FlyerIntel Research
Flights Into an Airport Versus Out of It
SFO is the airport where arriving flights run furthest behind departing ones, by 6.4 points.
Published October 18, 2026 · Source period July 2023 – July 2026
Summary
6.4 points separate flights landing at SFO from flights leaving it: 72.3% of arrivals were on time, 78.7% of departures landed on time. That is the largest gap with arrivals behind among 64 airports; flights into the airport were on time less often than flights out of it at 39 of them.
Key findings
- SFO has the widest gap with arrivals behind: 72.3% of flights arriving on time, 78.7% of flights leaving landing on time, 6.4 points.
- CLT has the widest gap the other way: 74.8% of flights leaving landed on time, 79.7% of flights arriving, 4.8 points.
- Flights into the airport were on time less often than flights out of it at 39 of 64 airports, and more often at 25.
- The median airport’s gap, flights leaving less flights arriving, is +1.3 points.
- The gap is wider than 2 standard errors with arrivals behind at 38 airports and with departures behind at 23.
Two ways to ask whether an airport runs on time
Every flight is counted at two airports: it leaves one and lands at another. This report puts the two counts side by side for 64 airports. Across them, 78.6% of completed flights landing at these airports were on time and 78.1% of those leaving them landed on time elsewhere. At SFO, the arrival side is 6.4 points behind. At CLT, the departure side is 4.8 points behind. Both rates ask the same question of a flight, did it land within 15 minutes of schedule, so the difference between them comes from where the flights go and come from, and from what happens at the airport itself.
Airports where arriving flights run furthest behind
| # | Airport | City | On time, flights arriving | On time, flights leaving | Leaving less arriving, points | Cancelled, arriving | Cancelled, leaving | Completed arrivals |
|---|---|---|---|---|---|---|---|---|
| 1 | SFO | San Francisco | 72.3% | 78.7% | +6.4 | 1.0% | 1.0% | 430,716 |
| 2 | ANC | Anchorage | 78.1% | 83.6% | +5.5 | 1.3% | 1.2% | 57,911 |
| 3 | SAT | San Antonio | 77.0% | 82.3% | +5.2 | 1.0% | 1.1% | 125,688 |
| 4 | MEM | Memphis | 76.4% | 81.2% | +4.8 | 1.8% | 1.7% | 71,381 |
| 5 | ONT | Ontario | 78.8% | 83.5% | +4.7 | 0.8% | 0.9% | 79,193 |
| 6 | BUF | Buffalo | 74.8% | 79.4% | +4.6 | 2.3% | 2.3% | 63,971 |
| 7 | ABE | Allentown | 78.3% | 82.8% | +4.5 | 1.6% | 1.5% | 11,553 |
| 8 | JAX | Jacksonville | 75.6% | 80.0% | +4.5 | 1.6% | 1.7% | 88,683 |
| 9 | TUS | Tucson | 79.8% | 84.3% | +4.4 | 0.7% | 0.8% | 58,821 |
| 10 | OMA | Omaha | 77.3% | 81.7% | +4.4 | 1.1% | 1.2% | 74,010 |
| 11 | MKE | Milwaukee | 76.6% | 80.7% | +4.2 | 1.7% | 1.7% | 86,232 |
| 12 | ABQ | Albuquerque | 79.0% | 83.2% | +4.1 | 0.8% | 0.8% | 75,360 |
Each rate is flights that landed within 15 minutes of schedule over flights that completed, taken from the airport’s arrival row for flights arriving and from its departure row for flights leaving. Cancelled shares are cancelled flights over scheduled flights on the same row.
Airports where leaving flights run furthest behind
| # | Airport | City | On time, flights arriving | On time, flights leaving | Leaving less arriving, points | Cancelled, arriving | Cancelled, leaving | Completed arrivals |
|---|---|---|---|---|---|---|---|---|
| 1 | CLT | Charlotte | 79.7% | 74.8% | -4.8 | 1.7% | 1.7% | 606,547 |
| 2 | BWI | Baltimore | 80.0% | 76.0% | -4.0 | 1.4% | 1.4% | 294,212 |
| 3 | DFW | Dallas–Fort Worth | 76.8% | 73.1% | -3.7 | 2.3% | 2.3% | 934,867 |
| 4 | MIA | Miami | 76.9% | 73.2% | -3.7 | 1.5% | 1.5% | 334,258 |
| 5 | ATL | Atlanta | 82.3% | 78.7% | -3.6 | 1.4% | 1.3% | 995,906 |
| 6 | DEN | Denver | 79.3% | 76.0% | -3.3 | 1.0% | 0.9% | 950,056 |
| 7 | MDW | Chicago | 79.5% | 76.4% | -3.1 | 1.2% | 1.2% | 236,430 |
| 8 | DAL | Dallas | 78.9% | 76.0% | -3.0 | 1.3% | 1.2% | 220,314 |
| 9 | IAH | Houston | 80.0% | 77.4% | -2.5 | 1.5% | 1.5% | 352,937 |
| 10 | ORD | Chicago | 76.9% | 74.5% | -2.3 | 2.0% | 1.8% | 930,756 |
| 11 | IAD | Washington | 82.7% | 81.0% | -1.7 | 1.3% | 1.3% | 167,387 |
| 12 | DTW | Detroit | 81.3% | 79.6% | -1.7 | 1.3% | 1.3% | 383,081 |
The same measure, the largest negative gaps. Each rate is flights that landed within 15 minutes of schedule over flights that completed, taken from the airport’s arrival row for flights arriving and from its departure row for flights leaving. Cancelled shares are cancelled flights over scheduled flights on the same row.
What this does not show
The two groups of flights are not the same flights. A flight leaving an airport lands somewhere else, so the departure rate carries the delays of every city the airport sends flights to, and the arrival rate carries those of every city that sends flights in. A wide gap can come from the mix of other airports as easily as from this one, and the report does not separate the two. Every flight is both a departure and an arrival, so across all airports the two groups are the same flights; for the airports listed here they are not. This is not a measure of delay at departure. It leaves out cancelled and diverted flights from the on-time rates, and shows the cancelled shares in the table beside them. Airports are included when both rows clear the airport floor of 5,000 scheduled flights.
Pages for what this report covers
Airports: San Francisco (SFO) · Anchorage (ANC) · San Antonio (SAT) · Memphis (MEM) · Ontario (ONT) · Buffalo (BUF) · Allentown (ABE) · Jacksonville (JAX) · Tucson (TUS) · Omaha (OMA)
Methodology
Calculated by FlyerIntel from 21,708,321 individual flight records in the US Department of Transportation Bureau of Transportation Statistics Reporting Carrier On-Time Performance dataset, covering July 2023 through July 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/. Each airport’s figures come from its arrival row and its departure row in the airport aggregate. Both on-time rates are arrivals within 15 minutes over completed flights on their own row, and the gap is the departure-row rate less the arrival-row rate. A gap is called wide only when a two-proportion z of at least 2 separates the two rates; this guards against naming a tie and is not an interval.
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 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