What this dataset measures
Built to test the common claim that the Wednesday before Thanksgiving is the busiest travel day of the year. TSA screenings are the only daily national count of air travelers; FHWA's continuous count stations are the only daily national count of road traffic; the Bureau of Transportation Statistics (BTS) survey is the only federal count of long-distance trips by day of the Thanksgiving period.
- Geography
- United States
- Unit of observation
- Calendar day (and, for roads, calendar day by road group)
- Update cadence
- One-time compilation; TSA publishes new days daily and FHWA publishes monthly count files with a lag of several months
- Curator
- FactHub Research Team
Supporting files
Source snapshots, manifests, quality reports, documentation,
and other materials used to produce or review this release.
-
tsa_parse.py
Transformation manifest · text/x-python · 635 bytes
Parses saved TSA yearly passenger-volume pages into a date-to-count file.
-
tmas_agg.py
Transformation manifest · text/x-python · 1 KB (1,052 bytes)
Reads FHWA monthly continuous-count zip files and sums each lane's 24 hourly volumes into a daily total, skipping records with restriction flags or non-numeric hours.
-
build.py
Transformation manifest · text/x-python · 4.6 KB (4,665 bytes)
Builds the TSA and road rank tables, the road index and the strict-panel check from the parsed files.
Saved analyses
Long-distance trips by day, Thanksgiving 2001 (survey)
Question: On which day of the Thanksgiving period did Americans make the most trips of 50 miles or more?
Result: In the 2001 survey, Thanksgiving Day had the most long-distance trips (about 28.0 million, 109% above an average day); Wednesday had about 19.5 million.
Operation: Copy of BTS table, Thanksgiving rows
Scope: November 20-25, 2001
View this analysis
- Counts departing plus returning trips each day, so a same-day round trip to dinner counts twice on Thursday.
- 2001 data, weeks after the September 11 attacks.
Sources
-
Transportation Security Administration, TSA checkpoint travel numbers (passenger volumes), yearly tables 2019-2025 and current year
Daily number of people screened at TSA airport checkpoints
Retrieved November 16, 2026. Public domain (U.S. government work). Allowed
Reliability:
Government.
TSA posts the total screened at all U.S. checkpoints each day; past years are at /travel/passenger-volumes/{year}. TSA sometimes revises a day slightly after its first press release (for example, 3,133,924 announced for November 30, 2025 versus 3,134,613 in the table).
-
Federal Highway Administration, Traffic Monitoring Analysis System (TMAS) continuous count station hourly volume data, monthly files 2023-2024
Hourly vehicle counts by lane from permanent road counters reported by state transportation departments
Retrieved November 16, 2026. Public domain (U.S. government work). Allowed; the raw monthly zip files are linked, not rehosted
Reliability:
Government.
State transportation departments run the counters and submit hourly volumes to FHWA, which publishes them by month. Each record names the state, road class (functional system), station, direction and lane.
-
Bureau of Transportation Statistics, America on the Go: U.S. Holiday Travel, Figures 1-4 table (Long-Distance Trips During Thanksgiving and Christmas/New Year)
Survey estimate of long-distance trips (50+ miles) by day of the 2001 Thanksgiving period, by all modes and by car
Retrieved November 16, 2026. Public domain (U.S. government work). Allowed
Reliability:
Government.
Built from the 2001 National Household Travel Survey, in which sampled households logged their trips. Trip counts are survey estimates and include departing and returning trips on each day.
License: FactHub's selection, cleaning and combined tables are
licensed under CC BY 4.0.
Source data keeps the terms listed above (terms).
Limitations and reproducibility
Reproducible from public files with the included scripts: tsa_parse.py turns TSA's yearly passenger-volume web tables into tsa_daily.csv. tmas_agg.py reads FHWA's monthly continuous-count-station zip files (hourly volumes per lane) into daily totals per lane, and build.py builds the road index, the rank tables and the strict-panel check. The FHWA zip files (about 26 MB each) are linked from the source, not hosted here.
- TSA counts people screened at checkpoints, not airline passengers or trips. Connecting passengers who stay airside are not counted again, and airport workers screened at checkpoints may be included.
- Road counters count vehicles passing a point, not people or trips. One long trip crosses many counters; a short trip may cross none. Counters sit mostly on major roads, so local streets are underrepresented.
- The road index weights every counted lane equally (each lane's daily volume divided by its own yearly average, then averaged across lanes). Busy lanes and quiet lanes count the same.
- Road data cover 2023 and 2024 only, from counters that reported on at least 355 days of the year. States differ in how many counters they send to FHWA, and some states send few or none for a full year.
- The BTS survey dates from 2001, weeks after the September 11 attacks, and counts trips of 50 miles or more each way, including same-day round trips.
Publications using this dataset
Version details
The exact citing versions, and the version of this dataset each cites.
-
Article
thanksgiving-wednesday-busiest-travel-day —
version 1
(published) cites version 1
-
Chart
thanksgiving-wednesday-air-vs-road —
version 1
(published) cites version 1
-
Chart
thanksgiving-busiest-air-day-strip —
version 1
(published) cites version 1
-
Briefing
thanksgiving-facts —
version 3
(published) cites version 1
-
Briefing
thanksgiving-facts —
version 2
(published) cites version 1
Node: dataset:thanksgiving-travel-daily-counts:1:analysis:bts-2001-long-distance-trips