Why Better Gas Mileage Doesn't Save as Much as You Think: A Field Guide to Fuel Economy's Broken Arithmetic

1. The MPG Illusion: Why Everything You Know About Gas Mileage Is Nonlinear

Ask someone which is better: replacing a 15 MPG truck with a 20 MPG truck, or a 35 MPG car with a 50 MPG car. The second number is a 15 MPG improvement. The first is only 5 MPG. The second one feels bigger. It is not.

Over 10,000 miles driven per year, at $4 per gallon:

The third row is the shocker. Going from 10 to 11 MPG — a single mile per gallon — saves more fuel than going from 35 to 50 MPG, which is a 15 MPG jump. The math isn't broken. It's just that MPG measures the wrong thing for the question people are actually asking.

When a person asks "how much fuel will I save," they are asking about gallons consumed per distance — which is 1/MPG. Gallons per mile. The relationship between MPG and gallons per mile is a reciprocal curve: flat at the high end, steep at the low end. A 1 MPG improvement at 10 MPG cuts your fuel consumption by about 9%. The same 1 MPG improvement at 50 MPG cuts it by about 2%. The MPG number went up by 1 in both cases. The fuel bill dropped by vastly different amounts.

Larrick and Soll called this the MPG Illusion, and their paper demonstrated it across hundreds of participants. When people were asked to rank fuel-saving upgrades using MPG, they consistently picked the wrong ones — favoring large MPG jumps from already-efficient vehicles over small MPG jumps from gas guzzlers. When the same problem was presented in gallons per 100 miles, almost everyone got it right. The unit changes the intuition.

The practical consequence of the MPG Illusion is that US fuel economy standards, expressed in MPG, systematically undervalue improvements to the least efficient vehicles. A CAFE standard that raises fleet average from 25 to 30 MPG sounds modest. But the vehicles that actually achieve those gains are predominantly trucks and SUVs moving from ~15 to ~18 MPG — because that's where the cheap gallons are. The standard does more good than it sounds like, but the MPG framing makes it sound insufficient. Every debate about US fuel economy standards is fought on a battlefield tilted by the unit choice.

2. 235.215: The Number That Lives on Every Car Forum

Every car forum has a thread where someone asks "how do I convert MPG to L/100km" and someone replies with a number: 235. Divide by MPG, get L/100km. Divide by L/100km, get MPG. The number gets quoted as gospel. But it's not a definition — it's a consequence of two older ones.

Here is every digit, accounted for:

The factor in that last line is 2.352145834. That's the liters-per-kilometer equivalent of one gallon per mile. But L/100km is liters per hundred kilometers, so multiply by 100: 235.2145834. Round to 235.215. That's the constant. It is the exact SI translation of an American gallon consumed over an American mile, rescaled to the European convention of reporting consumption per 100 km.

In reverse: if a European car consumes 8.0 L/100km, how many US MPG? 235.215 ÷ 8 = 29.4 MPG. The math is symmetric because it's a pure inversion after accounting for the unit translation. The constant is only needed because we're crossing both unit systems (imperial→metric) and measurement conventions (distance-per-fuel→fuel-per-distance) in the same step.

The UK constant: 282.481

For Imperial (UK) gallons, the constant is different. 1 Imperial gallon = 4.54609 liters (exact — UK Weights and Measures Act 1963, reaffirmed 1985). Chase the same chain: 4.54609 L/gal ÷ 1.609344 km/mi = 2.824810 L/km. × 100 = 282.481. Same logic, larger gallon, larger constant. UK MPG numbers are always about 20% higher than US MPG for the same car. A Volkswagen Golf that gets 48 MPG in a British review gets 40 MPG on the US window sticker. The car burns the same fuel. The gallon it's divided by doesn't hold the same volume.

This is also why you should never skip the "(US)" or "(UK)" qualifier when discussing MPG online. A number without a gallon origin is a number without meaning. A British driver bragging about "55 MPG" is getting about 45.8 MPG in American terms — still excellent, but not the number the American reader thinks they're hearing.

3. The 20% Gap: Why a "Gallon" Is Anywhere From 3.79 to 4.55 Liters Depending on Which Side of the Atlantic You Park On

The gallon is not one unit. It's three, and two of them are still in active use for fuel economy.

