Rankings / Methodology

Methodology

Everything here is calculated from airline filings to the US Department of Transportation. This page lists the formulas, the choices we made and the limits.

For a term used below, open its definition in the glossary.

The main figure

Passenger-miles per gallon is the number of passenger-miles an airline flew in scheduled service, divided by the gallons of jet fuel it burned on those flights. A passenger-mile is one paying passenger carried one mile. We use it because it counts both how big the aircraft are and how full they fly.

Two related figures sit beside it. Seat-miles per gallon uses seats offered instead of passengers carried, so it ignores how full the flights were. Load factor is passenger-miles divided by seat-miles.

CO2 per passenger-mile is gallons burned times 9.75 kg of CO2 per gallon, divided by passenger-miles. The 9.75 kg factor is the US EPA's figure for kerosene-type jet fuel in its GHG Emission Factors Hub. We apply the same factor to every year so the trend reflects fuel use and not changes in the factor.

Where the data comes from

• Fuel: BTS Form 41 Schedule P-12(a), gallons of fuel consumed by carrier, covering domestic and international flying. Some airlines also fly cargo or charter flights that BTS can file as scheduled service, so we keep only the share of each airline's fuel that matches its share of air time on scheduled passenger flights, taken from the same T-100 file that supplies the traffic. Regional airlines report their own fuel in BTS Schedule T-2 (gallons issued by aircraft type), and we use that for them.
• Traffic: BTS Form 41 Schedule T-100 segment data for US carriers. For each carrier, route and month it gives departures, seats and passengers. We keep scheduled passenger service and skip charter and cargo-only flying.
• Seat-miles are seats times the segment distance, added up over all segments. Passenger-miles are passengers times the segment distance.
• Years: only calendar years where all twelve months are present in both files. 2000 to 2025 are in the data. A part-year would flatter or penalise an airline depending on the season.
• Size cutoff: airlines with fewer than one billion seat-miles in a year are left out of that year. The airlines ranked in 2025 cover 100.0% of US scheduled seat-miles.

Regional airlines

This is the hardest part, and it matters most for United, Delta, American and Alaska. Many regional flights wear a major airline's name but are flown by a separate company under contract. SkyWest, Republic, Envoy, Endeavor and PSA are examples. In the traffic filings the regional reports the seats and the passengers. In the main fuel filing (Schedule P-12(a)) the regional reports almost nothing, and the major airline reports only the fuel for its own aircraft. The regional's fuel is in a different filing: Schedule T-2 gives the gallons issued to each aircraft type, and every regional reports its own there.

So a network's figure here is its own mainline fuel plus the fuel of the regionals flying for it, divided by the traffic of both. We split each regional's flying and fuel between the airlines it flies for with the method below, and we treat the airline selling the seats as the owner of the flying. Subsidiaries count as the airline itself: Horizon Air at Alaska, and Continental Micronesia at Continental.

Some regional filings are missing or wrong for some years: Chautauqua and Piedmont filed no fuel for years, and a few years show figures no aircraft could reach. A regional carrier-year with fewer than 30 or more than 70 seat-miles per gallon is treated as an estimate and flagged on its page. Most of these regionals still report their flight hours and aircraft types in T-2, so the first choice is to price their own hours at the burn per hour that other airlines report for the same aircraft type that year (the types nobody reports use the nearest year that has one). Where that figure is also off the scale, usually because the regional's traffic and hours don't match, the estimate falls back to the median seat-miles per gallon of the regionals that reported properly that year. Each page says which method it used.

The split works from hubs. For each airline that has regional partners, we count its own mainline seats at every airport. A regional flight between two airports is credited to the airline whose mainline presence best fits both ends. A SkyWest flight between Salt Lake City and a small town goes mostly to Delta, because Delta dominates Salt Lake City. A flight touching the hubs of two airlines is split by how well each fits.

Each regional also has an overall mix of partners. We estimate that mix by starting from an even split and repeatedly re-weighting it toward the airlines whose hubs the regional actually serves, until it settles. Regionals that fly for one airline come out at or near 100% for that airline. The table above shows the result.

