Angie · learn

Mean reversion

The tendency of a stretched price to drift back toward its own average. Angie measures how often that actually happened for a name rather than assuming it will.

Updated 2026-09-03

Mean reversion is the idea that a price which has moved far from its average tends to come back toward it. It is the intellectual opposite of trend following, and both are profitable in the right conditions and ruinous in the wrong ones.

When it holds and when it does not

It holds where there is a genuine anchor. Spreads between closely related instruments revert because arbitrage forces them to. Volatility reverts because panic ends. What does not reliably revert is the price of a single company, because a company's fair value moves, and a price that has fallen a long way may simply be repricing a business that got worse.

This is the failure mode behind the phrase catching a falling knife. A stock down 60 percent has an extreme z-score, and the extreme reading is generated by the same news that permanently lowered what the company is worth. The measurement is correct and the inference from it is wrong.

Angie is careful to state mean-reversion statistics as measured rather than as predicted. A note saying a stock has historically returned toward its average from this level is a description of history. It is not a claim about what happens next, and the distinction is the whole difference between an observation and a forecast.

See also

  • Z-score — How many standard deviations today's price sits from its recent average. Two sigma below means a move this far down happened on roughly one day in forty.
  • Bollinger bands — A moving average with bands drawn two standard deviations above and below it. Price at a band is unusual relative to its own recent range, not relative to any other stock.
  • RSI — A 0 to 100 gauge of how one-sided recent moves have been. Conventionally below 30 is oversold and above 70 overbought, though a strong trend can hold an extreme for weeks.
  • Backtest — Running a set of rules over history to see what they would have done. It shows what did happen, never what will, and it flatters any rule tuned on the same data.

← See this one in the deck

Educational content, not investment advice. Angie explains how a valuation is built so you can judge it yourself. What you do with that is your call.

IsiofiaIsiofia