Most people earn below average.

There are two everyday “averages” and they answer different questions. The mean is the one you were taught at school: add up every value and divide by how many there are. The median is simpler than it sounds — line everyone up from smallest to largest and point at the person in the exact middle. Half are below them, half above.

For a tidy, symmetric set of numbers — the heights of a class, say — the mean and the median land in almost the same place, and it does not matter which you quote. Money is not tidy or symmetric, and that is where the trouble starts.

Half of everyone sits below the median — that is what “middle” means. It only sounds alarming when we expect “average” to mean “typical”, and it doesn’t.

Incomes, wealth, house prices and returns are right-skewed: most people are bunched together at modest amounts, with a long, thin tail of the very rich stretching far out to one side. The mean is a balance point, so it is exquisitely sensitive to that tail. Move one person in our room of thirty from £50,000 to £50 million and the mean leaps by well over a million pounds — while the median does not budge a penny. You watched it happen above.

The famous version: when Bill Gates walks into a small bar, the average wealth of the drinkers rockets into the billions, and on paper everyone is a millionaire. Nobody’s actual wallet has changed. The average has simply stopped describing anyone in the room.

This is why “the average UK salary is about £35,000” and “average household wealth is in the hundreds of thousands” can be true and still mislead: the median salary and the median wealth are noticeably lower, because the means are inflated by the top few per cent. “Most people earn below average” is not a riddle — for skewed data it is simply how things are.

Once you can see it, the swap is everywhere a few big values hide in the data:

Pay and wealth. Statistics agencies report both the mean and the median; a headline can pick whichever flatters its story. “Average pay rose to £X” sounds like everyone’s pay rose, even when the median — the typical worker — barely moved and the mean was lifted by bonuses at the top.

House prices. A handful of mansions pull the average price well above the median, so the “average home” can cost far more than the home a typical buyer actually faces.

Investment returns. A fund’s quoted average annual return can sit comfortably above what its typical investor earned, because a few extreme years — and the order they come in — distort the simple average. That is its own illusion, the volatility drag.

“The average customer.” When a few “whales” spend enormously, the mean spend per customer describes none of them, and a business that plans for the average plans for a person who does not exist.

Ask which average it is. If a figure is only ever called “the average”, assume it is the mean, and ask for the median. Reputable sources give both.

For anything money-shaped, prefer the median. Income, wealth, prices, returns, waiting times and durations are all skewed; the median is the trustworthy “typical” for every one of them.

Treat a large mean–median gap as a skew alarm. The further apart they are, the harder a few extreme values are pulling the mean — and the more misleading the word “average” becomes.

Be wary of “X% are above average”. For skewed data, far fewer than half can be above the mean, so “most are below average” is ordinary, not a scandal — and “above average” is a lower bar than it sounds.

When you can, look at the shape of the whole distribution, not a single number. One number can only ever tell one story, and a skewed world has at least two.