
What Is Mean Reversion Trading? Definition & How It Works
Mean reversion trading is the practice of betting that a price, spread, or other market measure will move back toward its statistical average after it has drifted too far away from that average. The idea rests on a simple observation: most markets don't move in a straight line forever. They oscillate, and when a price pushes far above or below its typical level, the odds increase that the next move is a pullback rather than a continuation. A mean reversion trader isn't predicting direction in the usual sense — they're predicting that an unusually large deviation will shrink.
This only works as a premise, not a guarantee. The "mean" in question is almost always a specific, calculated reference point — a moving average over some lookback period, a historical price range, or the average spread between two related instruments. Deviation is measured against that reference, and the trade is built on the expectation that the gap closes. When a market genuinely changes character and starts trending, the same deviation that looked stretched can keep stretching, which is the central risk the rest of this explainer comes back to.
The Statistical Logic and Indicators Behind Mean Reversion
The reversion-to-the-mean premise
Mean reversion borrows its logic from statistics more than from forecasting. If you plot a market's price over time and it spends most of its life oscillating around a central value, that central value acts like a magnet: moves away from it tend to lose momentum and snap back, because the conditions that pushed price away — a burst of buying, a panic sell-off, a short-term imbalance — are usually temporary. The premise holds up best in markets that are genuinely range-bound, where supply and demand keep resetting around a stable equilibrium rather than shifting to a new one.
Measuring deviation with z-score and standard deviation
To act on this premise, a trader needs a way to measure "how far is too far." Standard deviation does that job: it quantifies how spread out recent prices have been around their average, giving a sense of what a normal fluctuation looks like for that specific market. A z-score then expresses the current price's distance from the mean in units of that standard deviation. A reading close to zero means price is sitting near its average; a reading well away from zero means the market has moved an unusual distance from where it typically trades. Mean reversion setups are built around watching that distance stretch and then start to contract.
Reading overbought/oversold signals with Bollinger Bands, RSI, and moving averages
Three tools do most of the practical work of spotting this stretch:
- Moving averages smooth out short-term noise to show the underlying "center" price is reverting toward.
- Bollinger Bands plot a band above and below a moving average, sized by the market's own standard deviation, so the bands widen when volatility rises and narrow when it falls; price pressing against or through a band is read as a statistically stretched move.
- RSI (Relative Strength Index) measures the speed and size of recent price changes on a bounded scale, so low readings suggest a market has sold off hard enough to be considered oversold, and high readings suggest it has rallied hard enough to be considered overbought.
None of these three signals the same thing in isolation as cleanly as they do together — a price at the edge of a Bollinger Band with RSI also at an extreme is a stronger case for reversion than either one alone.
A Concrete Example: Spotting an Overbought-to-Oversold Reversion
Picture a stock that has spent the past several weeks trading in a fairly contained band, drifting between roughly 48 and 52 on its chart with no clear directional push either way. Its moving average sits near 50, and its Bollinger Bands track a few points above and below that line, widening slightly on busier days and narrowing on quiet ones.
Then a piece of short-lived news sparks a rally. The stock jumps to 58 in two sessions — well outside its usual band — and RSI climbs into territory that marks it as sharply overbought. A mean reversion trader watching this setup isn't asking "will this stock keep rising." They're asking whether the move looks like a genuine shift in the stock's trading range or a temporary overextension. With the price sitting far above its moving average, the band stretched, and momentum reading extreme, the statistical case points toward a pullback rather than continuation.
Over the following days, the stock drifts back down, first to 54, then toward the 50 level where its moving average sits. RSI eases out of overbought territory as it does. This is the shape of a textbook mean reversion trade: price deviates, indicators flag the deviation as unusual, and price retraces toward the average that defined the range in the first place. The same logic runs in reverse during a sharp sell-off — price undershoots, RSI drops into oversold territory, and the bounce back toward the mean is what the trader is positioned for.
Mean Reversion vs Trend Following: What Sets Them Apart
Trend following starts from the opposite assumption: once a market is moving decisively in one direction, that move is more likely to continue than to reverse, at least until clear evidence says otherwise. Where a mean reversion trader sees an overbought reading as a sell signal, a trend follower sees the same strength as confirmation to stay with the move.
The real dividing line between the two isn't the indicators — both approaches can reference moving averages, for instance — it's the market condition each one is built for. Mean reversion is suited to range-bound markets, where price repeatedly tests and retreats from the edges of an established band. Trend following is suited to trending markets, where each pullback is shallow and each new high or low extends the prior move rather than reversing it. The same RSI reading that signals "exhausted, likely to reverse" in a ranging market can simply mean "strong trend, likely to keep going" once that market has broken into a sustained directional move. Knowing which regime you're actually in matters more to the outcome than which indicator you've chosen.
Where Mean Reversion Breaks Down: Limits and Risks
The approach's core weakness is built into its core assumption: it only works while the market is actually range-bound. The moment a market transitions into a sustained trend, the same signals that used to mark a reliable turning point start marking the early stages of a much larger move instead. A stock that looks overbought at the top of its old range can keep climbing well past it if the range itself is being replaced by a new, higher one. A trader applying mean reversion logic in that environment ends up repeatedly betting against a move that keeps extending, taking a series of losses that each look, individually, like a reasonable bet on reversion.
False signals compound this problem. Oversold and overbought readings aren't binary triggers; a market can sit at an extreme reading for an extended stretch without reverting at all, especially when a trend is strengthening. Relying on a single indicator reading in isolation tends to produce more of these false starts than combining several, because any one measure can stay "stretched" far longer than intuition suggests. The specifics of how often this happens vary by market, timeframe, and the particular combination of indicators used, which is why mean reversion is generally treated as a tool suited to particular conditions rather than a strategy that holds up across every type of market.