As Chicago's streetlights turned to LED, what happened to crime after dark?
A month-by-month replay, 2014–2025. The top panel is the share of the city converted, dated from satellite imagery. The panels below it track outdoor crime at night (9 pm–4:59 am) against outdoor crime in daytime (9 am–4:59 pm), the comparison that isolates what lighting can plausibly touch.
By neighborhood: each area measured against its own conversion
Citywide curves mix the lighting rollout with everything else that moved in those years. The sharper test is within the city: South and West Side areas converted first, in 2018; the North Side a year later. If lighting matters, night crime should bend down in each area around its own conversion month, while areas not yet converted do not. The map replays the rollout; the chart lines up every area on its conversion start.
Neighborhood layer pending: this section fills in once the community-area crime pull (05_pull_crime_by_area.py) has run and its output is published alongside this page.
0%100% converted−40%+40% night crime (12-mo, vs area's pre-conversion years)
South & West Side areas (converted first)North & Central areas (converted later)All 77 areas
How to read this
Conversion dates come from space, not city records. NASA's Black Marble nightly radiance (VIIRS, ~460 m) falls 20–30% when sodium lamps are replaced by LEDs, because the sensor barely registers LED blue light and LED fixtures aim light downward. Each tract's start and end of conversion is the best-fitting flat → decline → flat break in its monthly radiance relative to an unconverted suburban ring. The share converted shown here is the mean across Chicago's 801 tracts.
Night versus day on the same streets. Lighting can only change what happens after dark. Comparing outdoor crime at 9 pm–5 am with outdoor crime at 9 am–5 pm removes citywide shifts that hit both (policing, the pandemic, reporting). The index uses rolling 12-month totals so seasons do not dominate.
What the citywide replay shows. The night-to-day ratio falls through 2017–2019 as the South and West Sides convert, and rises again from 2021, after the whole city is converted, when night activity patterns changed for many reasons. The citywide curve is a description, not an estimate of the lighting effect; the neighborhood section is the design that can separate them.
Outdoor only. Incidents at street, sidewalk, alley, parking, yard, park, transit stop and similar locations, from the City of Chicago's public crime data. Counts are reported incidents, so changes in reporting affect night and day alike.
Data: NASA VNP46A2 via Google Earth Engine; City of Chicago "Crimes – 2001 to Present." Analysis and visualization: Matthew DeMichele (Ponte Research and Policy) and Joshua Freilich (John Jay College of Criminal Justice), 2026.