How Roland Fryer’s Controversial Harvard Study on Racial Bias by Police Actually Shows Negligible Bias (or Brutality)

I’ll crunch the numbers. You can comment and debunk.

It’s been a month, and much has been written (often reactionary) about the new study (“empirical analysis”) on policing by a MacArthur “Genius” Fellowship winner Roland Fryer, a renowned economist from Harvard. The New York Times headline and detailed charts got the most coverage (and nearly 1300 comments): “Surprising New Evidence Shows Bias in Police Use of Force but Not in Shootings.”

I even made graphics (see above and below) replicating the use-of-force data New York Times editors placed at the top of their story, but added a new orange column with numbers I crunched on the right-hand side. The original graphic I cynically believe was positioned for readers “first” and prominently to tell them, “Hey, don’t worry: We’re still highlighting and insinuating evidence of police racism, where use of force is 16–25% MORE likely to happen to African Americans.” The “24% LESS likely to be shot by police if black” chart and statistic, Dr. Fryer’s second major conclusion and obviously the most newsworthy, was positioned far below by The Gray Lady. My analysis will show how both data sets show a trivial difference — being 20% more or 20% less with small raw numbers doesn’t mean much — and the overall occurrences predictably align with crime statistics.

People still freaked out, specifically about the evidence dealing with police shootings — it went directly against the massively popular narrative that began after Trayvon Martin’s death and erupted in Ferguson, Missouri. Fryer himself tried to answer reader questions. Though, it doesn’t take a “genius” to conclude for the past few years that if a demographic that is 25% of those being killed by police is also 27% of the those arrested (36% for violent crimes), and 43% of the persons killing cops, then there might very well not be nationwide systemic racial bias against that demographic.

Americans and media could also do some homework and point out, for example, that according to the Centers for Disease Control and Prevention the rate of police killings per million for African Americans has fallen by 70 percent over the last four decades but the risk remains as high or higher for whites, Latinos and Asians. So-called “state-sanctioned violence” got better for some since the 1960s, but not for others.

Regardless, few arguments against Fryer got to the heart of the other matter on use of force which I’m addressing here: “How exactly are police approaching and treating citizens? Per encounter (‘stop’), what does the data translate to?”

It is by taking this “next step” with Fryer’s data where a non-opaque image emerges on how use of force is applied “on the ground” and whether racial disparities exist. My conclusion: It’s negligible.

And evidence of prevalent police brutality? Arguably, it’s not there. Unless you believe that police placing handcuffs roughly 3% of the time on BOTH white and black persons they stop in New York City constitutes unequal and pervasive “rough” treatment. This is ironic in the context of what Fryer stated publicly to The Wall Street Journal upon delivering his new data, “For all the eerie similarities between the current spate of police interactions with African Americans and the historical injustices which remain unhealed, the current debate is virtually data free.” And I wonder about those similarities with Jim Crow and other “historical injustices” when that “roughly 3%” figure I calculated is from Fryer’s own data, which mirrors other verified raw numbers of U.S. law enforcement actions.

Also, it’s worth noting that Fryer’s Wikipedia page is currently updated with the following:

In 2016, the National Bureau of Economic Research published a working paper by Fryer concluding that although minorities (African Americans and Hispanics) are more likely to experience police use of force than whites, but that they were not more likely to be shot by police than whites.[17] The study generated considerable controversy and criticism.[18][19][20]

Indeed, it did. Mostly, I’d hypothesize, because stakeholders and the public are married to preordained notions, propping up narrowly defined evidence to defend their point-of-view a (however thin) and giving a knee-jerk dismissal of new evidence (however thorough). Or set an extremely low bar for charges of racism or unfair treatment. This is a trend, I theorize, spawned in corners of academia and think tanks and running rampant in news organizations, activist groups, and social media — often by over-embellishing certain findings and ignoring others.

