Real-world vs official fuel economy (WLTP, NEDC and EPA)
Updated 23 September 2026
The official figure comes from a lab test. On the road, almost every car uses more. This is the average gap measured with real data from millions of vehicles, and how to account for it when working out what a trip will cost.
Why doesn’t the official figure match real life?
Official fuel economy is measured on a rolling road, over a fixed drive cycle, at around 23 °C, with no air conditioning, no hills, no wind and the car in ideal condition. It is useful for comparing models, but it does not reflect how people actually drive: higher motorway speeds, climate control, luggage, traffic or cold weather.
Since September 2018 every new car in Europe has been certified under WLTP, a longer, faster and more realistic cycle than the old NEDC. In the United States, the EPA label has been calculated since 2008 from five test cycles plus an adjustment designed to match real-world use.
How much more cars use in reality
Since 2021, new cars sold in the EU record their real consumption with an on-board meter (OBFCM). Using that data, the European Commission found that cars registered in 2021 emitted — and therefore burned — 23.7% more than their WLTP value for petrol and 18.1% more for diesel. Weighted by the distance each car drives, the gap is 20.4% and 16.7%.
In 2026 the ICCT analysed about 8 million cars and found the gap has grown slightly, from 17–18% in 2021 to 19% in 2023 for both fuels. Data from private German drivers (Spritmonitor.de) shows a gap 5–7 points smaller: company cars tend to use more than average.
| Official figure | Petrol | Diesel | Source |
|---|---|---|---|
| WLTP (cars from 2018) | +20% (12–25%) | +17% (12–22%) | European Commission, ICCT |
| NEDC (older cars) | +39% (30–45%) | +39% (30–45%) | ICCT, From laboratory to road |
| EPA label (US) | +2% (0–10%) | +2% (0–10%) | Consumer Reports, Greene et al. |
Under the old NEDC cycle the gap was much larger: the ICCT put it at 39% for 2017 cars. If your car is older than 2018 and its spec sheet shows an NEDC figure, the correction is almost twice as large.
What about plug-in hybrids?
For plug-in hybrids (PHEVs) the gap is huge: the ICCT found real consumption to be 265% to 400% higher than the official figure (3.5 to 5 times as much), because the test assumes frequent charging while many are in practice driven mostly on the combustion engine. An average percentage makes no sense for a PHEV: use the figure from your trip computer.
How to measure your real fuel economy
- Fill the tank until the pump clicks off and reset the trip meter.
- Drive normally until you need to refuel (the more miles, the more reliable).
- Fill up again until the pump clicks off and note the litres or gallons.
- L/100 km = litres ÷ km × 100, or mpg = miles ÷ gallons.
The trip computer also works, although it usually reads slightly low. If you know your real figure, it is the best input for the calculator: it already includes your driving style, which can change consumption by 10–40% according to fueleconomy.gov.
How the calculator applies it
When you enter your fuel economy you choose where the figure comes from: WLTP, NEDC, EPA or your own real-world figure. For an official figure the calculator applies the correction in the table for your fuel; for a real-world figure it applies none. The uncertainty of the correction feeds, together with the other factors, into the “between X and Y” range of the result.
Sources
- European Commission (2024), COM(2024) 122 — real-world CO₂ emissions
- ICCT (2026), On the way to “real-world” CO₂ values? OBFCM data 2021–2023
- ICCT, From laboratory to road: A 2018 update
- Wali, Greene et al. (2018), fuel economy gaps (cites Consumer Reports 2013)
- fueleconomy.gov — Driving More Efficiently