Why is it so hard to buy when everyone else is selling?
Copying other people is usually excellent policy. If a crowd runs out of a building, running too is not a bias — it is survival. Markets are the rare arena where the heuristic inverts, because here the copying itself moves the evidence.
The feedback loop
In an ordinary domain, other people's behaviour is independent evidence about the world. In markets it is not. When many buy, the price rises; the rising price is then read as confirmation that buying was correct, which attracts more buyers. The crowd is producing the very data it is treating as proof. "Everyone is doing it" is not information — it is a correlation of ignorance, priced in real time.
Two forces keep the loop turning. Social proof answers uncertainty with imitation. FOMO — the fear of missing out — is loss aversion pointed at a gain you have not made: watching others profit registers as a loss, and the pain drives action long after analysis would have stopped.
The arithmetic of joining late
This is the part that surprises people, because the asset does not have to fail for the investor to.
An investor commits €5,000 to a theme early, at a price of 100 — 50 units. The theme doubles. Vindicated and impatient, they add €20,000 near 200 — 100 more units.
- Total invested: €25,000. Total units: 150. Average cost: about 167.
- The price then falls back to 120 — still 20% above where they first bought.
- Position value: 150 × 120 = €18,000. Loss: €7,000, about 28%.
The asset is up. The investor is down 28%. Nothing failed except the sizing over time: the small money was early and the big money was late. This is the behaviour gap of the next unit in miniature, and it is why "but the theme was right" is such a poor defence.
Why the bottom is worse than the top
Herding on the way down is more expensive still, because the crowd is loudest exactly when prices are lowest. Selling into a decline converts a fluctuation into a permanent loss, and the decision feels overwhelmingly prudent at the time — everyone agrees, the news confirms it, and the price is falling as you read.
The professional's counter is not contrarianism for its own sake. It is to notice that crowd behaviour is a fact about the crowd, not about the asset, and to require an independent reason before acting. If you cannot state a reason that would still make sense with the crowd removed, you do not have a reason.
Unit checkpoint ahead
Loss aversion, overconfidence, recency, anchoring, herding — five mechanisms, one common shape: a feeling arrives, and the portfolio pays for it. Unit 2 adds the sixth, then measures the whole bill in a single number.
Try it now
- Take your largest position and write the reason you own it in one sentence. Now delete every clause that refers to other people, the price trend, or what is popular. Is a reason left?
- Reconstruct one of your positions the way this lesson did: how much money went in at which price? Compute your average cost, then drop a Level at that price on the chart below and look at where it sits relative to the path. Did your money arrive early or late?
- Write down what you would need to see before adding to a position that has already doubled. Decide it now, while nothing has doubled.