There’s no solid evidence police violence has increased in recent years (despite what academic papers titled “In the Shadows of the War on Terror” might imply), or is disproportionately directed at minority Americans at rates vastly higher than the crime rate. In fact, the most credible evidence says the opposite: It’s gone down significantly for blacks, but not for whites. Let’s go back to lethal force by cops for a minute. Indeed, police-involved shootings in Chicago today are four to six times less than they were in the 1970s, as a 1982 sociology text confirms. Black conservatives like Larry Elder make the same point, and citing Centers for Disease Control and Prevention (CDC) data, one of the more trusted U.S agencies, he asks, “So what’s driving this notion that there is now an ‘epidemic’ of white cops shooting blacks when in the last several decades the numbers of blacks killed by cops are down nearly 75 percent?” According to the same CDC data, the rate for all other racial groups have remained the same, or even went up slightly. Indeed, it’s noteworthy that the liberal group Center on Juvenile and Criminal Justice refers to the same CDC data showing the rate of black people killed by law enforcement is three times less than 40–50 years ago. Yet facts like that are largely ignored and almost never mentioned in mainstream news reports. In the age of “viral videos,” the “isolated instance” of police violence gets twisted into an epidemic — and nearly exclusively happening to people of color. That’s not true. Especially knowing that two out of four people killed by police are white, and only 10% of all people killed were unarmed — and some of those were police-fighting assailants were still lethally dangerous.

When cries of racism pop up, maybe it can be be pointed out that 80% of the people black officers kill are black. Also, black cops shoot their guns 3.3 times more often than their counterparts according to a 2015 University of Pennsylvania Department of Criminology statistical study — again, a reflection of the dangerous areas they serve more than any other factor. This indisputably counters the narrative that it’s white police officers using lethal force against black and brown people disproportionately compared to their peers. In fact, the University of Pennsylvania study by criminologist Greg Ridgeway says point-blank that while diversifying police ranks adds legitimacy, “Recent research suggests diversity does not make officers safer and this research does not suggest diversity will reduce the risk of police shootings.

And it’s worth noting: Applying “use of force” doesn’t mean it’s “unjustified use of force.” These are trained techniques used for compliance and the safety of the officer and citizens, when used correctly. I do sense some people — activists mostly — live in an imaginary world where police can’t protect themselves, and every perpetrator cooperatively goes to jail as easily as a ten-year-old heading to Disneyland. If use of force happened 1 in 25 times for both white and black persons per arrest (about 4% of the time), wouldn’t most people say that’s both low and negligible? That’s what the Center for Policing Equity found this summer, as well as Dr. Fryer, regarding citizen stops.

Thus, I won’t delve into Fryer’s self-proclaimed “most surprising result of my career” of NO racial bias in police shootings. I wrote about those results earlier in August when discussing Chicago’s recent police shootings and the rare use of deadly force, which contradicts news headlines and far-too-common assumptions. In fact, some paragraphs here are directly lifted from the “addendum” content at the bottom of that post. I want to focus on the area that DOES show racial bias — and then show that it’s really quite small. (Again, refer to the graphic at the start of this post.)

I do ask: To the many befuddled critics of Fryer’s “no racial bias” shooting data, from Slate to Snopes, can you also downplay his “yes there’s racial bias” use-of-force data from the same analysis? Because I don’t believe you can use one and not the other, especially when those numbers match up. That’s cherry picking for a preconceived narrative.

The Fryer use-of-force data itself is credible as it’s nearly identical to similar reports, such as one put out this summer by the Center for Policing Equity titled “Race, Arrests, and Police Use of Force.” This analysis has been shown as clear evidence of racial bias by the Daily Kos and the New York Times — even though I argue it’s not at all. Still, I prefer Fryer’s openness and his team’s 3000 hours of data-analyzing and clarity of presentation, even if it is a “working paper” and not yet peer-reviewed and only conservative websites tend to highlight the rigor of the more “controversial” shooting analysis. (If Fryer would have only published his use-of-force analysis, no one would have blinked.) Both papers utilize data from diverse cities in the United States, and Fryer focuses on New York City for use of force.

Regardless, these are twin reports that claim all-things-being-equal (applying crime benchmarks to the mix), the use-of-force rate is still around 20% higher for blacks than whites (a rate of around 1.2, or “1.2 times more likely, all things being equal”). The rate higher is 3.6 (Center for Policing Equity) or 3.335 (Fryer) if one does not control for highly important interaction points with law enforcement, such as arrests and known crime data. So we’re in the same ballpark, with or without controls. But on the street — and this is my main point — “20% higher” translates to about the SAME percentage of use-of-force actions by police for blacks and whites, within a margin of 1% (look at the orange column on the right).

How much the same? How often is use-of-force taking place? According to Fryer’s data during Giuliani-Bloomberg’s New York City, suspects are “pushed to the ground” once out of every 73 stops if black (1.3% of the time) or 1 in 87 times if white (1.1%). And the Center for Policing Equity’s definition of “use of force” by police says it happens 3.6% of the time for white people and 4.6% of the time for black people per arrest.

