BYD’s 623-Mile Range Figure Comes From a Test Most Buyers Never See

Two range figures can describe the same battery without either of them being wrong. A BYD announcement this week put a five-seat electric SUV above 623 miles, and a separate report on the Denza N8 quoted roughly the same number. Read as a promise about a drive, both figures are misleading. Read as the output of a specific laboratory procedure, both are exact.

The procedure is the part nobody prints.

A worn car tyre against a concrete wall with a 623 MI data sticker

What a test cycle actually measures

Range is not measured by driving until the battery dies. It is computed from a standardised cycle: a scripted pattern of acceleration, cruising, braking and idle time run on a dynamometer, with the result scaled by energy consumption. Change the script and the same pack produces a different headline number.

The Chinese CLTC cycle used for these figures leans heavily on low-speed urban driving, includes a long idle share, and assumes mild conditions. Regenerative braking recovers energy in stop-and-go segments, which flatters an electric drivetrain. The US EPA procedure includes higher speeds, more aggressive acceleration, and a correction applied to the raw laboratory result. WLTP, used elsewhere, sits between them.

None of this is a scandal. Every market needs a repeatable yardstick, and a repeatable yardstick is by definition artificial. The problem begins when a number crosses a border without the label that produced it.

Why the gap is so large for electric cars

Combustion engines are least efficient in the exact conditions where the urban-weighted cycle spends its time. Electric motors are the opposite. They are superb at low speed, recover energy under braking, and lose relatively little while idling. A cycle built around city driving therefore flatters an EV far more than it ever flattered a petrol car.

  • Aerodynamic drag rises with the square of speed, so sustained motorway driving is the single biggest range reducer.
  • Cabin heating draws directly from the pack in cold weather, with no waste heat from an engine to borrow.
  • Tyre choice, roof boxes, payload and headwinds all move the result, and none appear in a laboratory figure.
  • Battery chemistry changes how sharply cold weather bites, even when two cars share a headline number.

A range figure is a comparison tool, not a distance.

The practical consequence is that a 623-mile laboratory result and a 400-mile motorway day are not in conflict. They are two different measurements of one vehicle. Buyers who compare a CLTC figure against an EPA figure are comparing two scripts, not two cars.

What to look at instead of the headline

Usable battery capacity in kilowatt-hours is the least ambiguous starting point. Divide it by a realistic consumption figure for the driving you actually do and you get a number with fewer assumptions attached. Charging behaviour matters just as much: how long the pack sustains its peak rate, and whether the quoted range assumes charging to 100 percent every time.

The second useful number is the same car’s figure under a second cycle. When a manufacturer publishes both CLTC and WLTP for one model, the ratio between them tells you more about real-world expectations than either figure alone. It also makes cross-shop comparisons possible without doing the conversion in your head.

Third, look for the conditions attached to the claim. A range announced alongside a new model launch frequently describes a specific variant, wheel size, and temperature window. The version on a forecourt with larger wheels is a different measurement.

Why this keeps happening

A single large number is the most portable fact a car has. It survives translation, fits in a headline, and ranks well against a rival’s single large number. Test-cycle labels do not survive any of those journeys, so they get dropped at the first relay.

That is unlikely to change while the marketing value of the number exceeds the cost of the caveat. What can change is the question a buyer asks. Not “how far does it go” but “under which script, at what speed, in what weather” — which turns an impressive figure into a usable one.

So the number is not wrong. It is just local.

A city cycle, a mild day, a light foot, a full charge and no roof box all stack in the same direction, and because each of them quietly removes a penalty that a real drive would impose, the laboratory result ends up describing a journey that almost nobody takes from start to finish.

For reference, the US EPA electric-vehicle label guidance explains how its range estimate is derived and adjusted, and BYD’s own material states figures under the applicable local cycle. Put the two side by side and the discrepancy stops looking like a contradiction.