Cops Talk Council Member Into Changing Her Mind On ShotSpotter With Data That Doesn’t Actually Show It’s Worth Paying For
from the I-think-these-numbers-are-sufficiently...-large dept
ShotSpotter claims its gunshot detection tech is something cities battling gun violence just can’t (almost literally) live without. Data generated by cities paying millions for the tech often says otherwise.
On multiple occasions over the past few years, cities have terminated their contracts with ShotSpotter, citing the tech’s overall uselessness. Cops in Newark, New Jersey ditched the tech after it generated false alarms three-quarters of the time. Another town came to the same conclusion when ShotSpotter produced a 41% false positive rate. The city of San Diego, California refused to renew its $1.5 million contract with the company, claiming the tech did little more than encourage over-policing in areas already subjected to biased police practices.
Perhaps the biggest loss for ShotSpotter came in Chicago, Illinois, a city absolutely plagued with gun violence. An investigation by the PD’s Inspector General came to the following conclusion:
OIG concluded from its analysis that CPD responses to ShotSpotter alerts can seldom be shown to lead to investigatory stops which might have investigative value and rarely produce evidence of a gun-related crime.
[…]
The CPD data examined by OIG does not support a conclusion that ShotSpotter is an effective tool in developing evidence of gun-related crime.
The company continues to tout its success while real-world applications tend to indicate otherwise. In Houston, Texas, lawmakers have been reevaluating the city’s relationship with ShotSpotter. In January 2022, the city’s council voted overwhelmingly to start spending taxpayers’ money on unproven tech. There was only a single dissenting vote.
All but one council member — Letitia Plummer, At-Large Position 4 — voted to approve the contract.
Plummer said she voted “no” because she hasn’t seen any data to indicate the technology would lead to safer communities.
This despite the fact that other council members seemed to have some serious reservations about ShotSpotter, but just decided to go with the flow.
At least two council members who voted to approve the contract Wednesday appeared to agree that the program will likely not prevent gun violence in the city.
Responding to a woman from northeast Houston who testified in support of ShotSpotter, District B Councilmember Tarsha Jackson said she was voting for the technology in response to feedback from constituents.
While the $3.5 million is only a rounding error in the Houston PD’s $1.02 billion budget (20% of the city’s total budget), there’s no reason local lawmakers should be paying for something that even they agree probably doesn’t work. While Houston legislators gave SpotShotter the green light, San Antonio officials were ditching the program after doing the depressing math that showed the tech was costing taxpayers about $136,500 an arrest.
Given this background and the initial hesitance of council members who ultimately decided to vote in favor of spending money on ShotSpotter, one has to wonder why the lone holdout in the 2022 vote has changed her mind.
Houston City Council Member Letitia Plummer voted against the city of Houston using the gunshot detection system ShotSpotter technology in January 2022 because, she said, the data gave her a negative impression of the technology. Her feelings have since changed.
At the end of last month, when the council had to vote again on funding ShotSpotter in order to fix a clerical error that left them underpaying for the service, technically, Plummer was out of town, so her vote counted as a default “yes.” But in an interview, she said that had she been here, she would have voted in favor anyway.
“According to what I’m seeing, the data is showing positive,” Plummer said. “I believe that it’s working in Houston. This is the data given to us, and it’s all I can go by.”
Really? Because the cited data shows plenty of data, but very little that shows the tech is helping reduce crime.
According to a slide provided by Plummer’s office, since ShotSpotter was implemented in the Southeast Patrol Division in December 2020, the technology has had 5,203 published alerts, leading to 1,026 offense reports, 94 arrests, 63 misdemeanor charges, 4,216 fired cartridges recovered, and 93 guns recovered.
If this is all there is, it isn’t much. 94 arrests on more than 5,000 reported gunshots means the tech’s arrest rate is less than 2%. There’s no mention of felony charges, which leaves only the 63 misdemeanors, something that suggests reported gunshots aren’t leading to meaningful arrests. Recovering 93 guns means nothing without more context. What the stats show is a lot of busywork is being generated by ShotSpotter, but very little of it will have any effect on violent crime.
What the Houston PD handed to Plummer shouldn’t have been enough to change her vote. But that’s what has happened here. And taxpayers are now paying for the image ShotSpotter has cultivated for itself (fearless high-tech crime fighter!) and the hours wasted by officers rushing to reported gunshots to… um… pick up shell casings and engage in misdemeanor arrests.
