The most economical used cars in the real world, not the brochure

Every "most economical used cars" list you've ever read was built from brochure figures, which means it ranks cars by how well they did in a laboratory. Ours is built from what owners actually report getting. The two lists don't look the same, and the differences are the interesting part.

What are the most economical used cars in the real world?

Suzuki's naturally aspirated Dualjet petrols and small European diesels, mostly. Not the downsized turbo petrols that dominate the brochure rankings. The top real-world diesel in our Real MPG database is the Dacia Sandero Stepway 1.5 dCi at 71.8 mpg, and the top petrol is the Suzuki Celerio 1.0 Dualjet at 69.9 mpg, both averaged from owner submissions rather than from a test cycle.

Everything below is the owner-reported average for that engine, one entry per model, from generations with at least 100 submissions behind them. Every car links to its full page if you want the rest of its engines.

The 10 most economical used diesels

An asterisk means the percentage is measured against the newer WLTP official figure rather than the older NEDC one, so asterisked and plain percentages aren't comparable. The Sandero is the standout: 71.8 mpg against an official NEDC figure of 72.4 and a WLTP figure of 58.8, so it essentially matches the old test and comfortably beats the new one. Four of the ten beat their official figure outright, which is a rarer thing than these tables make it look: across the whole database only about one engine in ten manages it.

The 10 most economical used petrols

Four Suzukis in a top ten looks like favouritism until you notice what they have in common. Dualjet engines are naturally aspirated, they're fitted to light cars, and they're not being asked to haul 1,400kg on boost. They also do what they claim: three of the four sit within a few points of their official figure, and the Swift beats its WLTP number.

The Volkswagen Golf 1.0 TSI 85 is the one to look at hardest, because it has 2,635 submissions behind its generation, easily the largest sample on either list. 57.7 mpg from a family hatchback is genuinely good, and it's 98% of official, which for a downsized turbo is rare.

What about plug-in hybrids?

They top the raw table and the number means something completely different, so they get their own list and no percentage column:

A plug-in hybrid's real-world mpg is a statement about how its owners charged it, not about the car. Plug it in every night and do short trips and the figure is enormous. Buy one second hand, never charge it, and you're driving a heavy petrol car carrying a dead battery around. That's why the official figures for these are the least useful numbers in the whole database, and why we're not printing percentages next to them. If you're buying one, the questions that matter are whether you can charge it at home and what state the battery is in.

Why aren't the usual "economical" cars on these lists?

Because most of them earned their reputation on the old NEDC test, and small turbocharged petrols flatter that test more than any other type of engine. Across our whole database, engines of 1.0 litres and under return a median 79% of their official figure while engines over 2.0 litres return 86%, so the cars with the best brochure numbers are frequently the ones that miss them by the most. There's a full breakdown in why your MPG is lower than the official figure.

The clearest example isn't on either list precisely because of this. The Ford Fiesta 1.0T EcoBoost 100 from 2013 to 2017 claims 65.7 mpg, which would have put it near the top of any brochure-based ranking. Owners average 46.3. The same car's older, simpler 1.25 petrol claims much less and delivers a higher share of it.

How were these lists worked out?

Worth spelling out, because a ranking is only as good as its method:

• Source. Owner-reported averages from our Real MPG section, rebuilt from the archived Honest John Real MPG dataset alongside the official NEDC and WLTP figures and new reader submissions. 776 models, more than 319,000 submissions behind them in total.
• Threshold. Only engines whose generation carries at least 100 submissions. Submission counts are held per generation rather than per engine, so a popular generation lends its count to its rarer engines. That's the biggest weakness in the method and it's why the threshold is set high.
• One entry per model, so a single well-liked car can't fill half the table with its own engine range.
• Fuel types were checked by hand, not guessed from the engine name. Mild hybrids and plug-ins were pulled out of the petrol and diesel lists deliberately, because an engine name like "1.2 Dualjet SHVS" or "D6 AWD" hides electrical assistance that changes what the figure means.
• Self-selecting sample. People who submit fuel figures are people who track fuel figures, and they tend to drive more gently than average. Read these as what the car can do in careful hands, not as a promise.

Economical isn't the same as cheap to own

This is the bit these lists always leave out. A car that returns 70 mpg and needs a £3,000 gearbox is not a cheap car, and fuel is rarely the line that decides whether a used car was a good buy.

Take two off the tables above. The Vauxhall Corsa is genuinely frugal in 1.3 CDTi form, and the Corsa D also carries a documented ECU failure caused by rainwater getting to it, at £950 to £1,200. The Toyota Aygo makes both lists and the 2005 to 2014 car has documented premature clutch wear from around 20,000 miles and a habit of letting water into the boot and footwells, with the automated manual gearbox a much bigger bill again at £1,500 to £2,500. Neither of those is a reason not to buy the car. They're the reason to check the specific car before you do, because the difference between a good example and a neglected one is worth more than several years of fuel savings.

What should you actually check before buying one of these?

Starting from the honest bit: on provenance records we get exactly the same data as every other check. Outstanding finance comes from the lenders, theft markers from the police, write-off records from the insurance industry, and everyone reads the same registers. There's no private source and anyone claiming one is selling.

What no plate lookup does is read the advert. Every other check in the country starts with a registration and has never seen a word the seller wrote, which leaves the description as the one part of a listing nobody examines. Our free scan takes the advert text and reads it against the car's full DVSA MOT advisory history and the known faults for that exact model, engine and year, then gives you a verdict, a risk score and the biggest thing it found. No card, no password. So "economical, cheap to run, drives faultless" on a Corsa D gets checked against whether the advisory history is telling a different story.

It won't spot unrecorded accident damage or judge how well a past repair was done. That's not in any database at any price, which is why nothing here replaces looking at the car.

So pick the model off these tables, then run the free scan on the actual advert before you drive anywhere to see it. If it turns something up worth going further on, the free report upgrades on the same link afterwards without re-running anything, so there's nothing lost if the answer is walk away.

The short version

Ranked on what owners really get rather than on brochure figures, the most economical used diesels are led by the Dacia Sandero Stepway 1.5 dCi at 71.8 mpg, and the petrols by the Suzuki Celerio 1.0 Dualjet at 69.9. Naturally aspirated small petrols and modest diesels beat the downsized turbos that win brochure rankings, because those flatter the laboratory test and struggle on the road. Plug-in hybrids top the raw numbers only if their owners charge them. And fuel economy is the cheapest part of running a used car, so check what the specific car is carrying before you buy it for its mpg.