4 ms·

>It's uncanny how the cameras are almost always off whenever something controversial happens. I've seen this claim pretty frequently, but never any actual data

>It's uncanny how the cameras are almost always off whenever something controversial happens. I've seen this claim pretty frequently, but never any actual data, so I started tracking the stats in my own city. Since I started ~January, there have been 25 police shootings/use of force events(this includes things like tazing, bean bags, and punches/kicks that require medical treatment. It doesn't include violence against animals or desk pops/negligent discharges). 23 out of 25 events had bodycam footage. The two that didn't both involved off-duty officers, which I think is a reasonable exception to body cam rules. There were no instances of "malfunctioning" body cams, dead batteries, or officers who "forgot" to activate their cams. Even though 6 months of data from one city isn't enough to make any sweeping statements, I haven't found any studies more thorough than my own spreadsheet. If you know of any please send them my way!

kurthr 5y ago
Here's a recent report for a large department that apparently takes body cam footage seriously. 22% failed to have cameras on for "use of force" incidents... which seems like the most extreme example. If the usage was higher in any other circumstances you would have to wonder why, because it would seem sinister. https://www.latimes.com/california/story/2021-07-20/22-of-lapd-officers-failed-to-promptly-activate-body-cameras-in-force-incidents https://www.latimes.com/california/story/2021-07-20/22-of-la...
- Rebelgecko 5y ago
  >22% failed to have cameras on for "use of force" incidents IMO that's a an oversimplification compared to what the report actually says[1]. The 22% number isn't how common it was for officers to not have their cameras on, it was how common it was for officers to have cameras that were off OR the camera was on but they were late in hitting the record button[2]. Situations where the officer just straight up didn't record were more rare (that happened 3.7% of the time and if you exclude one instance that was considered justified it's only 3.2%). Funnily enough, LAPD is the department whose data I was tabulating (over a slightly different time period than this study). I think there's 2 main discrepancies between my numbers and the OIG report: I only counted if at least one officer present had body cam footage, not if every officer had body cam footage. Also, I still counted it if the footage of the shooting/use of force was in the 2 minute buffer. [1]: https://a27e0481-a3d0-44b8-8142-1376cfbb6e32.filesusr.com/ugd/b2dd23_65617839b7fc47549eb66001b06096ac.pdf https://a27e0481-a3d0-44b8-8142-1376cfbb6e32.filesusr.com/ug... [2]: Keep in mind that just because they were late to hit the record button doesn't mean the shooting wasn't captured on camera. If an officer talks to witness to ask where a suspect went, radios for help, activates the camera, finds the suspect and then a shooting occurs, they'd be considered non-compliant (e.g. https://www.youtube.com/watch?v=UDThHbrtQE4&t=355s https://www.youtube.com/watch?v=UDThHbrtQE4&t=355s). And even if the officer doesn't activate the camera until after a shooting, it'll still be captured in the 2 minute buffer (e.g. https://www.youtube.com/watch?v=WT4Ss6ASOrE&t=3m30s https://www.youtube.com/watch?v=WT4Ss6ASOrE&t=3m30s)
  - kurthr 5y ago
    Wait, so they're supposed to be on all the time (that they are actively patrolling) so that the 2 min buffer exists, but as long as they turn on the cameras immediately before using lethal force it's just fine and really shouldn't count? I mean close to 2 minutes is kinda the minimum here. All they have to do is press the record button sometime before they pull their weapon or at least within a few seconds of discharging it to have 1:55 saved. Otherwise, they can inform their victim/perp that they are going to execute them and then turn the camera on for the few seconds before as a warning to anyone who would challenge them. Is this not a corner case you considered?

kurthr 5y ago

Here's a recent report for a large department that apparently takes body cam footage seriously. 22% failed to have cameras on for "use of force" incidents... which seems like the most extreme example. If the usage was higher in any other circumstances you would have to wonder why, because it would seem sinister. https://www.latimes.com/california/story/2021-07-20/22-of-lapd-officers-failed-to-promptly-activate-body-cameras-in-force-incidents https://www.latimes.com/california/story/2021-07-20/22-of-la...

Rebelgecko 5y ago
>22% failed to have cameras on for "use of force" incidents IMO that's a an oversimplification compared to what the report actually says[1]. The 22% number isn't how common it was for officers to not have their cameras on, it was how common it was for officers to have cameras that were off OR the camera was on but they were late in hitting the record button[2]. Situations where the officer just straight up didn't record were more rare (that happened 3.7% of the time and if you exclude one instance that was considered justified it's only 3.2%). Funnily enough, LAPD is the department whose data I was tabulating (over a slightly different time period than this study). I think there's 2 main discrepancies between my numbers and the OIG report: I only counted if at least one officer present had body cam footage, not if every officer had body cam footage. Also, I still counted it if the footage of the shooting/use of force was in the 2 minute buffer. [1]: https://a27e0481-a3d0-44b8-8142-1376cfbb6e32.filesusr.com/ugd/b2dd23_65617839b7fc47549eb66001b06096ac.pdf https://a27e0481-a3d0-44b8-8142-1376cfbb6e32.filesusr.com/ug... [2]: Keep in mind that just because they were late to hit the record button doesn't mean the shooting wasn't captured on camera. If an officer talks to witness to ask where a suspect went, radios for help, activates the camera, finds the suspect and then a shooting occurs, they'd be considered non-compliant (e.g. https://www.youtube.com/watch?v=UDThHbrtQE4&t=355s https://www.youtube.com/watch?v=UDThHbrtQE4&t=355s). And even if the officer doesn't activate the camera until after a shooting, it'll still be captured in the 2 minute buffer (e.g. https://www.youtube.com/watch?v=WT4Ss6ASOrE&t=3m30s https://www.youtube.com/watch?v=WT4Ss6ASOrE&t=3m30s)
- kurthr 5y ago
  Wait, so they're supposed to be on all the time (that they are actively patrolling) so that the 2 min buffer exists, but as long as they turn on the cameras immediately before using lethal force it's just fine and really shouldn't count? I mean close to 2 minutes is kinda the minimum here. All they have to do is press the record button sometime before they pull their weapon or at least within a few seconds of discharging it to have 1:55 saved. Otherwise, they can inform their victim/perp that they are going to execute them and then turn the camera on for the few seconds before as a warning to anyone who would challenge them. Is this not a corner case you considered?