New York has the lowest per-capita rate of alcohol-involved traffic deaths in the nation, yet the gap between its deadliest and safest counties is more than twenty-fold. An analysis of federal crash records from 2019 to 2023.
New York is, on paper, one of the safest states in the country for drunk driving. Its per-capita rate of alcohol-involved traffic deaths is the lowest in the nation, roughly 40 percent of the national average. Yet that statewide figure hides enormous local variation. This analysis maps every alcohol-involved fatal crash in New York from 2019 through 2023 using federal crash records, and the gap between the deadliest and safest counties is more than twenty-fold.
The deadliest counties for drunk driving
Alcohol-involved fatal crashes per 100,000 residents, by county, 2019 to 2023 (five-year total). Hover or tap a county, and the full ranking is below.
FARS counts a crash as alcohol-involved when at least one driver was recorded as drinking. Figures aggregate five years because single-year county counts are very small.
| County | Alc. fatal crashes, 2019–23 | per 100k (5 yr) |
|---|---|---|
| Lewis | 10 | 37.6 |
| Sullivan | 20 | 25.3 |
| Oswego | 27 | 22.9 |
| Cortland | 10 | 21.6 |
| Yates † | 5 | 20.3 |
| Hamilton † | 1 | 19.6 |
| Madison | 12 | 17.8 |
| Orleans † | 7 | 17.6 |
| Chenango † | 8 | 17.1 |
| Schuyler † | 3 | 16.9 |
| Genesee † | 9 | 15.5 |
| Jefferson | 18 | 15.5 |
| Wayne | 14 | 15.4 |
| Franklin † | 7 | 14.9 |
| Warren † | 9 | 13.7 |
| Washington † | 8 | 13.1 |
| Columbia † | 8 | 13.1 |
| St. Lawrence | 14 | 13.0 |
| Wyoming † | 5 | 12.5 |
| Cattaraugus † | 9 | 11.8 |
| Herkimer † | 7 | 11.7 |
| Chemung † | 9 | 10.9 |
| Tioga † | 5 | 10.4 |
| Suffolk | 156 | 10.2 |
| Montgomery † | 5 | 10.1 |
| Oneida | 23 | 10.0 |
| Otsego † | 6 | 10.0 |
| Ontario | 11 | 9.8 |
| Albany | 31 | 9.8 |
| Orange | 37 | 9.2 |
| Putnam † | 9 | 9.2 |
| Seneca † | 3 | 9.1 |
| Broome | 18 | 9.1 |
| Delaware † | 4 | 9.0 |
| Clinton † | 7 | 8.9 |
| Chautauqua | 11 | 8.7 |
| Rensselaer | 14 | 8.7 |
| Steuben † | 8 | 8.6 |
| Schenectady | 13 | 8.2 |
| Livingston † | 5 | 8.1 |
| Essex † | 3 | 8.1 |
| Cayuga † | 6 | 8.0 |
| Dutchess | 23 | 7.7 |
| Erie | 70 | 7.4 |
| Ulster | 13 | 7.1 |
| Tompkins † | 7 | 6.8 |
| Monroe | 49 | 6.5 |
| Allegany † | 3 | 6.4 |
| Onondaga | 29 | 6.2 |
| Fulton † | 3 | 5.7 |
| Niagara | 12 | 5.7 |
| Nassau | 66 | 4.8 |
| Rockland | 15 | 4.4 |
| Greene † | 2 | 4.2 |
| Saratoga † | 8 | 3.4 |
| Schoharie † | 1 | 3.3 |
| Westchester | 30 | 3.0 |
| Bronx | 40 | 2.8 |
| Queens | 66 | 2.8 |
| Richmond † | 9 | 1.8 |
| Kings | 41 | 1.6 |
| New York (Manhattan) | 24 | 1.5 |
† Small sample, fewer than 10 alcohol-involved fatal crashes over five years. Rates for these counties are statistically noisy and should be read with caution.
A safe state with glaring local exceptions
Rural Upstate counties dominate the top of the ranking. Lewis County leads the state at 37.6 alcohol-involved fatal crashes per 100,000 residents over five years, followed by Sullivan, Oswego and Cortland. At the other end sit the New York City boroughs, where Manhattan records just 1.5, roughly one twenty-fifth of Lewis County's rate.
Analysis. The ranking is the mirror image of the raw counts. Suffolk and Nassau see the most alcohol-involved fatal crashes by sheer number, but their large populations pull their per-capita rates down. Dense, transit-rich downstate counties give people alternatives to driving and shorter trips, while rural counties pair long drives on fast two-lane roads with few transportation options. The per-capita danger is concentrated exactly where the population is thinnest.
When drunk-driving crashes happen
Share of alcohol-involved fatal crashes by day of week, 2019 to 2023.
| Sun | 20.7% |
|---|---|
| Mon | 9.8% |
| Tue | 10.1% |
| Wed | 10.2% |
| Thu | 9% |
| Fri | 15.2% |
| Sat | 25% |
Weekends carry the load. Friday through Sunday account for 61 percent of alcohol-involved fatal crashes, and Saturday alone for a quarter. The single deadliest hour is midnight, and the window from 9 PM to 3 AM accounts for 46 percent of these crashes.
How impaired the drivers were
Among New York drivers in fatal crashes with a measured blood-alcohol level above zero.
| At or over 0.08 (legal limit) | 80.1% |
|---|---|
| At or over 0.15 (nearly double) | 57.1% |
Four in five drinking drivers in fatal crashes were at or above the 0.08 legal limit, and well over half were at or above 0.15.
Alcohol-involved fatal crashes by year
New York, 2019 to 2023.
| 2019 | 177 | 16% |
|---|---|---|
| 2020 | 193 | 17.5% |
| 2021 | 263 | 23.8% |
| 2022 | 253 | 22.9% |
| 2023 | 220 | 19.9% |
Alcohol-involved fatal crashes climbed during the pandemic, peaking at 263 in 2021, before easing to 220 in 2023.
How New York compares nationally
Drunk driving remains a national crisis even as New York leads on prevention. Each figure below links to its source.
Two definitions are at work. The county ranking uses FARS alcohol-involved crashes, meaning at least one driver was recorded as drinking, with no statistical imputation. NHTSA's headline alcohol-impaired figures estimate drivers at or above 0.08 and include imputation for untested drivers, so they run somewhat higher. The two are labeled separately throughout.
If a drunk driver hurt you in New York
A drunk-driving crash can support claims well beyond a standard car-accident case, both against the impaired driver and, in some circumstances, against the bar or restaurant that over-served them under New York's Dram Shop Act. Injured people may be entitled to compensation for medical bills, lost wages, and pain and suffering, and in cases of especially reckless conduct, punitive damages.
If you were hurt by a drunk or negligent driver in New York, the team at Porter Law Group can review your crash and your options at no cost. There is no obligation.
Methodology
Crashes are counted as alcohol-involved when at least one driver was recorded as drinking in the federal Fatality Analysis Reporting System, which is the only alcohol measure available at the county level and uses no statistical imputation. For 2021 through 2023, where the derived drinking-driver field was discontinued, it was reconstructed from the vehicle-level records and validated against 2019 and 2020, where it reproduced the official counts exactly. Figures aggregate five years, from 2019 through 2023, because single-year county counts are very small. Thirty-two of the sixty-two counties recorded fewer than ten alcohol-involved fatal crashes over the period and are marked as small samples. Population is the Census American Community Survey five-year 2023 estimate. FARS captures fatal crashes only, not all alcohol-related crashes.