That’s the “next step.” That’s reality. It’s tangible and something an average reader can visualize. Why isn’t that done by media or academia?

Quick Aside, Can We Trust The Numbers? Answer: Yes. (Also, please feel free to skip to the next yellow & orange graphic if these details weigh you down.)

Unfortunately, far too many news outlets, from the New York Times to CNN, simply emphasize the misleading and hyperbolic “Police use of force among blacks is 3.6 times as high as among whites” (or “360% higher”) and leave it at that. I don’t and won’t. Even while recognizing the racist and economically exploitative American history that leads to poverty, spawning crime, and therefore shaping those interaction points. With that in mind, it’s uncontroversial in criminology circles, but it certainly is in public forums, that African Americans commit much more crime than their percentage of the population, and white Americans slightly less.

If a demographic is arrested, say, 3.5 times more and it’s reflective of actual crime levels, there’s going to be 350% more moments use of force as a baseline. If it’s shown there’s 175% more moments of use of force then police are exacting HALF the proportional use of force on that demographic. Some would even call this hypothetical result a form of “reverse discrimination” — an unhelpful term, but one gets the point. The New York Times muddies this issue often, as the mainstream media regularly does, by not providing context of criminality and obfuscating.

Use of force by police is estimated, in several studies, to be about 2.5 times higher for African Americans in raw numbers compared to the general population, and The Guardian and The Washington Post consistently shows killings are 2.0 times higher (13% of population, but 26% of deaths). The difference with whites is slightly greater. Think about those two data points, 2.0 and 2.5, and ask yourself: “Shouldn’t those numbers be fairly close?” But I’m often fed the argument, “Well, you don’t know the many unofficial times police mistreat black people, so your analysis is highly limited and you should question use-of-force data.” Yes, racial bias and line-crossing brutality by police does occur. Police lie. Though, I argue it’s isolated and often shared with the outside world through viral videos and anecdotally with little evidence of one key aspect: prevalence. This stands in contrast to current systemic arguments of racial discrimination, to name some examples, in housing, hiring practices (the bias against “black-sounding” names, also see: Roland Fryer’s research with a Freakanomics co-writer), and prison terms (10% longer sentences for blacks than whites).

The linchpin underneath my argument that use-of-force data and death-by-police data are intertwined rests on two uncontroversial points: 1. Yes, police can theoretically hide abused people through non-reporting; 2. No, they can’t hide dead bodies.

But how much are police “hiding” data for one group of people (black males), but not another (white males), the implication behind point #1? Logic dictates that if one set of violence, like use of force, really happened 4.0 times more for blacks than the general population (and, again, it’s documented as 2.5) then police involved killings — the most extreme use of force on the continuum — would also have to be 4.0 times higher (and it’s basically a fact it’s 2.0). I’ll also point out that anecdotally, nearly every person I’ve polled said they believed police killings of blacks make up 50–90% of total, which would be 4–7 times higher than their percentage of the population. They’re surprised when I say, “No, despite what social media and cable news tells us, it’s twice the black population rate, so 25%.” Bush and Obama-era Justice Department oversight, usually instigated after high profile shootings, doesn’t necessarily benchmark stops to crime demographics and may have unintended consequences of even more crime, though that’s not fully tested. For now, the crowdsourcing data vigorously gathered by The Washington Post and The Guardian is the gold standard until the FBI and law enforcement get their act together in terms of data collection. Thus, I argue the “common knowledge” or “ethnographic” explanation — often used as an argument stopper — of racially motivated police harassment (“white privilege”) is deeply suspect. I’ve yet to be convinced that police are beating people up at double or triple the rate they’re killing them, though I’m open to any evidence or argument. Also, those are the levels necessary (4.0–6.0) to indicate clear racial bias and systemic abuse towards African Americans. What’s telling is the rate they’re being killed (2.0) is possibly lower than to be expected based on crime and encounters with police, as I’ll get to shortly.

