Method
How we choose
Every pick here is built from the manufacturer's published specs, checked against real fit and weight arithmetic, and shown with our working. You get the measurement that decides the purchase, the top pick for your child's stage, and a live price to act on.
Last updated November 4, 2026
What we actually do
Step 01
Verify the product is real and currently sold
Every one of the 72 products in our registry was checked against the manufacturer's own live product page. If a model is discontinued, it comes off the site rather than being left to rot at the top of a roundup.
Step 02
Read the published specs from the source
Weight, minimum and maximum saddle height, inseam range, brake type, gearing and frame material come from the brand's own spec page, and every product card links to the page we read. Where a manufacturer does not publish a figure, the cell says so, and the pick is made on the figures that are published.
Step 03
Check fit against inseam, not age
We compare each bike's minimum saddle height against the inseam ranges that actually correspond to it, and we flag where brands disagree with each other. That disagreement is usually the most useful thing on the page.
Step 04
Do the weight arithmetic
Bike weight is meaningless in isolation, so we express it as a percentage of the child's body weight using the CDC's published 50th-percentile weight-for-age data, and scale it to an adult equivalent so the number means something. Both inputs are linked; neither is estimated.
Step 05
Rank buyer-first
The top pick is the one we would buy for our own kid at that stage, not the one that pays best. Where the right answer is "buy used" or "do not spend this much", that is what the page says.
What you get on every page
Every number linked to its source. Each product card carries the manufacturer's own spec page, so the weight and the saddle range you are deciding on are one click from being checked.
A clear top pick, and the runner-up worth knowing about. Every roundup names the bike we would buy at that stage and the cheaper one that does most of the same job, with the measurement that separates them stated plainly.
A live price on the button. Prices come from a daily data refresh with the date attached, so the number you see is the number you pay — never a figure typed in a year ago.
The scale we score on
Every product in our registry carries a score out of 10, and the scale runs from 8.4 to 9.8. That narrow band is the point: this is not a list of everything on sale, it is a list of the 72 products we are willing to put in front of a parent. A product that would score below 8.4 is not rescaled to look acceptable — it comes off the site and a stronger, in-stock alternative takes the slot.
| Score | What it means |
|---|---|
| 9.5–9.8 | Category-defining — the bike or helmet we would buy first at this stage |
| 9.0–9.4 | Outstanding — a pick we are happy to recommend without qualification |
| 8.4–8.9 | Excellent and strongly recommended — right for a specific buyer or budget |
| Below 8.4 | Not recommended on this site, so it is not listed here at all |
Scores are editorial and ours. They are not Amazon star ratings, they carry no review count, and they are never published as review structured data.
How the number is worked out
The six criteria below, and the weight we give each of them, have not changed. What they have always produced is an order — the ranked list on every roundup, comparison and review here. The score expresses that order on a 10-point scale, using one site-wide mapping: we take each product's strongest position anywhere on the site, its average position across every list it appears in, and how many lists it earns a place on at all, then stretch the result linearly across 8.4–9.8. The strongest all-round picks land at the top of that range; the narrowest-use picks land at the bottom of it.
Two things follow, and both are deliberate. A product's score is the same on every page it appears on, because the score belongs to the product, not the page. The score never reorders a page, because each page answers a narrower question than the score does — on "best balance bike for a five-year-old" the bike that fits a five-year-old best is ranked first, and a stronger all-round bike further down the page can carry a slightly higher score. Where that happens, the page tells you which figure separated them.
We do not claim to have ridden every bike here. Every card says exactly what the research basis was and links the spec page it came from, and we would rather publish that than imply a test we did not run.
The criteria, in order
1. Fit — minimum saddle height against inseam
It decides everything else. A bike your child cannot put their feet down on does not get ridden, and an unridden bike is the most expensive outcome available.
2. Weight relative to the child
Under about a third of body weight. It governs starting from a standstill, lifting the bike after a fall, and how far they get before they have had enough.
3. Brakes appropriate to the stage
Hand levers a small hand can actually squeeze. We prefer a freewheel to a coaster brake, and we treat single-lever systems as a legitimate third option rather than a compromise.
4. Adjustability — how long it keeps fitting
Saddle travel is the difference between one purchase and two. We publish it wherever the manufacturer does.
5. Assembly and serviceability
Whether a parent can maintain it with a multi-tool and a pump, and whether the parts are standard enough to replace.
6. What the family actually gets for the money
Including resale value, which is a real and routinely ignored part of what a kids bike costs.
How we make money
Affiliate commission on some outbound links, disclosed above the first link on every commercial page. The clearest evidence that it does not drive the rankings: Guardian and Prevelo appear at the top of several roundups here, are not sold on Amazon, carry no buy button and earn us nothing at all. Full detail on our affiliate disclosure.
How to tell us we are wrong
Email us through the contact page. If a figure is wrong we will correct it and note the correction with a date on the page, so you can see exactly what changed. That policy is written down on our editorial policy.