The UK Car That Nearly Lost To A Wrong Gallon

Two Cars, One Notebook Page

The short version first, because the whole lesson sits inside it. A UK car's fuel figure passed through a converter set to the wrong gallon, lost more than nine MPG on paper, and nearly cost that car the sale. On one honest scale the UK car used 5.0 L/100km and the Japanese auction car used 5.13 L/100km. The car that looked thirstier was the thriftier one, and the only thing wrong with it was the gallon hiding behind its number.

Here is how it happened. Last year my uncle in Multan was choosing between two imported hatchbacks and asked me to compare them, because comparing things on paper is most of my job. Car one was a Toyota Aqua hybrid on a Japanese auction sheet, quoted at 19.5 km/L, with a landed price of about Rs 3.85 million through the importer. Car two was a Volkswagen Golf diesel on a UK listing, priced at about Rs 4.35 million by the time it reached Karachi. The Golf's downloadable spec sheet led with its European figure, 5.0 L/100km, and the family wanted everything in MPG, because MPG is the unit my uncle has used for forty years.

The importer had already done his own conversion for the Aqua and quoted it in his message as 55 MPG. That number was correct, as far as it went. It was built on the UK gallon. Nobody wrote the word gallon next to it, and that missing word is where this story lives.

The First Comparison, And Where It Went Wrong

My job was to add the Golf to the same page. I took its 5.0 L/100km from the spec sheet, opened the first converter that search gave me, and typed it in. The converter returned 47.04 MPG. I wrote it down without a flicker of doubt, because converters are supposed to be the boring, reliable part of the internet.

The page then read like this. Aqua: 55 MPG. Golf: 47 MPG. The Golf cost half a million rupees more and, on my page, drank noticeably more fuel doing it. My uncle read the page twice, shook his head at the Golf, and started asking the importer about Aqua colors. The decision was effectively made, on the strength of one conversion I had not checked.

The converter had not lied, exactly. It had answered a different question. It converted 5.0 L/100km into MPG with the US gallon, silently, the way most converters do. Run the same figure through the UK constant and the answer is 56.50 MPG. One input, two outputs, and a gap of 9.46 MPG between them, which is not a rounding error. It is a different car, on paper.

The cheaper car on paper was the thirstier one in real units. I wrote that sentence under the page that night, and it is why I now check the gallon behind every MPG figure before I check the figure itself. A conversion I have not interrogated is not a fact yet. It is a rumor with a decimal point, and this rumor had nearly picked my uncle's car for him.

What saved the Golf was an accident of tidiness. I keep the raw figures in my notebook before I convert anything, and that evening I looked at the raw column again: 19.5 km/L on one side, 5.0 L/100km on the other. Those two numbers sit close to each other however you convert them honestly. They did not sit close in my MPG column, and a column that disagrees with its own raw data is a column with a mistake in it.

Both Cars On One Scale

The fix took ten minutes and one rule: pick a single scale, convert everything onto it with named constants, and write the gallon next to every MPG figure. I chose L/100km, because both raw figures were already metric, and I used the converter on this site, which prints the gallon type on the result instead of burying it in a menu.

The Aqua first, three ways, so the auction figure stops being mysterious. Its 19.5 km/L becomes 5.13 L/100km, because 100 divided by 19.5 is 5.128. In MPG that is 55.08 with the UK gallon and 45.87 with the US gallon. Same car, same sheet, three honest numbers, each labeled.

Read the last two columns slowly, because they hold the whole trap. The Golf's US-gallon figure, 47.04, is lower than the Aqua's UK-gallon figure, 55.08, and that is the exact pair my first page compared. Mixed-gallons comparisons do not look broken. They look like ordinary tables, and they quietly reverse rankings whenever the two cars sit within about twenty percent of each other, which is most of the time in the same class.

The twenty percent is not a coincidence. NIST lists the US gallon at exactly 3.785411784 liters and the imperial gallon at 4.54609 liters, so the UK gallon holds about a fifth more fuel. A car travels further on a bigger gallon, earns a higher MPG number, and needs the bigger constant, 282.481, to land back on L/100km correctly. The US constant, 235.215, belongs to the smaller gallon. Swap them and you have not made a small error. You have graded one car with another car's ruler.

A MPG figure without its gallon is half a number. The other half decides whether the car wins or loses.

One more check kept me honest. The EPA publishes official ratings for the US versions of common models at fueleconomy.gov, city, highway, and combined. Those ratings run on US gallons, so they enter a comparison like this one only after conversion, never as raw numbers dropped next to UK figures. Two of the three sources on my page that week were fine. The third, my converter, was the one that needed a chaperone.

What The Flip Costs Over 100,000 Kilometers

Rankings matter because they steer real money. On the corrected scale the Golf uses 5.00 L/100km and the Aqua 5.13, a gap of 0.13 liters per 100 kilometers. That sounds like nothing until you stretch it over the life of a family car. Across 100,000 kilometers the Golf burns 5,000 liters and the Aqua 5,128 liters. The 128 liter difference, at the late October 2026 price of about Rs 379 a liter, comes to roughly Rs 48,600.

Honesty needs a second sentence here. The Golf also cost about Rs 500,000 more to buy, and no fuel gap of 0.13 L/100km repays that on its own. The family was never choosing on fuel alone. They were choosing between a dearer car they believed was thriftier and a cheaper car they believed was thriftier still, and the belief was the problem. Corrected, the picture read properly: pay more for the Golf and get the lower running cost and the stronger motorway manners, or save with the Aqua and accept a touch more fuel per kilometer. That is a real decision. The first page offered a fake one.

They bought the Golf. A year on, its tank log sits near 5.4 L/100km in mixed Multan and motorway use, a bit above the sheet figure once city traffic takes its share, which is the normal direction for a laboratory number meeting real roads. The Aqua would very likely have logged a similar small rise over its own sheet figure. Test numbers flatter every car a little. The gallon error flattered one car by a fifth, and that is the difference between a measurement and a mistake.

The habit this episode built in me is cheap, and my family has adopted it since: before money moves, somebody repeats the fill-to-fill routine on any car being bought used, the same routine I set out in The Used Car Fuel Check. Brochure numbers rank cars. Pump numbers confirm them.

The Check I Run Before Any Comparison

The use case that repeats most in my inbox is a cousin of my uncle's: a buyer in Pakistan holding a UK-import listing in one hand and a US-spec review of the same model in the other. Try it with round numbers. The UK brochure says 40 MPG. The American review says 36 MPG. The bigger number looks better, and it is worse: 40 MPG on the UK gallon is 7.06 L/100km, while 36 MPG on the US gallon is 6.53 L/100km. The review car, on paper the loser, is the thriftier figure. Test cycles differ between the two sources as well, which adds its own gap, but the gallon swap alone is enough to flip the ranking before cycles even enter the argument.

So the check is three questions, asked in order. Which gallon stands behind this figure? Which test produced it? And what does it become on the one scale I am using for every car in this comparison? The second question has its own guide, because European listings carry WLTP figures that need reading on their own terms: The WLTP Guide walks through that cleanup the way I now do it for clients.

None of this takes long. Converting a figure takes seconds, and naming its gallon takes one word typed next to it. What it costs is the willingness to distrust a tidy table, including, especially, your own. My notebook page from that week is still in the drawer, wrong column and all. I keep it as a reminder that the most dangerous number in a car comparison is not the one you cannot find. It is the one you converted without asking what it was standing on.