FlyerIntel Research

The Best Time of Day to Fly From America's Largest Airports

Departure time is one of the few things a traveller controls, and it is worth more than most people think.

Published August 29, 2026 · Source period July 2023June 2026

Summary

Across 58 major US airports, flights scheduled to depart 5–9 AM arrived on time 88.4% of the time, against 69.2% for 6–9 PM departures — a gap of 19.2 percentage points that holds at almost every airport measured. The effect is not uniform: at DAL the spread between the best and worst departure window reaches 31.1 points, while at SLC it is only 9.6.

Key findings

  • Morning departures outperformed evening departures at 58 of 58 airports measured.
  • Network-wide, 5–9 AM departures arrived on time 88.4% of the time.
  • 6–9 PM departures were the weakest, at 69.2%.
  • DAL (Dallas) shows the widest swing: 91.2% for 5–9 AM departures against 60.1% for 9 PM–5 AM.
  • SLC (Salt Lake City) is the most consistent through the day, with only 9.6 points between its best and worst windows.

Why departure time matters this much

Delay is cumulative. An aircraft that starts the day late tends to stay late, and the disruption compounds through every subsequent leg it flies. A 6 AM departure is usually operated by an aircraft that overnighted at the airport; a 7 PM departure is operated by one that has already flown several legs and inherited whatever went wrong on each of them. That is why the late-aircraft category is consistently among the largest single causes of delay in the DOT data, and why the pattern below repeats at airport after airport regardless of geography or weather.

On-time arrival rate by departure window, all covered airports

5–9 AM88.4%
9 AM–12 PM82.9%
12–3 PM77.7%
3–6 PM72.1%
6–9 PM69.2%
9 PM–5 AM73.1%

Where timing matters most

AirportCityBest windowOn timeWorst windowOn timeSpread
DALDallas5–9 AM91.2%9 PM–5 AM60.1%31.1 pts
MDWChicago5–9 AM89.0%9 PM–5 AM58.6%30.4 pts
BWIBaltimore5–9 AM88.5%9 PM–5 AM59.0%29.5 pts
FLLFort Lauderdale5–9 AM88.7%9 PM–5 AM59.3%29.5 pts
MCOOrlando5–9 AM90.3%9 PM–5 AM61.8%28.5 pts
ABQAlbuquerque5–9 AM91.9%9 PM–5 AM65.3%26.6 pts
OAKOakland5–9 AM92.5%9 PM–5 AM66.2%26.3 pts
TPATampa5–9 AM90.7%9 PM–5 AM64.6%26.2 pts
BURBurbank5–9 AM90.9%9 PM–5 AM65.3%25.7 pts
BUFBuffalo5–9 AM87.6%9 PM–5 AM62.4%25.2 pts
RDURaleigh–Durham5–9 AM88.9%9 PM–5 AM63.9%25.1 pts
JAXJacksonville5–9 AM90.3%9 PM–5 AM65.5%24.8 pts
STLSt. Louis5–9 AM89.5%9 PM–5 AM64.6%24.8 pts
MIAMiami5–9 AM88.7%6–9 PM64.7%24 pts
HOUHouston5–9 AM92.0%9 PM–5 AM68.4%23.7 pts
PITPittsburgh5–9 AM90.3%9 PM–5 AM66.6%23.7 pts
RSWFort Myers5–9 AM92.9%6–9 PM69.4%23.5 pts
MCIKansas City5–9 AM90.2%9 PM–5 AM67.4%22.8 pts
AUSAustin5–9 AM89.9%9 PM–5 AM67.2%22.7 pts
INDIndianapolis5–9 AM89.7%9 PM–5 AM67.2%22.5 pts

Ranked by the gap between an airport’s best and worst departure window.