And to critics who say, “Well, this is data provided by ‘honest’ police departments with nothing to hide (wink wink). How can we trust it?” The problematic police departments might not cooperate, for example, or some police may be “cooking the books” on what they document. Besides, arrest data isn’t necessarily accurate, statistics are limited and don’t reveal the whole picture, and you need qualitative ethnographic studies to truly understand what’s happening on America’s streets. PBS investigations like Frontline’s Policing the Police have full access, cooperation, and the officer-on-the-street’s POV but still show “bad stops.” Even Fryer suggests it, saying in the analysis, “It is possible that these departments only supplied the data because they are either enlightened or were not concerned about what the analysis would reveal. In essence, this is equivalent to analyzing labor market discrimination on a set of firms willing to supply a researcher with their Human Resources data!” He had indicated a desire to work with Chicago crime data, for instance.

OK, I hear that. Well, you can’t hide dead or gunned down bodies, right? Compared to killings, use-of-force data shows slightly more racial bias. By looking at both, any rational observer could say, “OK, these numbers are close … and possibly should be.” And even Fryer trusted his numbers and esteemed reputation enough to trot out his paper to the world beset with the ideologically beholden. After all, much of the standard narrative that began on social media (Black Twitter) and spread through the media assumes that police are “executing” young African-American men en masse. Much of that thinking comes from exaggerating studies or statistics — or hyping tragic, and even illegal, isolated incidents perpetrated by the state.

Indeed, we can go back to the numbers around police involved shootings and deaths to prove Fryer’s data is sound and can be trusted. If the rate-of-death without controls were equal for blacks and whites the number you’d see would be “1.0.” In reality, African Americans die at hands of law enforcement in the United States at twice their rate of population (2.0); and compared to whites, they are two-and-a-half (2.5) times as likely. However, with controls, which are essential to statistical analysis, Fryer and I would argue it’s closer to 1.0 (or even .8 or .9 favoring the black populace). The New York Times posted in the middle of their story, “Mr. Fryer found that in such situations, officers in Houston were about 20 percent [.8] less likely to shoot if the suspects were black.” The use-of-force disparity rates that Fryer and other studies present (1.2 with controls, 3.335 without) are consistent with the impossible-to-miss evidence of police shootings and deaths. In the next section (Breaking Down Fryer’s Data “On the Ground”), I’ll show how 20% greater or lesser can appear negligible with low number of occurrences, which is what we’re dealing with in violent police interactions.

In fact, use-of-force percentages are slightly higher, even though that data is easier to “hide” by law enforcement because they self-document. Could they be putting their pens and contact cards down for some stops? Possibly, as it’s easy to do. But where’s the evidence that it’s a prevalent act or is not uniform in all racial categories? Is there an assumption anywhere that one-third of stops, let’s say in Newark, are “official” and two-thirds are “off the books”? Are there any news stories of a police officer shooting someone and the later discovery they tried to cover it up, such as dumping the body? I haven’t heard of any. The closest thing to it is the 100% immoral and illegal actions of Officer Michael Slager towards Walter Scott, the one case of dozens I’ve seen where it’s plain-as-day there’s a cover-up by the officer involved. Though, why should cops worry about any-of-the-above when the norm is 99% of the time police face no discipline? This is a significant problem, too.

Thus, in raw numbers, are we to understand the fact police kill or shoot Black Americans two-and-a-half times (2.5) as often as White Americans is hard evidence (with mortal bodies to prove it), but hearing that they’re about three-and-half times (3.5) more likely to be subject to use force sounds ostensibly low? I’d say no. Wouldn’t it make sense that violent use-of-force situations by police could escalate to a shooting, of which there are more than two thousand annually, thus tying together those data sets? Can one reasonably argue that the nature of implicit bias allows a “stop valve” to keep officers from shooting or strangling one demographic disproportionately but be OK delivering significantly higher moments of use of force to that same demographic. To me, verifiable police violence like killings (or shootings) are the “canary in the coal mine” to indicate injustice or not. Or at the very least show the percentage of interaction. And seeing how blacks are more than 3.5 times as likely to commit many violent crimes than whites in many categories (8 times more for robbery and murder, for instance), then where is the evidence of blatant injustice? All these numbers pretty much align. The only counter-argument is anecdote.

I ask all serious-minded and fair people: How can bias be a bigger factor than behavior in outcomes with police if we’re talking 20% difference with controls and 350–360% difference without controls? How can it be remotely controversial to say, “Well, if a certain race is being described to police 3.5 total times more via 911 calls with matching arrest rates — and for murder, rape, assault, larceny , etc. as opposed to arbitrary drug arrests — then maybe there’s going to be 3.5 more moments of use of force.”

