How we review

The sentence that matters most

We have not physically ridden, weighed, timed or measured any of the 22 bikes on this site. Nobody here has thrown a leg over a V9-G60 or taken a FLASH 2.0 up a hill. Every review says so, because a reader who does not know it will read our work as something it is not.

What we do instead is read what the merchants publish, transcribe it with a date, do the arithmetic the specification sheets imply but rarely show, and say what we think it means. That is a real method with real limits, and this page is where both are written down.

Where the facts come from

Every figure on this site comes from one of two places, and we keep them apart.

• The merchants' own published material. The product page, the specification table, the imagery and the manuals where those exist. All four brands run on Shopify, so alongside the visible page there is a structured product record — the Shopify product JSON — carrying variants, options, shipping weights and prices. We read both, because they do not always agree, and a disagreement is worth reporting.
• Arithmetic on those figures. Volts times amp-hours gives watt-hours. Price divided by watt-hours gives price per watt-hour. Watt-hours divided by an assumed consumption gives a range estimate. None of that is measurement; all of it is shown.

Each brand's data is transcribed into a dated file. The nine oldest records — the AMYET, VTUVIA and EUNORAU listings we started with — were last checked on 21 and 22 August 2026; the ten most recent additions, and all six Velotric records, on 27 August 2026. When a figure changes on the merchant's side, the answer is to re-read the listing and re-date it, not to leave the old number looking authoritative.

Merchant claim and editorial note stay separate

There are three kinds of statement on this site, and we keep them apart visually and grammatically.

• Merchant claim. Something the brand publishes, quoted in its own wording and always attributed: "AMYET lists the V9-G60 at up to 32 MPH", never "it does 32 MPH". In our product data this is a field called merchantClaim, and copy is never allowed to promote it into a statement of fact.
• Arithmetic. Something that follows from a published figure by calculation. 48V × 20Ah = 960Wh is arithmetic, not a claim, and we show the working rather than presenting the result as measured.
• Editorial note. Our reading of what the specification means in practice — what a 55 Nm figure feels like against a 220 Nm one, why a 110 lb bike changes how you store it. This is opinion, labelled as ours, with the reasoning visible so you can disagree with it.

A claim and our note about it sit in different blocks, in different type. If you cannot tell which you are reading, that is a defect — tell us through the contact page.

Where a brand publishes nothing, we print "Not published"

This is the rule that does the most work.

If a specification is missing from a merchant's listing, the entry on our page reads "Not published". We do not substitute a plausible figure, carry a number across from a sibling model, or quietly drop the row so the table looks complete. A blank in a specification sheet is information about the product, and hiding it would be the easiest way to mislead a reader while appearing thorough.

The gaps we found

Documenting the gaps is part of the job. Across the 22 bikes, these are the ones that matter most.

• AMYET publishes no motor torque figure for any of its five models — not the V9-G60, not the EB26, not the S8, not the K10, not the Ares 2. Torque in newton-metres predicts hill climbing far better than watts do, and sixteen of the seventeen non-AMYET bikes here publish one, which makes the absence conspicuous rather than unlucky.
• AMYET publishes no assist sensor type either. A cadence sensor switches assist on when the cranks turn; a torque sensor scales it to how hard you push. They produce completely different bikes, and AMYET does not say which is fitted.
• AMYET publishes no bare bike weight for any model. The EB26 listing carries a 30 kg shipping figure in its variant data, but that includes packaging and is not a bike weight, so we report it as what it is.
• EUNORAU publishes no payload figure for the FLASH 2.0. Every other bike here publishes one, and the 1,000 lb rating quoted for that bike's rear shock is a component specification, not the bike's limit.
• EUNORAU publishes no assist sensor type for the FLASH 2.0, an unusual omission on a mid-drive, where the sensor shapes the whole character of the ride.
• Several bikes state no certification at all. The AMYET S8, the VTUVIA Zeal LT7, the VTUVIA SX20 and the EUNORAU META26 1.0 carry no UL standard on their listings as we read them.

