The Number That Makes Crypto Look Better Than It Is

Cryptocurrency metrics displaying exaggerated performance data

Imagine a room with a hundred people in it. Ninety-nine of them reach into their pockets and hand £100 to the one person standing at the front. That person walks out with £9,900 and everyone else goes home skint. Now imagine someone runs the numbers the next morning and announces: "Great news! The average participant gained £98!"

Technically true. Utterly useless.

Cryptocurrency metrics displaying exaggerated performance data — detail

This is, more or less, what happens every time someone quotes you the average return on a crypto investment.

Why the Average Is Lying (Without Technically Lying)

The number people throw around when talking about crypto gains is almost always the mean – you add up everything and divide by the number of people. That works fine when the outcomes are bunched together, like heights in a classroom or scores on a test. But it falls apart completely when the distribution is wildly skewed, with a tiny cluster of colossal wins and a much larger crowd of people who broke even or lost.

Crypto returns are about as skewed as it gets. Bitcoin's most spectacular price rises happened in a handful of short, violent bursts – a few months in 2013, a stretch in 2017, another in 2020-2021. If you happened to be holding during those windows and, crucially, actually sold, you may have done extraordinarily well. The mean return across all buyers across all time looks remarkable partly because of those moments. But the median – the return that the person exactly in the middle of the queue actually experienced – is a very different, much quieter number.

The Difference Between Mean and Median (And Why It Matters Here)

The mean is dragged upward by the outliers at the top end. The median ignores them entirely and just tells you what a typical person got. In a power-law distribution – which is the rough shape of crypto returns – those two numbers can be miles apart.

Back to the room with a hundred people. One person gains £100,000. Everyone else loses £100 each. Mean return: roughly £990 profit per person. Median return: minus £100. The average looks extraordinary; the typical experience was a loss. Both numbers are arithmetically correct. Only one of them tells you what happened to most people.

The mathematics here isn't an argument against buying crypto. It is an argument against trusting headline return figures without asking which measure of "average" is being used and who, exactly, was in the room when those returns were made. If you want to go further and understand the practicalities of how buying works, there are straightforward guides at places that explain the process without dressing it up – Coins.co.uk is one example worth a look for UK buyers.

What the Person in the Middle Experiences

Most buyers arrive after the headline-grabbing run has already happened. They buy somewhere in the middle of a cycle and either sell too early, too late, or hold through a correction that wipes most of the gain back out. They didn't make a mistake, exactly. They just experienced what the median always experiences in a skewed distribution: something considerably less exciting than the mean.

The number everyone quotes when they say crypto has delivered extraordinary returns is, by construction, a number built mainly by people who aren't you – a handful of early holders, a handful of traders who got the timing right, and a mathematical formula that adds them all in regardless. Worth knowing before the room fills up again.

Questions this raises

  • What is the median return on a crypto investment?
  • Why do people quote the mean instead of the median?
  • How many crypto investors actually lose money overall?
  • Does the same averaging problem apply to stock market returns?