas to give Haaland the endorsement in hopes of taking the D5 seat, then it simply ain’t happening with these pieces on the table.

    We cannot defund the SFPD fast enough. We need to be hiring social workers with specialties in psych and substance issues. And we need to reconceptualize the SFPD away from the model of slave catchers and Philippine counter insurgency veterans at the dawn of Jim Crow.

    votes. Sign in to vote
    1. Jesus Christ, Marc,

      That’s a great comment.

      One of your best of thousands.

      Thank you for being my friend.

      And, for defending me when Julian Davis threatened to thrash me.

      Remember that?


      I miss Julian.

      Hated the beatings he took just for being randy as was I.

      Read, ‘Kiss My Gay Ass’?


      votes. Sign in to vote
  6. I assume these meetings are live-streamed or otherwise on line. How does one view them?

    votes. Sign in to vote
  7. More unsettling to the law abiding citizens will be the precipitous drop in all the numbers. Why will all the numbers decrease significantly?
    1. Defunding the police means less numbers?
    2. Something else?

    I select # 2

    votes. Sign in to vote
  8. What is the “contraband” being found/seized after stops? What other than probation or parole status serve as “reasonable suspicion” or “probable cause” that serve as searches. These are very elastic terms and it would be helpful to understand how they are used in real life situations.

    votes. Sign in to vote
Leave a comment
Please keep your comments short and civil. Do not leave multiple comments under multiple names on one article. We will zap comments that fail to adhere to these short and very easy-to-follow rules.

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