How far to trust it

We can test the method against regionals whose partner is public record. 7 regional and period combinations are checked, such as Envoy and Piedmont for American and Endeavor for Delta, covering 89 regional-years. The method puts 96.2% of that flying with the right airline. The worst single case is PSA Airlines in 2016, where 73% went to American Airlines. A test fails the build if this accuracy falls.

Regionals that fly for several airlines, such as SkyWest and Republic, can't be tested this way, and they're where the error is largest. One gap is known: SkyWest flies for Alaska as well as United, Delta and American, and about a tenth of its flying touches Seattle and Portland, but the hub method gives Alaska none of it, so Alaska's figure may be a few percent too high. To show it, we run the split twice: once with the partner-mix step and once without it. For airline-years with regional allocations or a fuel adjustment, the home ranking and airline directory cards show the stored allocation bounds as an amber line on the meter and as text beside the point value, even when the rounded bounds are equal. Cards flag neighbouring results whose ranges overlap, including a reported point inside an estimated range: their order is uncertain, so treat them as tied. Airline-years with neither regional allocations nor a fuel adjustment show only the reported point, with no amber band or allocation label. Frontier has one exception, described next.

Frontier's filed fuel doesn't hold together for 2010 and 2011. Its Airbus A319s burned about 810 gallons per flight hour in 2008 and 2009, about 1,100 in 2010, about 1,030 in 2011 and about 840 in 2012, on the same number of hours and departures. Both the P-12 and T-2 filings show the jump, so it isn't a download error, and no aircraft burns a third more fuel an hour for two years and then stops. For those two years the main figure takes Frontier's own hours on each aircraft type and prices them at that type's burn per hour in 2009 and 2012. The fuel as filed is kept as the other end of the range and is marked with an asterisk in the year table. Other airlines and years weren't changed.

Fleets: age and seats

The fleets page uses BTS Form 41 Schedule B-43, the annual inventory every large airline files, with one row per aircraft: tail number, model, the year the airframe was built, the usual number of seats and whether it's in operation. We count aircraft in operation with 10 or more seats, which leaves out freighters, helicopters and business jets, and for each airline we report the average age (the filing year minus the year built), the share under five years old and twenty or older, and the average seats per model. The inventory runs from 2006.

The models are typed by each airline in its own style and change from year to year, so we group them into families by their text (for example every 737-800 variant together) and check the result against the airline's known fleet. The seat count is the airline's own figure and can lag a cabin refit. The same inventory covers only the aircraft an airline files for itself, so a mainline airline's fleet leaves out the regional jets flown by its partners, which appear under the regional airline. To test whether age explains efficiency we compare average age with seat-miles per gallon for the airlines that fly no regional partner and have at least 20 aircraft; with so few airlines the result is a rough guide, not a finding about aircraft age in general.

Routes, aircraft types and the guides

The route pages, aircraft pages and guides use the same filings at a finer level. Schedule T-2 gives, for each airline and aircraft type, the fuel issued and the hours in the air; we keep the share of those hours flown on scheduled passenger flights, the same rule as the airline figures. Dividing fuel by hours gives gallons per airborne hour for that airline and type, and the aircraft pages show it directly.

Airborne hours run from take-off to landing. Block hours include ground time too; they aren't the denominator here. Browse the manufacturer hubs for type comparisons.

The private jet comparison explains how hourly fuel burn is divided between the people aboard.

Each route page also compares the average time in the air in the two directions, from the same T-100 flights (air time is wheels-up to wheels-down, scheduled passenger service only). We call a difference meaningful when it's at least 6 minutes and 4%. Which direction is westbound comes from the longitude of each city's busiest airport. The extra fuel is the extra minutes at the route's average gallons per flight hour, so it's an estimate of the cost of the longer direction, not a separate measurement.

