Ten thousand steps was invented by a pedometer company

The figure comes from a Japanese pedometer launched before the 1964 Tokyo Olympics, called the manpo-kei — literally “ten-thousand-step meter”. It was chosen because it was a round number and the character for 10,000 looks a little like a walking person. That is the entire origin story.

It has since been studied properly, which is the interesting part, because the real numbers are more encouraging than the invented one.

What the evidence actually shows

Mortality benefit starts appearing at around 2,500 steps a day compared with near-total inactivity. It climbs steeply between 2,500 and roughly 7,500. Above about 8,000 the curve flattens; above 10,000 it is close to flat.

The steepest part of that curve is at the bottom. Which means the person who gains most from walking more is the person currently walking least — and the gain from 2,000 to 5,000 steps is considerably larger than the gain from 8,000 to 12,000.

If you are currently doing very little, you do not need to find ten thousand steps. You need to find three thousand.

Intensity does more than total

Cadence matters independently of volume. Around a hundred steps a minute is roughly the threshold for moderate intensity in most adults — brisk enough that you could still talk but would rather not sing.

Thirty minutes at that pace does more for cardiovascular risk than two hours of ambling, and it takes less of your day.

The part people skip

Walking does very little for muscle strength, and from around forty we lose muscle steadily unless we do something that loads it. Two sessions a week of anything resistance-based — body weight is fine — protects the thing that determines whether you are independent at eighty.

Steps are the easy half. The resistance work is the half that gets dropped, and it is the half I spend most of my time arguing for.