Filed Under: guns, gunshot detection, houston, houston pd, letititia plummer, texas, wasteful spending
Companies: shotspotter
Comments on “Cops Talk Council Member Into Changing Her Mind On ShotSpotter With Data That Doesn’t Actually Show It’s Worth Paying For”
So, if I read that right, less than 20% are readily linked to any wrongdoing having occurred? I’ll also note that it doesn’t say anything like “gun related offense reports” which means… it could be as bad as ~20%[1] of the time officers are able to “salvage” the situation by finding someone near by and some offense they can write up (which might be totally unrelated to the alert, other than it got officers looking around in the area)?
[1] This is speculation. My point is that the report seems very vague. Lots of ways a causal reading to come a conclusion that might not be real.
Missed opportunity
The immediate flood of posts on Nextdoor about “was that a gunshot?” in the most innocuous suburban neighborhoods where it was almost certainly a backfire or firework seems like Shotspotter is missing a trick by not targeting HOAs for installations.
On the other hand, police departments might monitor social media and get the same information that Shotspotter provides with the added benefit of having a user to ask for followup.
One might wonder why PDs can’t rely on 911 reporting, but the unwillingness of police to respond to unsexy calls that won’t result in immediate arrests, or super willingness to show up guns blazing could factor into people being reluctant to call the cops
Not like she'll ever be on the wrong end of a false positive...
“According to what I’m seeing, the data is showing positive,” Plummer said. “I believe that it’s working in Houston. This is the data given to us, and it’s all I can go by.”
Given the accuracy of the tech sure isn’t in it’s favor I can only assume either the company just flat out lied to her or they pointed out that it sure would look nice if she was able to boast about how she’s ‘tough on crime’ come next election and it’s not like it’s her money that’s being spent.
Hucksterism and snake oil are common ingredients in modern policing technology. Disappointing but this has been the case ever since policing was invented.
Re: The digital equivalent of probable-cause on four legs
It’s even worse than that I’d say as police have a vested interest in technology that gives them as many ‘hits’ they can use to justify searches/arrests as possible, all the more so since there’s basically no penalty for getting it wrong even when they can’t just offload blame onto the tech in question.
A working brain is, unfortunately, not required for most elective positions…
More magic computer pixie dust
Several communities not part of the City of San Antonio had ponds were water fowl would land and congregate. In some of those areas, propane devices were installed to, at irregular intervals, release a quantity of propane, igniting it, and making a “BOOM!”.
3 guesses as what could possibly go wrong there and how long it took PD to ignore Shot Spotter reports for a mile around that area.
So the real question we need to ask…
Did the vote change because of a contribution or because of someone playing the ‘hates police’ & ‘you’ll be unsafe’ card thats been so popular.
you’ll never be able to put sense where there’s only room for BS and ‘contributions’
Re:
“Never attribute to malice what can adequately be explained by stupidity.”
“It is difficult to get a man to understand something, when his salary depends on his not understanding it.”
“Between evil and stupid, take evil. Because stupid, my friend, is all the time, but Evil only gets uppity now and again and is usually predictable.”
you’ll never be able to put sense where there’s only room for BS and ‘contributions’. what should be most concerning is why the hell does the cops want this SOOOOOOO badly? either someone in their midst is gaining a backhander and/or it can be altered to NOT show what happened when they shoot to death another poor fucker for no no reason!!
Re:
As I noted above while it’s possible that there might be some ‘financial incentives’ being kicked around to encourage departments to make use of the tech the biggest likely draw and one that doesn’t need any corruption of that sort is the ability for them to justify searches/arrests when they otherwise might not have been able to.
When all you care about is covering your ass should a fishing expedition get legal pushback having a ‘the tech said someone did something bad here’ excuse is a great thing to have.
Mostly Useful Report
Maybe ShotSpotter would be more useful if they added psychics into the product so it could predict when it will detect actual shots …
cops like shotspotter
Cops like shotspotter, not because it catches crooks, but because it gives them pretext to stop and search anyone they like or dont like.
Maybe the cops actually showed her some “data” that convinced her that changing her mind would be in her best interests, rather than actual data that would show the efficacy of the tech in question, and she’s just hiding it.
I don’t think that’s true, but it makes about as much sense as what apparently did happen.
It didn’t matter in this case, but I think that generally an absence should not count as a vote or should default to whichever option maintains the status quo.