When a route page compares 2019 with 2025, each year uses that year's filings. A few filings in 2019 can't be right for the aircraft they describe: Endeavor reported 145 to 245 gallons an hour for jets that burn 450 to 640, and Alaska's A320 and A319 figures were 1,482 and 239 where the two types burn about 920 together, which looks like fuel assigned to the wrong type. For any airline and aircraft type whose gallons per hour fall outside 0.75 to 1.35 times the median for that type that year, the route figures use the type average instead.

For a route, T-100 gives each airline's flights, seats, passengers and air time on a pair of city markets, by aircraft type. Fuel per flight is gallons per airborne hour times the air time of that flight, and CO2 per passenger is that fuel, at the EPA factor, divided by the passengers per flight. If an airline reports no fuel for a type, the type's average across airlines is used. Averages over all flights of a type include taxi and holding time spread over the hours, so short routes are probably understated, and the figures don't weight cabins, so a premium-heavy flight looks worse per passenger than its economy seats really are. Only domestic city pairs are covered, and a city market is a whole metro area, so New York means Newark, JFK and LaGuardia together.

For driving we take road distance as 1.17 times the straight-line distance. That's the median of 30 of these city pairs, measured between their busiest airports on OpenStreetMap roads, with a range of 1.07 to 1.32. Cars use the EPA figure of 0.297 kg of CO2 per vehicle-mile (0.394 for a pickup or large SUV), divided by the people in the car. Bus and Amtrak use the EPA figures per passenger-mile, and rail is shown only for city pairs with a direct Amtrak train, taking rail distance as about the road distance. The private jet comparison uses published hourly fuel burn for the jets, as listed on the page, against our measured airline figure per passenger-hour.

What the figures leave out

• Method change, 22 October 2026. Earlier versions of this site used the scheduled service fuel total as filed. That overstated the fuel burned by passenger flights at airlines with freighters or charters, most of all Sun Country (its figure rose by 30 to 51 percent in each year from 2020 on once corrected) and Hawaiian in 2024 and 2025 (up 4 and 9 percent). It also moved ATA, which flew military charters alongside its scheduled service, down by about a fifth to two-fifths in every year it appears. The check that supports the new method: on the Boeing 737-800 in 2024, Sun Country, Southwest and Alaska each burned about 900 gallons per airborne hour in BTS Schedule T-2, so fuel follows hours flown.
• Method change, 22 October 2026 (regional fuel). An earlier version assumed the mainline airline's fuel already included the regional flying it sold, and added regional seat-miles with almost no regional fuel. BTS Schedule T-2 shows that regionals report their own fuel and the mainline figure covers only its own aircraft. Adding the regional fuel lowered the 2025 figures for the four network airlines: American from 68.3 to 55.9 passenger-miles per gallon, Delta from 66.6 to 58.0, United from 65.3 to 58.2 and Alaska from 79.5 to 71.6. Airlines without regional partners didn't change.
• Cargo in the hold. Fuel burned to carry freight and mail in passenger aircraft is counted against passengers. Airlines with big widebody fleets carry more belly cargo, so their passenger figure looks slightly worse than it would on a split basis.
• Other warming effects. The CO2 figure counts carbon dioxide from burning fuel. Contrails, nitrogen oxides and the emissions from making and moving the fuel aren't included.
• Foreign airlines. Only US carriers file these forms. Foreign airlines flying to the US aren't in the data.
• Mergers. BTS combines merged airlines' filings from a set date. A merged airline's line can jump in that year because it starts to include the other airline. We keep the earlier airline as its own entry before that date.
• Years we dropped. ATA Airlines in 2008 (it stopped flying part-way through the year) was left out. A year is dropped when the airline stopped flying mid-year, so its fuel and traffic cover only a few months, or when the figure is one no jet airline could reach, which means the fuel and traffic filings don't describe the same flying.
• Different routes. An airline flying long routes with few stops and one flying short hops aren't directly comparable. Average trip length is listed on the home ranking and airline directory cards so the difference is visible.

Rebuilding it

All inputs are public BTS files and one EPA table. The data page has the output as a CSV. Figures update when BTS publishes a new year.

Source references: BTS TranStats terminology and the US EPA GHG Emission Factors Hub.

For a possible error, follow the corrections policy or contact us with the page and data year.