• US liquid gallon = 231 cubic inches = 3.785411784 L. Defined by an English wine-gallon statute from 1707, adopted by the US after independence, metricized exactly in the 1959 agreement.
• Imperial (UK) gallon = 4.54609 L. Defined by the UK Weights and Measures Act of 1824 as the volume of 10 pounds of distilled water at 62°F (16.67°C) weighed in air against brass weights. Metricized in 1963 to exactly 4.54609 L.

The 1824 act was supposed to unify British measures. It abolished the wine gallon, the ale gallon, and the corn gallon — all of which had different volumes — and replaced them with a single Imperial gallon. The US, having left the empire 48 years earlier, kept the old Queen Anne wine gallon. That fork in 1824 is why a "gallon of milk" means different things in London and Los Angeles. And why a British car review's MPG figure means something different from an American window sticker's MPG figure, even though both say "MPG."

The difference is 20.1%. That's not a rounding error. A car that consumes exactly 8.0 L/100km gets:
· 29.4 MPG (US) — using 235.215 ÷ 8.0
· 35.3 MPG (UK) — using 282.481 ÷ 8.0

The car is identical. The fuel in the tank is identical. The test cycle might differ (EPA vs WLTP), but the unit-accounting difference alone shifts the number by 20%. Car magazines that publish UK MPG figures without saying so are creating a 20% phantom that American readers unconsciously import.

Car reviewers who cross the Atlantic constantly convert between these numbers in their heads. The rough shortcut: UK MPG × 5/6 ≈ US MPG. 48 UK × 5/6 = 40 US. Close enough for a podcast, not close enough for a purchase decision. Use the MPG to L/100km converter for the exact factor, and make sure you know which gallon you're starting from.

4. km/L: Japan's Quiet Compromise

Japan, Korea, India, and much of Southeast Asia measure fuel economy in kilometers per liter. It's distance-over-volume, like MPG — so it shares MPG's nonlinearity problem — but it's fully metric. No conversion to miles, no gallon variant to specify. One liter, one kilometer. The number means what it says.

A car rated at 15 km/L travels 15 kilometers on one liter of fuel. That's 35.3 MPG (US) or 6.67 L/100km. Multiply km/L by 2.352 to get US MPG. Divide 100 by km/L to get L/100km. The conversions are clean, decimal, and unambiguous.

Japan's 2030 fuel economy standard for passenger cars is 25.4 km/L (about 60 MPG US, or 3.9 L/100km). The Japanese test cycle (JC08, now transitioning to WLTC) is gentler than the EPA cycle, so the numbers aren't directly comparable to US window stickers. The unit at least is honest: km/L hides no gallon ambiguity, no MPG Illusion. It still carries the nonlinearity — the jump from 10 to 12 km/L saves more fuel than 20 to 25 km/L — but it eliminates the transatlantic gallon confusion.

Indian fuel economy labels use km/L. South Korean labels use km/L. Thai stickers use km/L. For the majority of the world's car buyers by population, fuel economy is expressed in kilometers per liter. The US, UK, Canada, and Europe are the outliers — and even Europe moved away from MPG decades ago.

If you drive a Japanese-market car and need to convert km/L to something an American mechanic or a European fuel app understands, the km/L to MPG and MPG to km/L converters handle it directly.

5. Why Europe Regulates in L/100km — and the US Doesn't

In 1975, the US Congress passed CAFE (Corporate Average Fuel Economy) standards in response to the 1973 oil embargo. The unit: miles per gallon. It was the unit Americans knew, and the standard was designed for a world where American cars averaged 13 MPG and Japanese imports were hitting 25. The unit made sense when every improvement was in the steep part of the reciprocal curve. At 13 MPG, a 2 MPG gain is a big deal.

Europe took a different path. The EU began regulating CO₂ emissions from passenger cars in 2009, and CO₂ is directly proportional to fuel consumption — grams of CO₂ per kilometer equals liters of fuel per 100 km times a constant (roughly 23.2 for gasoline, 26.5 for diesel). Europe's unit for fuel economy followed the carbon accounting: L/100km. It's consumption-based, linear with emissions, and avoids the MPG Illusion entirely.

The difference is not academic. Under CAFE, an automaker's fleet-wide target is an MPG number, and every vehicle's contribution is harmonic — the fleet average is the harmonic mean of individual MPG ratings. Under EU CO₂ regulation, the fleet target is a grams-per-km number, and every vehicle's contribution is arithmetic — the mean of individual g/km ratings. The arithmetic mean is simpler to calculate, easier to optimize against, and doesn't penalize low-MPG outliers as harshly as the harmonic mean. The unit choice cascades into the regulation design, and the regulation design shapes what cars get built.

