Fuel Efficiency Converter

Fuel efficiency can be measured two opposite ways, and the two don’t move in the same direction, which is the source of endless confusion. Distance per fuel (mpg, km/L) goes up as a car gets thriftier; fuel per distance (L/100km) goes down. They describe the same thing inside out.

Key conversions:

• L/100km = 100 ÷ (km/L) (an inverse relationship)
• km/L to mpg (US) = km/L × 2.35215
• km/L to mpg (UK) = km/L × 2.82481

For a sense of scale:

• An efficient car: 5 L/100km (47 mpg US)
• An average car: 8 L/100km (29 mpg US)
• An SUV or truck: 12 L/100km (20 mpg US)
• A hybrid: 4 L/100km (59 mpg US)

The big trap is comparing US and UK mpg figures as if they were the same. They aren’t, because the Imperial gallon is about 20 percent larger than the US gallon. A car rated “40 mpg” in Britain is really about 33 mpg in American terms. Always check which gallon a number uses before comparing cars across the Atlantic.

The inverse nature of L/100km also makes savings deceptive. Dropping from 15 to 12 L/100km saves more actual fuel than dropping from 8 to 6, even though the second looks like the bigger improvement, because the gains shrink as efficiency climbs.

Run the same comparison in mpg and it gets worse, because mpg exaggerates at the top of the range. Going from 10 to 12 mpg saves about 17 gallons over 1,000 miles. Going from 40 to 50 mpg, a much more impressive-looking jump, saves 5. That is why the US window sticker now prints gallons per 100 miles beside the mpg figure: it runs the same way round as L/100km and it tells you the truth about where a fleet actually saves fuel. Anyone trading a truck for a slightly better truck is doing more good than someone trading a good hybrid for a slightly better one.

One more thing worth knowing when you compare two cars on paper. Published economy figures come from a standardised test cycle, not from driving, and the gap between the sticker and reality is larger for hybrids and small turbo engines than it is for a plain petrol car, because both do unusually well on a gentle test loop. Treat the numbers as a ranking rather than as a promise.