Every covered airport, morning against evening

AirportCity5–9 AM6–9 PMDifference
MCOOrlando90.3%62.0%28.3 pts
FLLFort Lauderdale88.7%61.6%27.1 pts
DALDallas91.2%64.2%27.0 pts
TPATampa90.7%65.8%24.9 pts
MIAMiami88.7%64.7%24.0 pts
RSWFort Myers92.9%69.4%23.5 pts
BWIBaltimore88.5%65.6%22.9 pts
LASLas Vegas90.0%68.1%21.9 pts
AUSAustin89.9%68.2%21.7 pts
MDWChicago89.0%67.3%21.7 pts
BOSBoston86.9%65.4%21.5 pts
CLTCharlotte87.1%65.9%21.2 pts
JAXJacksonville90.3%69.0%21.2 pts
DENDenver88.3%67.2%21.1 pts
MSYNew Orleans90.1%69.1%21.0 pts
SATSan Antonio91.3%70.4%21.0 pts
RDURaleigh–Durham88.9%68.0%20.9 pts
DFWDallas–Fort Worth84.8%64.2%20.6 pts
MCIKansas City90.2%69.6%20.6 pts
INDIndianapolis89.7%69.3%20.4 pts
PITPittsburgh90.3%70.0%20.3 pts
HOUHouston92.0%71.9%20.1 pts
BNANashville88.9%68.8%20.0 pts
CLECleveland88.7%68.8%19.9 pts
PHLPhiladelphia85.7%66.0%19.7 pts
STLSt. Louis89.5%70.0%19.4 pts
DCAWashington87.2%67.8%19.3 pts
OAKOakland92.5%73.2%19.3 pts
LGANew York88.8%69.5%19.3 pts
EWRNewark87.1%67.9%19.2 pts
SJUSan Juan88.3%69.3%19.0 pts
BUFBuffalo87.6%69.3%18.3 pts
CMHColumbus89.5%71.3%18.3 pts
ORDChicago84.4%66.2%18.2 pts
PHXPhoenix90.1%72.8%17.3 pts
ABQAlbuquerque91.9%74.6%17.3 pts
SMFSacramento92.1%74.9%17.2 pts
OMAOmaha89.6%72.5%17.1 pts
SJCSan Jose93.0%75.9%17.0 pts
ATLAtlanta88.1%71.1%17.0 pts
JFKNew York85.5%68.6%17.0 pts
IAHHouston88.0%71.0%16.9 pts
SANSan Diego87.0%70.2%16.8 pts
MEMMemphis87.9%71.2%16.7 pts
MKEMilwaukee87.2%70.7%16.5 pts
ONTOntario91.9%75.6%16.2 pts
IADWashington88.1%72.5%15.5 pts
DTWDetroit86.3%72.2%14.2 pts
SNASanta Ana91.1%77.2%13.9 pts
BURBurbank90.9%77.2%13.7 pts
SFOSan Francisco89.2%77.0%12.2 pts
PDXPortland89.2%77.1%12.1 pts
MSPMinneapolis88.2%76.4%11.8 pts
ANCAnchorage89.2%78.0%11.2 pts
LAXLos Angeles88.5%77.6%10.9 pts
SEASeattle86.6%76.5%10.1 pts
SLCSalt Lake City89.0%79.4%9.6 pts
HNLHonolulu93.1%86.4%6.7 pts

What to do with this

If a trip can be flown either in the morning or in the evening for a similar fare, the morning option is historically the more reliable choice at almost every major US airport. The effect is largest where an airport already runs close to capacity, and smallest at airports with slack in the schedule. It is a tendency measured across years of flights, not a guarantee about any individual departure.

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 bucketed by scheduled local departure time at the origin airport. A departure window is only included for an airport where it recorded at least 120 flights in the period.

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.

Sources

  • FlyerIntel analysisFlyerIntel · source
  • Reporting Carrier On-Time PerformanceUS Department of Transportation, Bureau of Transportation Statistics · source

Full FlyerIntel methodology · Report a problem with this analysis

More FlyerIntel Research