FlyerIntel Research
The Delay You Inherit
Most delay is not created by your flight. It arrives with the aircraft, and it accumulates all day.
Published August 30, 2026 · Source period July 2023 – June 2026
Summary
Departures scheduled at 05:00 leave on time 93.0% of the time. By 20:00 that has fallen to 66.9% — a drop of 26.1 percentage points across a single day. The reason is visible in how the delay is attributed: at 05:00 only 15.7% of delay minutes come from a late-arriving aircraft, but by 20:00 that cause accounts for 49.0% of them. An early flight is not flying in better weather or through emptier skies. It is flying on an aircraft that has not yet had time to fall behind.
Key findings
- On-time departures fall from 93.0% at 05:00 to 66.9% at 20:00.
- Late-arriving aircraft cause 15.7% of delay minutes at 05:00 and 49.0% by 20:00 — it becomes the single largest cause of delay in America by mid-afternoon.
- DAL (Dallas) decays most: 91.7% on time before 09:00 against 58.0% in the evening peak, a spread of 33.8 points.
- HNL (Honolulu) holds up best, losing only 6.9 points across the same comparison.
- Taxi-out time barely moves across the day, so congestion on the ground is not what is driving this.
Why the first flight of the day is a different product
An aircraft operating its first flight of the day starts from a stand where it has sat all night. Nothing upstream can have gone wrong yet. Every departure after that inherits whatever happened to the aircraft, the crew and the gate before it, and airlines schedule aircraft to fly several legs a day with little slack between them. A twenty-minute delay in the morning does not disappear; it is handed to the next flight, and often the one after that. This is why departure time is worth more than almost any other choice a traveller makes, and why it costs nothing.
On-time departure rate by scheduled hour
Share of delay minutes caused by a late-arriving aircraft
What this is not
These two curves move together, and it would be easy to read one as causing the other. The honest statement is narrower: as the day goes on, a larger share of the delay that does occur is attributed by the operating carrier to a late inbound aircraft, and on-time performance falls. Evening flights also face accumulated air traffic control flow, afternoon convective weather in summer, and crew duty limits. What the attribution data shows is that knock-on delay is the largest single named cause by the middle of the day, and it is the one a traveller can avoid entirely by flying earlier.
Where the day decays fastest
| Airport | City | Before 09:00 | Evening peak | Points lost |
|---|---|---|---|---|
| DAL | Dallas | 91.7% | 58.0% | 33.8 |
| MCO | Orlando | 91.2% | 61.0% | 30.2 |
| FLL | Fort Lauderdale | 89.5% | 59.8% | 29.7 |
| BWI | Baltimore | 89.2% | 60.6% | 28.7 |
| MDW | Chicago | 90.4% | 62.7% | 27.7 |
| TPA | Tampa | 92.0% | 65.1% | 26.9 |
| LAS | Las Vegas | 90.7% | 64.2% | 26.5 |
| AUS | Austin | 91.4% | 65.0% | 26.4 |
| HOU | Houston | 92.0% | 65.8% | 26.2 |
| BNA | Nashville | 90.7% | 64.8% | 25.8 |
| RSW | Fort Myers | 94.2% | 68.6% | 25.6 |
| BOS | Boston | 89.6% | 64.9% | 24.7 |
| MSY | New Orleans | 91.6% | 67.2% | 24.4 |
| MIA | Miami | 90.0% | 65.6% | 24.3 |
| STL | St. Louis | 90.8% | 66.5% | 24.3 |
| OAK | Oakland | 92.8% | 68.5% | 24.3 |
| DEN | Denver | 90.0% | 66.4% | 23.6 |
| MCI | Kansas City | 92.7% | 69.2% | 23.5 |
| SJU | San Juan | 90.4% | 67.0% | 23.4 |
| JAX | Jacksonville | 92.1% | 68.8% | 23.3 |
| SAT | San Antonio | 92.6% | 69.6% | 23.0 |
| RDU | Raleigh–Durham | 91.1% | 68.4% | 22.7 |
| CLT | Charlotte | 88.9% | 66.7% | 22.2 |
| PIT | Pittsburgh | 93.1% | 70.8% | 22.2 |
| BUF | Buffalo | 90.4% | 68.4% | 21.9 |
Morning is departures scheduled 05:00–08:59; evening peak is 17:00–20:59. Ranked by the gap between them.
What to do with this
If the itinerary allows it, take the earliest departure you can tolerate — the advantage is largest at the airports at the top of the table above. If you are connecting, the argument is stronger still, because an inherited delay on the first leg becomes a missed connection rather than a late arrival. And if you must fly in the evening, prefer an airport near the bottom of that table, where the network holds its schedule better late in the day.
Methodology
Calculated by FlyerIntel from 21,076,354 individual flight records in the US Department of Transportation Bureau of Transportation Statistics Reporting Carrier On-Time Performance dataset, covering July 2023 through June 2026. Following BTS methodology, a flight counts as delayed when it arrives 15 or more minutes after its scheduled arrival time. Cancelled and diverted flights have no arrival time and are excluded from on-time rates rather than counted as late; cancellation rates are reported separately. Full definitions are published at /methodology/. Flights are grouped by scheduled departure hour in the airport's local time. Hourly rates are computed from summed counts across covered airports rather than by averaging rates, so busy airports carry proportionate weight. Delay-cause shares are of minutes the operating carrier attributed under the DOT's five cause categories, which apply only to delays of fifteen minutes or more. An airport appears in the decay table only where it recorded at least 5,000 flights.
Limitations
The source dataset covers scheduled domestic flights operated by carriers above the DOT reporting threshold; smaller carriers and international flights are not included. Flights sold under a mainline brand but flown by a regional carrier are attributed to the operating carrier, which is how the DOT records them. Past performance describes what has happened and is not a prediction about any individual flight. Delay causes are self-reported by the operating carrier, and a single delayed flight can carry more than one cause. The relationship shown here is an association across millions of flights, not a prediction about any individual departure: an early flight can still be delayed, and an evening flight can be on time.
Sources
- FlyerIntel analysis — FlyerIntel · source
- Reporting Carrier On-Time Performance — US Department of Transportation, Bureau of Transportation Statistics · source
Full FlyerIntel methodology · Report a problem with this analysis
More FlyerIntel Research
The Same Airport, Different Airlines
Airport reliability tables treat a field as one thing. Ask which carrier you are flying and the picture changes completely.
America's Most and Least Reliable Flight Numbers
Two flights can leave the same airport for the same city on the same day and arrive on time at wildly different rates. The number on the ticket is not decoration.
How Long US Customs Actually Takes
Two years of federal records for every major US arrival hall — and the queue you join depends on the passport you carry.
The Best Major Airports for a Tight Connection
Where an inbound flight is most likely to land on time, and how long you then spend taxiing before the clock really starts.