It’s a canard to simply attack racist police or claim “institutional racism” leads to these outcomes. Studies conclude similarly stating, “Relative to violation frequency as reported by crime victims, the likelihood of arrest for white and black offenders is roughly equal.” And in a 2016 Criminal Justice Review published paper that went further, “Results indicate that race does have an indirect effect on police contact, but it is white individuals who are more likely to be questioned and arrested.” Basically, there can be an argument that blacks get away from being caught by police more than whites in some key categories. Of course, in areas like hard drugs and marijuana arrests, that’s not necessarily the case. But people aren’t put in prison for simple possession. According to the Pulitzer Prize-winning website Politico when correcting Bernie Sanders’ oft-presumed notion that America jails people for smoking marijuana, “99.9 percent of those sentenced to federal prison for any drug-related crime during that year-long period [2012–12] were sent to prison for something more serious than simple possession.”

And yet, even when public radio reporting points out that in NYC during stop-and-frisk, blacks are nearly twice as likely (85.4%) than whites to be charged with resisting arrest per encounter (that’s “rate” not “total”), the narrative is still spun to place blame squarely on “aggressive police.” And thus, racial bias is to blame. Again, the controls will show about a 20% difference in regards to use of force in Fryer’s study, so behavior manifests violent outcomes far more often than bias. Yet, a key question is how “controls” in studies include attitudes both police and citizenry bring to the interaction. This may not be quantifiable. For example, a Pew poll showed that for nearly 10 years only one-third of African Americans believe police treat blacks and whites equally, while seven-in-ten whites have confidence in the fairness of law enforcement. As I’ve stated before, “Issues of trust around police make this a chicken or egg situation, which I don’t deny needs fixing.”

Add the fact that twin groundbreaking 2015 analyses on police killings by the Washington Post and The Guardian both say indisputably “200% more likely to get killed if black” sans controls — i.e. African Americans are a little more than 25% of total deaths by law enforcement and just under 13% of the population. Interestingly, these same media outlets tend to spin their numbers to foment racial injustice, while a scientist with free time on his hands can write an “impartial data analysis” that makes clear, logical sense indicating there’s no racial bias.

Of course, conspiracy minded folks — and I have a few in my social network — will say there’s many times more murders and shootings of blacks committed by police that aren’t recorded. I honestly directly hear “opinions,” to say it nicely, about the CPD as if they’re the famed Vice Lords gang committing “hits” left and right and leaving the bodies. It’s true the FBI has admitted their undercount for years, but now newspapers are crowdsourcing that data. I’m receptive to uncovering potential data on all shootings from hospitals, as academics have proposed (“Who Shot Ya? How Emergency Departments Can Collect Reliable Police Shooting Data”). But I remain skeptical as even the academic proposing this radical method for measurement via hospitals can’t say the percentage would be marginally higher; and one doctor he quoted preposterously said 10% of his patients could very well be shot by police, so I smell pre-formed agenda in this approach. In verifiable numbers this year, for my city of Chicago as of the last week of August 2016, there were 17 shootings by cop out of 2800-plus total — well under 1% of total citywide shootings were by law enforcement. Even the past six years, Chicago police shoot around 40 persons annually. So are we talking 5 more shot every year in the perceived undercount? 10? 2? One can see how absurd this line of thinking becomes. Yet sometimes it derives from people with PhDs as well armchair critics across the United States.

Again, my point is “You can’t hide dead or shot bodies” and those bodies — white or black, pink or brown — match interaction points by police based on actual crime. So how can you hide a sizable number of undocumented bruised or beaten bodies from use of force when the police data nearly matches the “dead or shot” ones?

Breaking Down Fryer’s Data “On the Ground”

But back to the use-of-force data. To better understand the Harvard study by Dr. Fryer, a celebrated black economist interested in racial disparities, see the graphic below for the “stop-and-frisk” era of New York City, 2003–13. Then note my additional numbers in the right column, and the further percentages I breakdown below the image. Then ask, “Is it really that racist or excessive?”

This is during the period of zero-tolerance “broken windows” theory style of policing, started by Mayor Rudy Giulani and Police Commissioner William Bratton and continued with Mayor Michael Bloomberg and Police Commissioner Raymond Kelly. This was a highly criticized period for “racist policing” and excessive use of force, which led to the anti-stop-and-frisk Mayoral victory of Bill De Blasio. Recorded police stops have dropped from 684,000 in 2011 to 191,000 by 2013 — more than three times less! Yet, polling shows as recently as last year widespread and consistent support in white, black and Hispanic communities for Bratton, Kelly, and police issuing arrests or summons for “quality of life, low-level offenses”(a.k.a. “broken windows”).