The US has been repeatedly advised to switch to a consumption-based metric. A 2010 EPA report recommended gallons per 100 miles as a supplementary label. The National Research Council recommended it in 2011. The US window sticker now shows gallons per 100 miles in small print below the MPG number — but MPG remains the headline. No US regulation has ever been proposed in GPM. The inertia of a unit that 300 million people think they understand is stronger than the arithmetic case against it.

6. The Gimli Glider: 22,300 Pounds of Fuel Doesn't Equal 22,300 Kilograms

On July 23, 1983, a 4 PM Air Canada flight from Montreal to Edmonton — via Ottawa — pushed back from the gate at Ottawa Macdonald-Cartier International Airport. The Boeing 767-200, registration C-GAUN, was Air Canada's fourth 767 and the first to be delivered with metric instrumentation. Until then, the airline had operated entirely in imperial units: pounds of fuel, feet of altitude, nautical miles of distance. The 767 spoke kilograms, meters, and kilometers.

The flight management computer required fuel in kilograms. The ground crew, trained on pounds, calculated the fuel requirement using the specific gravity of jet fuel — 0.803 kg/L — but they used a pounds-per-liter figure by mistake. The combination of a broken fuel quantity indication system (both channels inoperative, deferred under the MEL with a dipstick as backup), a misapplied conversion factor, and an aircrew that triple-checked the arithmetic without catching the unit error resulted in 22,300 pounds of fuel being loaded instead of 22,300 kilograms.

22,300 pounds is 10,115 kilograms. Less than half the required fuel for the Ottawa-Edmonton leg.

At 41,000 feet, over Red Lake, Ontario, the left engine flamed out. Then the right. A 767 without engines is a 90-ton glider with a glide ratio of about 12:1. Captain Bob Pearson — an experienced glider pilot — calculated that Winnipeg was out of range but that a decommissioned Royal Canadian Air Force base at Gimli, Manitoba, might be reachable. First Officer Maurice Quintal did the math on the standby instruments as the aircraft descended at 2,500 feet per minute.

Gimli's runway had been converted to a drag racing strip. A sports car event was in progress. The 767 came in fast, no flaps (hydraulics gone), no nose gear (gravity drop failed to lock). Pearson executed a forward slip — a maneuver no one had ever attempted in a 767, taught in gliders to steepen descent without gaining speed — to shed altitude over the crowd. The aircraft touched down hard, nose gear collapsed, and the 767 ground to a stop on its nose and main gear, 100 feet from the spectators.

Zero fatalities. Sixty-one passengers, eight crew, and everyone on the ground walked away. The aircraft was patched up in Gimli, flown out two days later, and returned to service until 2008. It earned the nickname "Gimli Glider" and became the most famous case study in aviation human factors — a textbook example of how a chain of individually minor errors, any one of which could have been broken, cascaded into a near-catastrophe.

The unit conversion at the heart of the Gimli Glider — pounds to kilograms, multiplied by a specific gravity that was itself ambiguous between metric and imperial — is not strictly a fuel economy problem. But it is the same species of error. A fuel unit is a fuel unit. When the number is 22,300 and the label is ambiguous, the plane loses both engines at cruise altitude. The difference between MPG(US) and MPG(UK) is 20%. That is not a rounding problem. It is the same gap between a flight that lands and a flight that glides into a drag strip.

7. Master Fuel Economy Conversion Table

Every fuel economy unit relationship covered by the converters on this site. All conversion factors are exact — anchored to the 1959 international inch (0.0254 m) and the US gallon (3.785411784 L) or UK gallon (4.54609 L) as specified.

Common conversions at a glance

8. Every Fuel Economy Converter on This Site

The L/100km number is the universal translator. It does not depend on what kind of gallon you started with. A car that burns 7.0 L/100km burns 7.0 L/100km whether you're filling up in Texas, Tokyo, or Toulouse. The conversion from that universal number to MPG depends on which gallon you want. The conversion from that number to km/L is 100 ÷ 7 = 14.3 km/L, with no gallon involved. When in doubt: convert everything to L/100km first. It is the only fuel economy representation that is simultaneously metric, linear, and gallon-free.