Cover story
The Range Lie
Of the 296 machines in this index that publish a range claim, 246 also publish a battery size. Run the physics and the two numbers rarely survive each other. We did the arithmetic on every one.
By The Data Desk Jul 11, 2026 7 min read 1,489 words
The number is printed where you cannot miss it — on the listing page, on the box, sometimes on the frame itself. Up to 160 kilometres. It is the first thing a buyer reads and, for most buyers, the reason the machine gets bought: that figure is the promise that the battery will outlast the day. The battery, meanwhile, sits lower on the same spec sheet, described in watt-hours, and says something quite different. This story is about what happens when you make the two numbers talk to each other.
They talk in one sentence of physics. A battery stores energy; a motor spends it; the distance a machine can assist you is the first number divided by the rate of the second. Everything else — the marketing copy, the mode names, the asterisk — is negotiation with that sentence.
The two numbers on every spec sheet
The index tracks 669 e-bikes across the Canadian and US markets. Of those, 296 publish a range claim, 340 publish a battery capacity in watt-hours, and 246 publish both. That last group is the one that can be audited, because a machine that states its battery size has already told you roughly how far it can assist you — whether or not the range claim agrees.
The audit needs one assumption, and we state it rather than hide it: on the road, assisted riding draws somewhere between 15 and 25 watt-hours per kilometre. Gentle assist on flat pavement with a light rider sits near the bottom of that band; hills, headwinds, cargo, cold and throttle push toward the top and past it. We take 20 Wh/km as the midline, which makes the arithmetic almost embarrassingly simple: a realistic range estimate is the battery’s watt-hours divided by twenty. Divide by 15 instead and you get the optimistic bound — the most the battery gives up under the friendliest realistic conditions. Divide by 25 for a hard winter commute.
The model is deliberately generous. It assumes every published watt-hour is usable, the cell is new, and the day is warm. Batteries age, controllers reserve a buffer, and January exists. If the arithmetic errs, it errs in the maker’s favour.
What the arithmetic returns
Across the 246 machines that publish both numbers, the median battery is 749 Wh. Divided by twenty, that supports about 37.5 km of assisted riding. The median range claim on those same spec sheets is 126 km.
For the typical machine, the claim runs roughly three times what its own battery midline supports — the median gap is +198%. And this is not a story about a few loud outliers: 238 of the 246 machines (96.7%) claim more than their midline estimate, and 234 (95.1%) claim more than even the optimistic battery-Wh-divided-by-15 bound. On this shelf, a range claim that survives its own battery arithmetic is the exception — about one machine in twenty.
There is a second way to read the same data. Take each machine’s claim at face value and ask what consumption rate would make it true. The median answer is 6.7 Wh per kilometre — less than half the bottom of our band. A motor sipping 6.7 Wh/km is a motor doing very little motoring. Most of that kilometre is you.
The pattern holds everywhere, but not evenly:
| Category | n | Median claim | Battery-math midline | Median gap | Claims past even Wh÷15 |
|---|---|---|---|---|---|
| Commuter | 90 | 123 km | 35.1 km | +285% | 98.9% |
| Fat tire | 34 | 125 km | 48 km | +160% | 97.1% |
| Mountain | 26 | 129 km | 48 km | +169% | 100% |
| E-trike | 24 | 137 km | 48 km | +185% | 100% |
| E-dirt | 22 | 130 km | 48 km | +170% | 72.7% |
| Folding | 20 | 113 km | 36 km | +220% | 100% |
| E-scooter | 15 | 65 km | 43.7 km | +94% | 73.3% |
| Cargo | 6 | 205 km | 40.8 km | +515% | 100% |
| Moped-style | 5 | 161 km | 36 km | +347% | 100% |
| Road | 4 | 124.5 km | 28.3 km | +241% | 100% |
One row deserves a pause. E-scooters — the one category with no pedals, where the claim cannot lean on your legs — sit closest to the battery math, at a +94% median gap and an implied 10.3 Wh/km. The moment a machine has pedals, the marketing gains a co-author: you.
The ten widest gaps
Below are the ten machines whose claims sit furthest above their own battery arithmetic. The columns are the maker’s two published numbers and the range band those watt-hours support at 15–25 Wh/km. We print them neutrally, because a gap this wide can mean one of two things: the claim assumes the motor is barely helping, or one of the two published numbers is simply wrong. Both numbers came from the maker’s own listing. The arithmetic only says they cannot both describe the same assisted ride.
| Machine | Claim | Battery | Battery math says | Implied Wh/km |
|---|---|---|---|---|
| Vtuvia Gemini 1000W | 145 km | 92 Wh | 3.7–6.1 km | 0.6 |
| Freesky Ranger Air M-540 | 153 km | 200 Wh | 8–13.3 km | 1.3 |
| Zeus Ebikes GT33 Cafe Racer | 160 km | 296 Wh | 11.8–19.7 km | 1.9 |
| Velotric Discover M VeloCore | 161 km | 374 Wh | 15–24.9 km | 2.3 |
| Velotric Tempo | 153 km | 374 Wh | 15–24.9 km | 2.4 |
| Velotric GoMad Utility | 241 km | 624 Wh | 25–41.6 km | 2.6 |
| Engwe P275 Pro | 260 km | 691 Wh | 27.6–46.1 km | 2.7 |
| Engwe P275 ST | 260 km | 691 Wh | 27.6–46.1 km | 2.7 |
| Freesky Cheetah MT-380 | 320 km | 880 Wh | 35.2–58.7 km | 2.8 |
| Tesway X5 Pro Cargo | 320 km | 880 Wh | 35.2–58.7 km | 2.8 |
Full disclosure, printed rather than footnoted: row three is sold by a store owned by this magazine’s publisher. The script does not know that, and we did not teach it. Every brand on the index runs through the same arithmetic — that is the point of having an index.
Some of these rows are almost certainly spec-sheet errors — a 92 Wh battery on a 1000-watt machine reads like a mistyped 920. But an error on the spec sheet is still the spec sheet, and a buyer comparing listings has no way to know which of the two published numbers to disbelieve.
Where “up to” comes from
None of this requires anyone to lie, which is what makes it durable. There is no standardized, mandated range test for e-bikes in North America — no equivalent of the fuel-economy cycles cars must publish. So “up to” is produced under the friendliest conditions the fine print can carry: the lowest assist setting, a flat course, a warm day, a light rider, no wind, no stops. Each choice is defensible. Stacked, they describe a ride nobody takes. The claim is physics performed in costume — real numbers, generated under conditions selected precisely because they will never recur on your street.
That is why this story’s headline names the practice and not a company. The lie is structural: an industry-wide agreement to measure the same machine in a way the road never will.
What to do with a range claim
The good news is that the defence takes ten seconds and one division.
Find the watt-hours and divide by 20. That is your honest midline. Divide by 15 if your rides are flat and gentle; by 25 if your city has hills, winters or cargo. If a 720 Wh commuter claims 120 km, you now know to plan around 36.
Treat a missing battery spec as information. Fifty machines on this index publish a range claim with no battery capacity at all. A claim that cannot be audited is asking for trust it declined to earn.
Compare claims only against claims, never against your commute. Between two machines measured in the same costume, the ratio still says something. Against your actual Tuesday, only the watt-hours do.
Every machine in the database carries its published battery capacity where it publishes one, so the division above is always available to you. The boards rank machines on price and spec — the numbers the makers cannot dress up. The claims will keep being printed. The arithmetic is now printed too.