What a car costs to run, and what the sticker leaves out
The EPA's annual fuel figure is a good number built on assumptions that are probably not yours. Here is what it assumes, how far real drivers land from it, and the costs it was never trying to measure.
Last reviewed September 2026
Every new car carries a fuel economy label, and the same figures are free to look up for anything sold since 1984. They are careful, laboratory-derived numbers. They are also frequently misread, in ways that cost real money at the point of buying.
The annual figure assumes a driver who may not be you
When this site says a 2019 Subaru Outback costs about $2,200 a year in fuel, that is the EPA's estimate on their assumptions: 15,000 miles a year, a fixed split of city and highway driving, and a national fuel price that was set when the figure was published.
Drive 6,000 miles a year and the number is roughly two and a half times too high. Drive 30,000 and it is half of what you will spend. Live somewhere fuel is expensive and every figure shifts together. The estimate is useful for comparing two cars on equal terms, which is what it was designed for, and much weaker as a prediction of your own bill.
The same nameplate is several different cars
“A 2019 Outback” is not one vehicle. The 2.5-litre four-cylinder is rated 28 MPG combined and about $2,200 a year; the 3.6-litre six is 22 MPG and about $2,800. Same model year, same showroom, $600 a year apart — roughly $3,000 across a five-year ownership, decided by an engine choice.
A 2018 Ford F-150 is worse: it exists in around twenty catalogue entries separated by engine, drivetrain and payload rating. Any tool that picks one and calls it “the F-150” is giving you a precise number for a truck you may not be buying. This site shows the configurations rather than choosing for you.
What drivers actually report
The EPA also collects figures submitted by owners. That is the one number in this whole area that is observed rather than modelled, and the gap between it and the label is genuinely interesting: for the 2019 Outback 2.5, twelve drivers average 27.0 MPG against a rated 28.
It comes with a trap. We sampled configurations that had submissions and sorted them by how many:
| Submissions | Spread against the EPA figure | Reported better than the label |
|---|---|---|
| 12 or more | 2.9% to 6.3% below | none of 4 |
| 4 or fewer | 18.9% below to 25.7% above | 4 of 8 |
- 12 or more
- Spread against the EPA figure:
- 2.9% to 6.3% below
- Reported better than the label:
- none of 4
- 4 or fewer
- Spread against the EPA figure:
- 18.9% below to 25.7% above
- Reported better than the label:
- 4 of 8
The pattern at healthy sample sizes is exactly what you would expect — real driving comes in a little under the laboratory figure, every time. Below about five submissions the numbers scatter wildly, and half of them claim better economy than the label, which is not how cars work. Two drivers of a 2015 Jeep Cherokee 3.2 average 25.4 against a rated 23.
A plug-in hybrid's reported economy measures its owner
Seven drivers of a 2021 Toyota RAV4 Prime report 81.6 MPG against an EPA petrol-only figure of 38. That looks like an extraordinary result and is really a statement about charging habits: the car has 42 miles of electric range, so someone who plugs in nightly burns very little fuel. Someone who never plugs in gets 38.
The figure is real and worth seeing. Presenting it as “115% better than the EPA estimate” would describe a habit as though it were a property of the vehicle, so this site shows the number and does not offer the comparison.
What none of this covers
Fuel is the part with good public data. It is not the biggest part.
- Depreciation — usually the largest single cost of owning a car, and often larger than fuel and maintenance combined over the first few years. No federal agency publishes it.
- Insurance — varies by driver, postcode and history more than by vehicle.
- Maintenance and repair — scheduled servicing is predictable; what actually breaks is what the complaint data speaks to, in kind rather than in dollars.
- Tyres, registration, financing, parking.
This site does not estimate any of them, because estimating them credibly needs data that is commercial rather than public, and an invented number is worse than an acknowledged gap. Where a figure would help, we point you at Edmunds and Kelley Blue Book, who do this properly.
Using the numbers well
- Take the ratio, not the total. If one configuration costs 25% more to fuel than another, that holds whatever your mileage is.
- Scale to your own driving. Your miles divided by 15,000, times the figure shown.
- Compare within a class. A saloon beating an estate on fuel is not news.
- Check the configuration matters before agonising. Sometimes the engine choice is $600 a year; sometimes it is $50.