None of that is an accusation: an unpublished figure is not a bad one. But it is the thing you cannot check, and a buyer is entitled to know which parts of the sheet are missing before spending $639.99 or $2,999.

Certification is reported per bike, never inferred

Certification is where a helpful-seeming inference would be most dangerous, so we make none.

AMYET states UL 2849 for the V9-G60 and the EB26; the S8 listing carries no such statement, so ours says it carries none. VTUVIA states both UL 2271 and UL 2849 for the SF20H; the Zeal LT7 and SX20 listings state no UL standard, so neither do we. EUNORAU states UL certification for the SPECTER-S 3.0 without naming the standard, and we report exactly that — UL 2271 covers the battery pack, UL 2849 covers battery, charger, controller and motor together, and they are not interchangeable.

A certification on one model in a range tells you nothing about the model beside it. If certification is the reason you are buying, ask the brand in writing and treat their answer as the answer, rather than anything on this site.

What we never do

We do not write "we tested this bike", because we did not. We publish no star ratings, no scores out of ten and no numerical verdicts, because a score implies a measurement we have not taken. We do not publish customer testimonials, review counts or awards. We do not invent stock levels, discount codes or sales figures. We do not fabricate a certification, or describe a bike as certified because a sibling model is.

If any page here ever appears to claim a physical test result, that is an error, and we want to be told.

Prices, and what happens when a check goes stale

Every price carries the date it was read from the merchant's listing, and every price in prose matches the figure in the underlying data file. Prices were last checked on 27 August 2026.

Each price also carries a shelf life. Once a check is older than that window, we stop showing the number and show a "Check current price" link to the merchant instead. A stale price is worse than no price: it looks precise, it reads as current, and it is wrong in a way the reader cannot detect.

We do not publish countdown timers, invented "was" prices or claims about how long an offer will last. Where a merchant publishes a recommended price alongside a sale price, we show both.

What we look at, bike by bike

Every model gets the same list in the same order: rated and peak motor output; torque; battery voltage, amp-hours and the resulting watt-hours; whether a second battery is offered; claimed range; top speed and its legal implications; brake type; sensor type; drivetrain and gearing; tires and wheel diameter; suspension; frame material and sizes; bike weight; maximum payload; display and lighting; ingress protection; certification; warranty; and price, price history and price per watt-hour.

Where a merchant publishes conflicting figures across their own pages — the AMYET S8 appears at up to 32 MPH in one place and up to 35 MPH in another — we show both rather than choosing the flattering one.

What this method cannot tell you

Plenty, and pretending otherwise would undo the rest of this page.

A specification sheet will not tell you how a fork feels over a root, whether a display is legible in direct sun, how loud a motor is at 20 MPH, how a bike behaves at the top of its payload rating, or what customer service is like when something breaks in month eighteen. Those things require riding the bike and living with it.

What specification analysis is good at is catching the things that cannot be true, and the things left out. A range claim the battery cannot physically support. A brake specification that does not match the mass and the claimed speed. A price that is worse per watt-hour than the bike beside it in the same catalogue. A sheet with four blanks where the competitor's has none. That work is worth doing before you spend the money.

Independence

Ebike Rider earns affiliate commission on some outbound links — the affiliate disclosure has the detail. No merchant pays for coverage, for a ranking position, or to have a criticism removed. We do not receive review units, and no brand sees a page before it is published.

The ordering is the test we would apply to anyone making this claim. Our overall ranking does not run from cheapest to most expensive, or from lowest commission to highest; a $689 bike and a $2,999 bike are recommended to different people for reasons written down each time.

Corrections

Specifications, prices and availability change without notice, and the merchant's own page is always the authority. If you find an error — a spec we have misread, a claim we have overstated, a figure that has moved — email hello@ebikerider.us.

We correct errors in place and note substantive corrections at the foot of the page, with the date. Reports that include a link to the listing are dealt with fastest, because they let us re-read the source rather than guess.