DERAY MCKESSON: A lot of people don't know is that any number you've ever heard about police violence comes through the aggregate of media reports. If you get killed in this country and a newspaper doesn't write about it, it's not covered on like a blog or like a TV or something, you literally don't exist in the data set. The federal government doesn't collect information about police killings in any systemic way. We can tell you the rainfall in Missouri in 1830, and we can't actually tell you how many people got killed, like, as a hard fact last year. We don't know it. What we know is like the aggregate of media reports, these incredible activists years ago set up these two big databases that essentially called like an advanced Google alerts of like police killings, and that is the source data for everything that you've ever seen on police killings. Some of the biggest databases that you might have heard of are like the Washington Post Database, fatal encounters killed by police.

We created Mapping Police Violence to create the single stop database that had the most comprehensive data about police killings. If you think about the Washington Post Database, for instance, they only have killings by officers on duty that use a gun. So say for example an officer goes home and runs somebody over with their car, like that's not counted. Say somebody is on duty and the officer runs you over with their car, not accounted. Eric Garner's death is not in the Washington Post Database. Why? Because he wasn't killed with a gun.

So we wanted to say that like whether you got killed by a taser, a chokehold, whether the officer was at home and like killed his wife off-duty, we consider all of those to be symptoms of the same sort of root problem. So that's why we created the database. And what we know is that, left to their own devices, that police will just never report this data. There are times where the state of Florida has reported zero police killings in entire spans of years, and you're like "We know that's not true, like we could just look at the news and see it wasn't zero!" So the data actually is really important for us to help locate what the problem is and what the solution should be.

And the last thing I'll say is that we have to figure out how to start talking about police violence beyond death. So we know that the police inflict damage in communities in ways like sexual assault, verbal abuse, those sort of things, and the data we have most readily available is about death, but because we only focus on death with the data, we're losing how the police impact women, how the police impact LGBT communities, like any other ways that don't result in death but are still really bad—and we have to figure out how to do that. One of the limitations is that most police departments definitely don't make that data publicly available or don't collect it in any systemic way. So you think about police departments like Baltimore where so much of the data is on paper, you're like "Well, who is sitting down analyzing ten million records on paper? Nobody right now," and that becomes like a challenge.

So what we found were a couple myth busters. We found things like there's this idea that community violence and police violence are related, so in communities where there's just a lot of violence people say that "the police just have to be there, because the communities are violent and so the police must be there. And because the police must be there it's just more likely that they'll probably engage in violence against communities." And we found that that's just not true. There are places where there's a lot of community violence and almost no police violence, and the inverse is also true that there's no real relationship between community violence and police violence. We also found is that black people are actually more likely to be unarmed than any other race of people who are killed by police.

So there's this myth that black people are just like carrying guns around and like they're in the presence of police, and that actually isn't true. And with regard to policy we found out a lot of things. So we created the first public database of use of force policies and police union contracts in the country because we were trying to figure out like why are the police not accountable? Is it really that prosecutors and mayors just don't care? Like what is it?

What we found is that there's literally just a different justice system. So you think about places like in Maryland, in Maryland the law literally says that a citizen can file an anonymous complaint against an officer for everything except brutality. I don't even know what that means. In California there is a law that says that any investigation of an officer that lasts more than a year can never result in discipline. Like I don't even know what that means. In places like in Baton Rouge you can't file a complaint over the phone against an officer. There are places where you have to file an affidavit, like where would you go to get an affidavit today? I don't know.

There are all these things that almost guarantee that officers won't be held accountable, and we shed light on them for the first time with this big database trying to help people see like - and actually you could get a great prosecutor, but if the rules are set up against everybody the great prosecutor is still hamstrung. You can get a mayor, you can get a city council, but if we don't focus on these structural things that aren't necessarily the most "sexy," but if they're actually the most salient in terms of outcomes like we'll never win. And that's why we created Mapping Police Violence to get the data and then Campaign Zero to talk about the policies and solutions that actually might change the system of accountability.

So we made the first valid database of use of force policies in the country and it was hard to get good—Use of Force policies, for people that don't know, are essentially the rules that police use to say whether they can use force against somebody, anything up to (and leading to) death or murder. And what we found is that, A, it was just really hard to get the record. So our data set right now is the hundred biggest cities in the country and some places sent us back almost entirely redacted policies, some places, after a long battle we got them, and what we found are a couple of things: one is that one of the things that we look for is do you ban chokehold? We know that there are almost no instances where like it makes sense for an officer to restrict somebody's airway. We also found that there's some slipperiness that happens and New York City is a great example of like chokeholds are banned in New York City. "Eric Gardner," the police would say, "was not put me in a chokehold, he was put in a stranglehold and strangleholds aren't banned in New York City." And you're like people experience them all the same, he still died.

So we mapped that across places. We also map is there a continuum of force? Is there like an officer has to use a verbal warning then use something else? There is a lot of places where there isn't a continuum, it's sort of like you just use force, and you're like "Well that doesn't really make a lot of sense." We also mapped to look at: can you shoot in moving cars? What we find is that when you look at the data the police would have you believe that they respond, that they're always responding to bank robberies, like every single 911 call is a bank robbery, and the data just actually doesn't support that. So when you see some of the stuff that is leading to these car chases it actually just isn't worth the collateral damage. It's not worth somebody getting shot, their legs still being on the gas, and them running pedestrians over. Like none of it is worth it.

Especially in a time where there's like helicopters, like if you wanted to find the person you're probably going to find them, so we should ban shooting into moving vehicles. Like we are mapping all of these things so that we can say to departments, A, we know that there's a problem in your city because we have the data on police violence. B, there is structural things that you can actually change that statistically suggest that they'll have a better outcome in those places, so we mapped them. Does your partner have to intervene if they see you using force unnecessarily? We think that they should. Do you have to engage lesser force before you use lethal force? We think you should. We're up against a system that's really challenging. The police for decades have been taught something called the 21-foot rule. I don't know if you've ever heard of the 21-foot rule, but they get trained that if somebody is within 21 feet of you they can kill you. That's like the rule that they get trained. So you're like, why did you shoot the guy with a knife? And they're like "we were trained that if they're within 21 feet of us they can run quick enough to kill you." And you're like, who is making this stuff up? And like it's been disproven. It's like crackpot science, but those things are actually like really dangerous and we wanted to say :what are the policies and solutions that either lead to these outcomes or that we can change?