To be a broken record, as I wrote at the very bottom of an earlier Medium post: “And is the mistreatment by ‘NYC Cops’ so extreme, in general, if your chance of getting handcuffed [per stop] is 3%, whether you’re complexion is black or white?”

3% of the time you’re handcuffed per stop by one of the most criticized police departments in recent U.S. history. And it doesn’t matter if you’re ‘Black or White,’ as Michael Jackson used to sing to young boys like myself — before he smashed a car and tore off his shirt in one of his more bizarre artistic statements. The results are pretty much the same.

How did I get this percentage? How is it “the same”? Again, divide 310 by 10,000 and 266 by 10,000 (i.e. 1 in 32 for blacks, 1 in 37 for whites), and you get 3.1% (blacks) and 2.7% (whites). This is the “next step” I’m doing with Fryer’s data published in The National Bureau of Economic Research as translated by the New York Times. Do the simple math on any interaction with New York City police (see graphic above) during the controversial “stop-and-frisk” decade and you get nearly the same percentage for black and white citizens:

• 1 out of every 64 blacks stopped will have a weapon DRAWN on them 
  (1.5 % of encounters)
• 1 out of every 77 whites stopped will have a weapon DRAWN on them 
  (1.2 % of encounters)

Or:

• 1 out of every 185 blacks stopped will have a weapon POINTED at them 
  (.5% of encounters)
• 1 out of every 232 whites stopped will have a weapon POINTED at them 
  (.4% of encounters)

You can do this 5th grade arithmetic for all those moments of “disparities” (from 16% to 25%) among the “use-of-force continuum,” which I started doing in the Center for Policing Equity Study that I debunked as a powerful example of institutional racism. Even news outlets like Fortune focus on the higher percentages, which lead readers to conclusions that racial bias is the problem, not behavior. But that shocking figure of “25% more likely to” use of force with a baton or pepper spray really just compares 5 times out 10,000 (blacks) versus 4 times out of 10,000 (whites). While technically true it’s “25% higher” likelihood of a baton beating — and that’s the number media and academia promote — you’re still talking about pretty much the same minuscule rate or percentage of occurrence, i.e. .05% (black) versus .04% (white). That’s a paltry figure that doesn’t fit the narrative (or the cliché anti-police t-shirts) of cops beating down citizens.

And I’m still not sure if other intangibles are factored into Fryer’s analysis, such as general demeanor towards cops (sometimes negative in minority communities, shaped by perceptions of not being treated fairly) or ‘revolving door’ targeted stops of known gang members — there’s lots of “above my pay grade” math going on there. I have no doubt racial profiling and mistreatment occurs. I just don’t see it translated into hard evidence from the data that is currently being sold to the world as “statistically significant” and “racial bias” and “these cops are out-of-control.”

I’m not spinning these numbers; I’m trying to break them down to their simplest, most understandable form. Frankly, I’m surprised by the low per encounter use-of-force result. That’s why the first time I saw The NY Times story and lead graphic, my reaction was, “Wait a minute. What is that saying, really?”

I also recognize that if stops are more than three times less than during the stop-and-frisk era, that’s possibly three times less occasions where “use of force” is applied — and people can debate the merits of that. But I’m not arguing that today. I’m arguing that information that shapes key political decisions is poorly shared by the mainstream media, at times.

Again, I’m just wondering: Why this context is being wiped off media and academic reports, or not being published in the first place? It’s not difficult. It creates a clear picture to the public. Shouldn’t this be the “best practices” approach in presenting information that explains how use of force is experienced?

Hard Numbers vs. Perceived Reality: Which is it?

And honestly, I have cognitive dissonance from watching HBO’s The Wire. Or reading about Baltimore’s “racist” police department beset with a “cultural resistance to accountability” (one among numerous American cities operating under the Department of Justice’s “consent decrees”). And the longtime “common knowledge” of racial profiling with the Chicago Police, an assumption upheld in 2016 by a mayor-directed task force.

But when I look at the numbers that supposedly argue “police racism,” I simply have to say: It doesn’t add up.