What it is
Mean-reversion strategies assume that a price which has moved unusually far from some reference will tend to move back toward it. The reference can be a moving average, a session VWAP, an opening price, or, in relative-value form, the fitted spread between two related instruments.
How it works
Implementations usually standardise the deviation. Compute a z-score of price against its rolling mean and standard deviation, enter when the score exceeds a threshold, and exit at the mean or after a fixed holding period. Bollinger band and RSI extremes are the retail expression of the same idea. Short-horizon reversal is one of the more robustly documented effects in equities, particularly over overnight and one-to-five-day windows.
How traders use it
The payoff profile is distinctive and needs to be understood before the strategy is traded: a high win rate, small average wins, and occasional large losses when the deviation keeps extending. Equity curves look wonderful right up until they do not, which makes this family easy to over-trust and easy to over-leverage.
Where it breaks down
The essential requirement is a regime filter and a hard stop. Mean reversion works in balanced markets and fails badly in trends and in repricing events, where the very signal that triggers entry, an extreme move, is the beginning of new information being absorbed rather than an overreaction. The belief that price has to come back is not a risk-management plan: the reference itself moves, and nothing guarantees a return before the capital runs out.
Educational reference. This entry describes how a concept is defined and used. It is not investment advice, not a recommendation, and not a signal. Any rule you build from it should be tested with realistic costs before it is traded, and no historical result guarantees a future one.