We analyzed data on 1,127 killings by police in 2020 - finding alternative responses to mental health, traffic violations and other non-violent situations could substantially reduce fatal police violence nationwide. A thread (1/x). policeviolencereport.org
Working with an incredible team of data analysts (h/t @moncketeer@MaryLagman@KirbyPhares@backspace), we compiled media reports, official statements and other info on how each incident began, what reportedly happened during the encounter and whether officers were charged.
First, 2020 was largely in line with a longstanding pattern of ~1,100 killings by police per year since at least 2013. The lockdown, economic crisis, changes in crime, etc didn’t appear to change this trend, the overall pace of fatal police violence in America.
Racial disparities in fatal police violence in 2020 were also clear and consistent with past years - Black people were killed by police disproportionately and were more likely to be unarmed when killed.
The majority of people killed by police in 2020 were killed after being stopped for an alleged traffic violation, disturbance, mental health crisis, other reportedly non-violent offense, or in situations where no crime was reported or alleged (car accident, wellness check, etc).
The initial 911 calls can differ from what happened after police arrived. Arrival of police often escalates the situation- e.g. turning a mental health/wellness check into a barricade situation. Proven alternatives to police should respond to these calls: usatoday.com/story/news/nat…
Last year, police killed 121 people after pulling them over for a traffic violation. Cities like Berkeley are moving towards a model that removes police from traffic enforcement, which could save lives. theappeal.org/traffic-enforc…
Colorado is considering legislation (SB21-062) prohibiting police from arresting people for misdemeanors, traffic violations and other low level offenses. 80% of all arrests. This could impact a huge number of cases where police currently use deadly force. leg.colorado.gov/sites/default/…
And, just like past years, only 1% of cases where police killed someone in 2020 have led to officers being charged with a crime. The murder of George Floyd was one of only 16 cases where officers were charged last year.
We were able to identify the officers by name in 444 of the 1,127 killings by police. Based on media reports, at least 14 cases involved officers who had shot or killed someone before and 5 of those officers had shot or killed multiple people before.
Statewide restrictions on police personnel records prevent the release of officer’s past misconduct history in all but 13* states.
In the states where this information is publicly available, the worst officer is Officer John Nobles of Jacksonville Sheriff’s Office. Killed Leah Baker last year in his *5th police shooting*, including killing a 16 year old child in 2015. He’s still employed as an officer today.
Removing police from traffic enforcement and mental health calls could impact nearly 1 in 5 killings. If police responded to remaining situations without killing people who were not allegedly threatening officers or civilians with a gun, police killings could be reduced 55-60%.
And that’s a conservative estimate based on info that has been released publicly, often by the police without video evidence. We’ve seen police overstate a “threat” or cover up details to exonerate themselves. The actual impact of alternatives/changes would likely be even bigger.
There’s growing evidence that there has been a substantial decline (~30-40%) in police shootings/killings *in major cities* since the protests started in 2014. I’ve written about this here. But WHY the reduction in cities and why not other places? Well... fivethirtyeight.com/features/polic…
Now this is where it gets complicated. Because the places that saw the biggest reductions in police violence seem to share a set of common factors, some of which might’ve contributed and others which might not have. For example...
Arbitration is essentially the qualified immunity of officer discipline. While qualified immunity prevents families from receiving financial payouts from the city for misconduct, arbitration prevents cities from actually holding those individual officers accountable.
Because holding officers accountable has historically been the exception, not the rule, both qualified immunity and arbitration tie current practices to past precedents in a way that blocks the expansion of financial or administrative accountability for police violence in the US.
Both qualified immunity and arbitration are important practices to target for change via legislation. The focus on ending qualified immunity - which won’t actually hold those officers accountable (due to indemnification) must be expanded to include a focus on arbitration as well.
This isn’t true. There are three primary databases that track killings by police - Mapping Police Violence (ours), FatalEncounters, and the Washington Post. Between those, WaPo is the least comprehensive and the only one which doesn’t include non-shooting and off-duty killings.
What you’re referencing is the fact that the WaPo database in particular does not comprehensively track killings by police and is cited frequently in ways that underreport the scale of violence (WaPo also misses some on-duty fatal police shootings as well btw).
The WaPo data also underreports unarmed people killed by police. They don’t categorize people holding objects that aren’t weapons as unarmed and also don’t track the huge proportion of unarmed people who are killed by police tasers, restraints, etc and not shootings.
What’s the plan B for people to be able to vote by mail if USPS doesn’t function?
I’m seeing a lot of responses that people should just drop off their mail-in ballots but if USPS doesn’t function/is delayed substantially then won’t many people simply not receive these ballots in the first place?
I know several people personally who requested mail-in ballots in Florida in the (extremely close) 2018 election and never received the ballots in the mail. Many didn’t end up being able to vote as a consequence.
The Bureau of Justice Statistics estimates 53 million people had contact with police nationwide in 2015. This report is key to understanding how to reduce encounters with police overall. A thread. bjs.gov/content/pub/pd…
15% of all police contacts (8M/year) were in response to car crashes. The rest of the contacts were evenly split between people who called police and people who were stopped by police. White people had *more* contact with police overall - mainly because they called police more.
Here’s a breakdown of who called/initiated contact with police. White people were more likely to call police to report a possible crime, to report medical emergency/situations that wasn’t a crime, and to seek help for another reason. But who gets stopped by police more...
The methodological choice made in some academic circles to “adjust” rates of police violence based on rates of violent crime is a clear and pernicious example of scientific racism. A methodology designed to artificially and arbitrarily erase the existence of racism in policing.
Why is this choice so problematic? Violent crime doesn’t sufficiently explain police violence. Only 1-3% of police calls for service and 5% of all arrests are for violent crime. Violent crime rates don’t determine whether a city will have a high rate of killings by police.
Don’t take my word for it. Read the actual research literature. Here’s a recent study’s findings: