The EPA prints a five-year fuel cost comparison on every new car sticker: what you save, or spend, against an average vehicle. Across 8,903 models from 2020 onwards, the number is negative far more often than not.
| Five-year fuel comparison | Models | Item |
|---|---|---|
| Costs MORE than average | 6,421 | 72% |
| Saves money | 2,482 | 28% |
| Market average | −$3,423 |
The average isn't the median
If "average vehicle" meant the middle of the market, half the cars would land on each side. Instead 72% are worse than it.
The benchmark is a fixed reference figure rather than a running median of what's on sale, and the market has drifted heavier and thirstier than it. Which means the comparison flatters nothing — most new cars are shown spending money relative to a standard almost nobody meets.
The spread is larger than the price of a car
| Brand | Five-year fuel vs average | Item |
|---|---|---|
| Lucid | +$7,684 | 100% of models save |
| Tesla | +$7,542 | |
| Rivian | +$6,035 | |
| Hyundai | +$1,882 | 67% save |
| Aston Martin | −$11,421 | 0% save |
| Ferrari | −$12,707 | |
| Lamborghini | −$17,335 |
The extremes: a Bugatti Chiron at −$31,750 over five years, against hydrogen models like the Toyota Mirai at +$11,000. That's a $42,750 range — more than a new mid-market car costs outright.
Where the figure misleads
Hydrogen tops the table because of how it's counted. The Mirai and Nexo score +$11,000 partly because fuel costs are calculated on manufacturer-provided free-fuel arrangements. Hydrogen is not free, and stations barely exist outside California.
Electricity costs vary enormously. The EPA uses a national average rate. Charging at home overnight in one state and at a public rapid charger in another produce completely different real numbers.
It's fuel only. Insurance, tyres, depreciation and repairs aren't in it — and on the brands at the bottom of this table, fuel is not the expensive part of ownership.
How to use it properly
Compare within a segment, not against the benchmark. The useful question is whether this crossover costs more to fuel than that crossover, not how either compares to a fixed reference.
Use your own mileage. The figure assumes 15,000 miles a year. If you drive 8,000, halve it — and we found actual mileage varies 5.5x by model.
Check the fuel grade separately. A premium requirement adds around $500 a year, which is where turbo engines quietly lose their advantage.
Pull the original sticker for a used car. It carries these figures for the exact vehicle as built — a window sticker lookup by VIN retrieves it, and a VIN check confirms the engine it left the factory with.
Frequently asked questions
What does the five-year fuel savings figure on a window sticker mean?
It compares the car's fuel cost over five years against an average vehicle. For 6,421 of 8,903 models from 2020 onwards — 72% — the figure is negative, meaning the car costs more to run than the benchmark.
Why do most cars cost more than the average vehicle?
The benchmark is a fixed reference rather than a running median of what's on sale, and the market has drifted heavier and thirstier than it. So the comparison shows most new cars spending money against a standard few meet.
Which brands save the most on fuel?
Lucid at +$7,684 over five years with 100% of models saving, Tesla at +$7,542 and Rivian at +$6,035. At the other end Lamborghini is −$17,335 and the Bugatti Chiron reaches −$31,750.
Is the EPA fuel cost estimate reliable?
It's fuel only — no insurance, tyres, depreciation or repairs — and it assumes 15,000 miles a year at national average energy prices. Hydrogen models score unusually well partly because free-fuel arrangements are counted.
Denis Kataev is a serial solopreneur and the founder of ForCar. With 15+ years in software engineering and 10 years in SEO, he builds data-driven products end to end — backed by a sharp eye for design. At ForCar he mines proprietary vehicle datasets, turning raw numbers into buying advice you can actually trust.
Keep reading
All articles →
14,007 Complaints Mention Heat. 3,896 Are About the Air Conditioning Stopping.
Cold exposes wear. Heat exposes capacity. The complaints filed in summer describe a different set of failures from the ones filed in January, and one of them is a safety problem people do not treat as one.
15,810 Complaints Describe a Fault That Happens in the Morning. Only 1.4% Involve a Crash.
Some faults only exist before the car has warmed up. They are harmless by the numbers and expensive by the time anyone takes them seriously.
Two in Five Complaints Mention the Dealer. 7,186 Say the Fault Could Not Be Reproduced.
Complaint text is usually mined for what broke. Reading it for who the owner dealt with produces a different picture: in two cases out of five, the visit to the dealer is the thing being described.
Mark ForCar as a preferred source and our data shows up first in your Google results. One click, no account.