Step Counts Against a Manual Count
Step counts set everybody's daily target and nobody has checked the number. We counted 42,000 steps by hand and compared four devices against the tally.
Tech Daily Review Desk
What actually changed
Step counts became the default measure of whether a person moved today — on watches, on phones, in insurance schemes and workplace challenges — without the accuracy of the underlying count ever being something most people saw a figure for.
A step is inferred from an accelerometer trace, which means the device is looking for a periodic bounce and deciding whether it is walking. Anything rhythmic can look like walking and anything smooth may not: a trolley absorbs the bounce, a pushchair damps it, and applauding does not. The algorithms are all tuned for the common case, and they diverge exactly where the common case stops.
What is at stake
Because targets are set on these numbers, and because the same walk can produce meaningfully different counts depending on which pocket the phone was in.
How we measured it
Two people, a hand tally counter each, 42,000 steps across eleven scenarios — pavement, treadmill, stairs, shopping with a trolley, pushing a pushchair, carrying a box, hands in pockets, and three sitting-still controls. Two watches, a ring and a phone in a front trouser pocket were worn simultaneously. The hand count is the reference; every figure is against it.
What we recorded
| Scenario | Watch A | Watch B | Ring | Phone in pocket |
|---|---|---|---|---|
| Pavement, arms swinging | +1% | +2% | −3% | −2% |
| Treadmill | +2% | +1% | −6% | −4% |
| Stairs up | −4% | −2% | −9% | −7% |
| Hands in pockets | −18% | −14% | −4% | −3% |
| Pushing a shopping trolley | −41% | −37% | −11% | −6% |
| Pushing a pushchair | −44% | −39% | −13% | −8% |
| Carrying a box, two hands | −52% | −48% | −14% | −7% |
| Sitting, applauding 200 times | +186 | +144 | +31 | +0 |
| Sitting in a car on a rough road | +212 | +97 | +18 | +41 |
What the figures say
For ordinary walking every device is within a few per cent, which is the good news and covers most of what most people do.
Anything that stops your arm swinging destroys a wrist count — a shopping trolley cost 41% and a box cost 52%, and neither is an unusual way to spend an hour.
The ring and the phone were far more robust to those cases and far less good at stairs, so no device won across the board.
False positives are real and consistently ignored: applauding added 186 steps to a watch that had not moved anywhere at all.
Where this could mislead you
Three things worth holding on to when you meet these figures somewhere else:
- A daily total from a wrist device systematically undercounts anybody who shops, pushes a pram or carries things for a living
- Phone and watch counts should never be added together, and several apps will happily do exactly that
- A change in your average may be a change in where you carry the device rather than in how much you walked
What we would do
Compare yourself against yourself and ignore the absolute number. Every device in this test tracked direction and change reliably even where it was 40% off in absolute terms, so a target of "more than last week" is sound and a target of exactly ten thousand is arithmetic about an estimate.
If your day involves a trolley, a pushchair or carrying things, use the phone in a pocket rather than the watch — it was three to seven times more accurate in exactly those scenarios. If your day involves stairs and free arms, the watch is the better instrument. Pick for your day rather than for the device.
Where we land
Accurate to a few per cent when your arms swing and out by up to half when they do not. Track the trend rather than the total, and use a pocket rather than a wrist if you push or carry things.
Behind this piece
Every product covered here was bought at retail and lived with for at least a fortnight before a word was written. Battery figures come from a scripted day rather than a manufacturer estimate, and sustained performance is measured warm — after half an hour of load — because that is the state a device spends most of its life in.
Where a firmware update changes a result, the piece is re-tested and the change noted rather than quietly corrected. No review unit arrives with conditions attached, and no score is agreed in advance with anybody.
Where to go next
If you are weighing this against the rest of the wearables we have covered, the category archive lists every review in date order. The closest comparisons we have published are The Best Wearables For Battery Life in 2026 and The Best Wearables For Beginners in 2026.
For the reference material behind these figures, the specification pages carry the measured numbers, and the common questions collect what people ask once something is in daily use.
