← The impacts · AItopiaOrAImageddon?

entry · p2026-09-06-043 · validated

ETAS predictive policing (PredPol)

led to a average 7.4% reduction in crime volume as a function of patrol time
tech
predpol-etas
passes the definition as
self-exciting point-process model fit to a division's own recent crime records to forecast next-shift hotspots
from
2011
region
US
domain
safety
side
benefit
people
3 LAPD police divisions under algorithmic patrol direction (2015)
effect
7.4 percent (2015) — reduction in crime volume as a function of patrol time — 4.3 fewer crimes per division per week, a projected $17,258,801 annual saving across LAPD
reversible
yes
who holds the switch
owner:LAPD
attribution
named
source kind
primary
the statistic
In a randomised controlled trial across three LAPD divisions, ETAS-directed patrols produced an average 7.4% reduction in crime volume (4.3 fewer crimes per division per week) while analyst-directed patrols showed no significant effect; ETAS predicted 1.4–2.2 times as much crime as the analysts.
source
https://escholarship.org/content/qt1br4975j/qt1br4975j.pdf
proposed by
researcher (opus), brief 2026-09-06, era-1990-2011
received
2026-09-06T09:39:39-04:00
harvest report
impacts/harvest/2026-09-06-era-1990-2011.md

Rulings, in order

  1. 2026-09-06T09:39:47-04:00 · validatorpass — quote found on the fetched page (exact match; HTTP 200 application/pdf)
The proposal as it arrived (JSON)
{
 "id": "p2026-09-06-043",
 "received": "2026-09-06T09:39:39-04:00",
 "proposer": "researcher (opus), brief 2026-09-06, era-1990-2011",
 "kind": "entry",
 "sha": "bc6a9ddef06a49e2",
 "data": {
  "kind": "entry",
  "tech": "predpol-etas",
  "tech_name": "ETAS predictive policing (PredPol)",
  "definition": "self-exciting point-process model fit to a division's own recent crime records to forecast next-shift hotspots",
  "from": "2011",
  "region": "US",
  "domain": "safety",
  "side": "benefit",
  "people": {
   "n": 3,
   "unit": "LAPD police divisions under algorithmic patrol direction",
   "date": "2015"
  },
  "effect": {
   "n": 7.4,
   "unit": "percent",
   "date": "2015",
   "what": "reduction in crime volume as a function of patrol time — 4.3 fewer crimes per division per week, a projected $17,258,801 annual saving across LAPD"
  },
  "reversible": true,
  "switch": "owner:LAPD",
  "statistic": "In a randomised controlled trial across three LAPD divisions, ETAS-directed patrols produced an average 7.4% reduction in crime volume (4.3 fewer crimes per division per week) while analyst-directed patrols showed no significant effect; ETAS predicted 1.4–2.2 times as much crime as the analysts.",
  "source": "https://escholarship.org/content/qt1br4975j/qt1br4975j.pdf",
  "quote": "led to a average 7.4% reduction in crime volume as a function of patrol time",
  "attribution": "named",
  "source_kind": "primary"
 },
 "from_file": "impacts/harvest/2026-09-06-era-1990-2011.md"
}