New York City's public crash database — the raw feed that powers independent safety maps and research citywide — has not been updated since June 11, leaving roughly three months of collisions invisible to the public. The freeze means tools that New Yorkers rely on to track dangerous intersections and corridors are working with dramatically incomplete numbers, even as city officials have offered little explanation for the gap.
As reported by Streetsblog New York City, the outage remained unfixed as of Tuesday. The database, maintained by the Office of Technology and Innovation, recorded 36,195 crashes through June 11 — an average of 223.4 reported crashes per day — but that figure has not budged since, even though the outlet's reporting shows the city separately logged 54,832 reported crashes through August 30. That works out to a reported rate of just 149.6 crashes per day for the more recent period, a figure Streetsblog's Gersh Kuntzman notes is roughly one-third fewer than the actual rate implied once the fuller total is factored in. According to NYC Open Data, the Motor Vehicle Collisions tables contain information from all police-reported motor vehicle collisions in New York City, with the data provided by the NYPD.
The gap is not just a technical inconvenience. Because the underlying data feed is broken, sites like Crashmapper and CrashCount — which map crash locations to help residents and researchers spot dangerous patterns — are missing thousands of incidents. CrashCount, created by Michael Freedman, is unable to provide independent analysis of police crash data while the pipeline stays down, according to Streetsblog's reporting.
A Small Brooklyn Street Shows the Stakes
The consequences show up in specific places. Per Vision Zero View data cited by Streetsblog, Court Street in Brooklyn had six injuries through July 31 — but Crashmapper showed only four, because two of those injuries occurred after June 11 and never made it into the public feed. Streetsblog notes that Court Street became safer after a protected bike lane was completed there, with just those six injuries recorded in the same period compared with 15 injuries over an equivalent stretch before the bike lane went in.
That's exactly the kind of before-and-after evidence that stale data can obscure. As Streetsblog's reporting puts it, stale crash data may give the public a false sense that roadways are safer than they actually are — a particularly sensitive point given that NYC DOT recorded 205 traffic fatalities in 2025, the city's lowest annual road death toll since 1910 and a 31% drop since Vision Zero launched in 2014. Traffic fatalities continued to decline, reaching a record low in 2025.
Officials Point Fingers Over Who's Responsible
Responsibility for the outage has been difficult to pin down. An NYPD spokesperson declined to explain why the open data portal lacked updates and instead referred questions to the Office of Technology and Innovation, per Streetsblog's account. A note posted to the NYC Open Data crash page reportedly said the automated update process was being fixed, with a projected completion of August 2026 — a deadline that has already passed without the data reappearing.
A DOT spokesperson said residents can still access the NYPD traffic crash website during the gap, but that site provides only a rolling, unarchived data feed and does not allow users to plot specific streets or areas over time, according to the same Streetsblog reporting. DOT's Vin Barone said the city is working with OTI to improve data presentation, while Gloria Chin said in a statement sent Aug. 28 that work to improve access to the data was complete and undergoing final review.
An anonymous activist quoted by Streetsblog was blunter, saying the NYPD has zero interest in making its data useful to the general public unless it supports its political narrative. Noel Hidalgo, executive director of BetaNYC, said crash data should be automated properly and accurately. BetaNYC has separately built open-source civic tools, including precinct-level dispatch maps and legislative trackers, that similarly depend on continuous municipal data flows to function.
Who Feels the Gap Most
The stakes of an incomplete map fall unevenly across the city. A June report by the Center for an Urban Future found lower-income neighborhoods like East Flatbush record serious crash rates of 67.0 per 10,000 residents, well above the citywide average of 47.0, Hoodline has previously reported. Eight of the ten highest crash-risk neighborhoods in the city are lower-income communities of color, meaning the areas that most need precise, up-to-date street-level data are also the ones most affected when it disappears.
The blackout also lands amid other enforcement concerns tied to crash data. Hoodline has reported that hit-and-run drivers killed nine people citywide in August alone, part of a longer trend in which police arrest rates for hit-and-run collisions involving injuries dropped to 3.6% by 2023, down from 8.6% in 2018 — leaving more than 96% of hit-and-run drivers unpunished annually even as the city logs roughly 46,000 such collisions each year.
New York's underlying reporting system dates back decades: NYPD collision tracking was standardized in 1998 through TrafficStat and formalized the following year with the Traffic Accident Management System, before electronic crash reporting was fully adopted departmentwide in April 2016. Under the city's Open Data Law, expanded in 2017, agencies are required to keep public datasets current on the city's central portal — but as Streetsblog's reporting makes clear, NYPD remains responsible for actually generating and transmitting the crash feed that other agencies and the public depend on.
