How Far Can an E-Bike Really Go?

Heads up: we earn a commission if you buy through the links on this page, at no extra cost to you. It never decides what makes the list — here’s how that works.

The honest answer, up front: we don’t know, and neither does anyone else who hasn’t ridden the specific bike with your body on it up your specific hill. We have no test bikes, no dynamometer and no published dataset for any of these machines. What we can do is take the sellers’ claims apart with arithmetic, which turns out to be more revealing than another opinion.

The claims in Amazon’s e-bike catalogue run from 40 miles to 105. That’s a two-and-a-half-fold spread, and it’s the number people shop on. Here’s what happens when you check it.

The tell: everyone claims the same efficiency

Range isn’t a property of a bike; it’s capacity divided by consumption. So take each seller’s own two numbers — the battery and the claimed range — and reverse the division. That gives you the watt-hours per mile the claim assumes.

Heybike’s Mars implies 8.9 Wh/mile. The Ranger, 10.0. The Cityscape, 9.4. Vivi, 10.0. Jasion, 9.0. Engwe, 9.9. Hiboy, 10.0. Cyrusher, 12.8.

Look at what those bikes actually are. A 20-inch folding fat-tyre bike with a 1800W peak motor. A 26-inch commuter with no published motor at all. A 27.5-inch hardtail. A dual-suspension 52V machine. These differ enormously in mass, rolling resistance, frontal area and drivetrain. They do not have the same efficiency. It is not physically plausible. And yet seven of the eight land between 8.9 and 10.0 watt-hours per mile.

That uniformity is not five or six companies independently measuring their bikes and coincidentally agreeing. It’s a convention — a number the category has settled on, divided into whatever battery a bike happens to have, and printed on the box. We can’t prove intent and we’re not going to try. We can point out that if these were measurements, they’d disagree with each other, and they don’t.

The exception proves it nicely: the Cyrusher, at 12.8 Wh/mile, makes the leastflattering claim in the table relative to its battery. It has by far the biggest pack and it claims proportionally less from it than everyone else. That’s what an outlier looks like — and it’s the only figure here that behaves like it came from somewhere.

What the claims quietly omit

Not one of these listings publishes a single condition under which its range figure was produced. Not one. Here’s what would need to be stated for a range number to mean anything, and what all twelve sellers leave out:

Rider weight.Unstated everywhere. The federal statute that defines a low-speed electric bicycle specifies a 170-pound rider for its speed test — that’s what a real testing condition looks like. No range claim in this catalogue names one.

Assist level.Unstated everywhere. Every one of these bikes has multiple assist settings, and the difference between the lowest and the highest is easily a factor of two in consumption. The claim was made on one of them. It wasn’t the top one.

Terrain and wind.Unstated everywhere. “PAS” on the isinwheel listing at least admits pedal assist was involved — meaning the rider was contributing power, which is a polite way of saying part of that 105 miles is your legs.

The bike’s own weight. Unstated everywhere, and we keep saying it because it keeps mattering: zero of twelve listings publish a weight.On a hill, mass is energy. A range claim from a bike that won’t tell you its mass is a claim with a missing variable.

So how should you estimate your own range?

You can’t, precisely, and anyone selling you a calculator that says otherwise is selling confidence. But you can reason, and reasoning gets you close enough to buy sensibly. Three steps.

One: start from capacity, not from the claim. The battery is the only fact. Get it into watt-hours — volts × amp-hours if you have to — and treat that as your fuel tank.

Two: assume you are worse than the claim implies.Not because the sellers are lying, but because every unstated condition in a marketing figure is chosen favourably. If a listing implies 9 Wh/mile and you plan on twice that, you’ll have built in room for being heavier than the test rider, using more assist than the test rider, and riding into actual wind. Halve every claim in the table above and the bikes still cover a normal commute several times over.

Three: check whether the answer even matters.This is the step people skip. A commute is a known, repeated distance that ends at a plug. If your round trip is 12 miles, the difference between a 40-mile claim and a 105-mile claim is irrelevant to you — you are buying an argument you’ll never have. See the commuter page for why the biggest battery is usually the wrong purchase.

What actually drains the battery

We can’t give you measured numbers for these bikes. We can give you the mechanisms, which aren’t in dispute and let you reason about your own riding.

Speed is the big one, and it isn’t close. Aerodynamic drag rises with the square of speed, and the power to overcome it with the cube. A bike held at 28 mph is doing something categorically more expensive than the same bike at 15 mph. This is why the single most effective range extender ever invented is the assist button, pressed downwards. Five bikes in this catalogue claim 28 mph; none of them claims its range at 28 mph, because nobody would like the answer.

Hills, in proportion to total weight.Lifting you and the bike costs energy directly proportional to mass. On the flat, weight barely registers; on a climb it’s the whole story. Which is why an unpublished bike weight matters more to a hilly rider than to a flat one.

Cold. Lithium cells deliver less usable energy when cold. Your winter range is genuinely lower than your summer range and there is nothing wrong with your bike.

Stopping and starting.Every acceleration from rest costs energy you don’t get back — there’s no regenerative braking published on any bike in this catalogue. A city commute with twenty junctions is harder work than an open road at the same average speed.

Age.The pack you have in year three isn’t the pack you bought. Not one seller here publishes a cycle-life figure, so every range number on this page — theirs and ours — is a day-one figure.

The one number we’d trust, and nobody publishes it

If a seller wanted to be genuinely useful, it wouldn’t publish a range at all. It would publish watt-hours per mile at a stated speed, with a stated rider weight, on stated terrain — and then you could compute your own range in one division and be right. That is how the car industry ended up with EPA figures, and even those are contested.

Nobody in this catalogue does it. Nobody in this category does it. So the most honest thing we can put on this page is the table above: the sellers’ own numbers, divided by each other, showing you exactly what each claim assumes — and letting you notice that they all assume the same thing.