Saved analyses
Average alcohol-impaired deaths on holiday nights, 2015-2024
Question: Which holiday nights have the most drunk-driving deaths, on average, and where does Thanksgiving Eve rank?
Result: New Year's Eve averages 51.9 alcohol-impaired deaths, the most of any holiday night. Thanksgiving Eve averages 31.9 and ranks 78th of about 365 nights in a median year; its best rank in any year was 55th.
Operation: Mean of alcohol_impaired_deaths across years for each named night; median, best and worst rank_in_year.
Scope: United States, nights 6 p.m.-5:59 a.m., 2015-2024 (New Year's Eve 2015-2023)
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Robustness checks
-
Sensitivity analysis:
Does the time window change the answer?.
With a 9 p.m. to 2:59 a.m. window or calendar days, Thanksgiving Eve still averages fewer deaths than a typical Saturday and never ranks in the top 25 nights of a year.
Typical nights by day of week, 2015-2024
Question: How many alcohol-impaired deaths does an ordinary night of each kind have?
Result: A typical Saturday night has 41.2 alcohol-impaired deaths, a typical Wednesday night 16.6, and the average night 23.6.
Operation: Mean of alcohol_impaired_deaths over all nights matching the filter.
Scope: United States, 3,652 nights, Jan. 1, 2015 to Dec. 30, 2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Holiday nights ranked by drunk-driving deaths (chart rows)
Question: How does Thanksgiving Eve compare with other holiday nights?
Result: New Year's Eve 51.9, Memorial Day weekend Saturday 48.6, Labor Day weekend Saturday 44.4, Fourth of July 44.2, Thanksgiving Eve 31.9 average alcohol-impaired deaths per night.
Operation: Subset of named-night-averages: ten named nights, shortened labels.
Scope: United States, nights 6 p.m.-5:59 a.m., 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Thanksgiving Eve, year by year
Question: Each year, how many drunk-driving deaths happened on Thanksgiving Eve, and how did that night rank?
Result: Thanksgiving Eve was the deadliest Wednesday night of the year in 8 of 10 years, but 30 to 52 Saturday nights had more alcohol-impaired deaths each year. Its best rank was 55th (2015).
Operation: Lookup of each Thanksgiving Eve in the nights table; rank among the year's nights and Wednesdays; count of that year's Saturday nights with more alcohol-impaired deaths.
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Holiday nights deadlier than Thanksgiving Eve, count of years
Question: In how many years did each holiday night have more drunk-driving deaths than that year's Thanksgiving Eve?
Result: New Year's Eve had more alcohol-impaired deaths than Thanksgiving Eve in 9 of 9 years; the Saturdays of Memorial Day and Labor Day weekends in 9 of 10.
Operation: For each named night and year, compare alcohol_impaired_deaths with Thanksgiving Eve of the same year; count years where it was higher.
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Holiday periods: deaths per 24 hours
Question: Counting the whole holiday period, is Thanksgiving deadlier per day than other long weekends or an ordinary day?
Result: Per 24 hours, the Thanksgiving period averages 37.6 alcohol-impaired deaths, against 43.8 for Memorial Day weekend, 44.1 for Labor Day weekend and 31.4 on a typical day.
Operation: Mean over 2015-2024 of deaths in each 108-hour period, divided by 4.5 to give deaths per 24 hours; typical 24 hours = all deaths with a known crash hour divided by 3,652 days.
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
- Memorial Day and Labor Day windows are set to the same 108-hour length as NHTSA's Thanksgiving window for a like-for-like comparison; NHTSA defines those periods by its own rules each year.
Thanksgiving holiday period (Wed. 6 p.m. to Mon. 6 a.m.), by year
Question: How many people died in alcohol-impaired crashes over each Thanksgiving holiday period?
Result: Between 124 (2019) and 200 (2021) alcohol-impaired deaths per Thanksgiving period; 169 on average.
Operation: Sum by hour over the 108-hour window each year.
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Does the time window change the answer?
Question: Does Thanksgiving Eve rank higher if night is defined differently?
Result: In every window, Thanksgiving Eve averages fewer alcohol-impaired deaths than a typical Saturday. Its best single-year rank is 55th (6 p.m.-6 a.m.), 28th (9 p.m.-3 a.m.) or 97th (calendar day).
Operation: Repeat the ranking with a 9 p.m.-2:59 a.m. window and with calendar days (midnight to 11:59 p.m. on the Wednesday).
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.
Robustness check for:
Average alcohol-impaired deaths on holiday nights, 2015-2024
Annual alcohol-impaired and total traffic deaths
Question: Do our totals match NHTSA's published national counts?
Result: Alcohol-impaired deaths ranged from 10,196 (2019) to 13,599 (2021).
Operation: Sum over all crashes each year, including crashes with unknown hour.
Scope: United States, 2015-2024
View this analysis
- Counts only deaths in crashes on public roads that FARS records; injuries and non-fatal crashes are not included.
- Alcohol-impaired deaths are estimates: when a driver's blood-alcohol test result is unknown, NHTSA fills it in with 10 imputed values.
- 2024 data are from NHTSA's first (annual report) release of the 2024 file, posted April 1, 2026; the final file, due about a year later, can change counts slightly.