Every investor has watched a stock plummet and thought: surely it has to bounce back. Sometimes it does. Sometimes it keeps falling. The difference between these outcomes isn't luck—it's understanding when mean reversion actually applies.

Mean reversion is one of the most misunderstood concepts in finance. The basic intuition feels obvious: extreme moves tend to correct themselves. Buy low, sell high. What could go wrong? Quite a lot, actually. Entire portfolios have been destroyed by assuming prices must return to some historical average.

The reality is more nuanced. Mean reversion is a statistical property that applies under specific conditions—conditions that markets don't always provide. Understanding when prices genuinely tend to snap back, versus when an apparent extreme is actually the start of a new trend, separates informed analysis from wishful thinking.

Mean reversion isn't a law of physics. It's a statistical property that emerges when a time series is stationary—meaning its statistical properties remain constant over time. Temperature in a given location tends to be stationary: winter cold eventually gives way to summer warmth, and vice versa. The average holds.

Stock prices, however, are not stationary. They follow what statisticians call a random walk with drift. There's no fixed average they're obligated to return to. A stock at $100 has no more reason to return to $50 than to climb to $200. The price level itself carries no mean-reverting signal.

This is where many investors go wrong. They see a stock that was once $150 now trading at $75 and assume it's 'cheap.' But cheapness relative to past prices means nothing if the underlying value has genuinely changed. What can exhibit mean reversion are certain ratios and relative measures: valuations like price-to-earnings, spreads between related assets, or deviations from fundamental value.

The mathematical test for mean reversion involves measuring whether extreme observations tend to be followed by moves toward the center of the distribution. For truly stationary series, the autocorrelation of returns is negative—today's gains predict tomorrow's losses. For stock prices broadly, this autocorrelation is approximately zero. Mean reversion in price levels is largely a myth; mean reversion in valuations and spreads is often real.

Takeaway

Price levels don't mean-revert because they're not stationary. Focus instead on valuations, spreads, and ratios—these can exhibit genuine mean reversion because they have natural equilibrium points.

Even when mean reversion applies in theory, it doesn't apply uniformly across time. Markets shift between regimes—periods where different dynamics dominate. In range-bound regimes, mean reversion strategies thrive. In trending regimes, they get systematically destroyed.

The challenge is that regime shifts are only obvious in hindsight. A stock declining 30% could be mean-reverting from an overvalued peak, or it could be the first leg of a fundamental re-rating driven by deteriorating business conditions. Both look identical in the early stages.

Several signals help distinguish regimes. Volatility clustering often accompanies regime changes—sudden spikes in volatility frequently mark transitions. Correlation breakdown between previously linked assets suggests structural shifts. And volume patterns matter: mean reversion typically occurs on declining volume as panic exhausts itself, while trend continuation often features sustained or rising volume.

Fundamental context provides crucial information. A valuation stretched to extremes during stable business conditions differs entirely from one stretched by genuine earnings collapse. The former tends to revert; the latter may be finding a new equilibrium. The price action alone won't tell you which you're facing. You need to understand why the extreme occurred. Sentiment-driven extremes revert. Fundamental repricing persists.

Takeaway

Mean reversion isn't always active—markets alternate between range-bound and trending regimes. The key skill isn't trading reversion; it's recognizing which regime you're actually in.

Trading mean reversion profitably requires accepting an uncomfortable truth: you will be wrong sometimes, and when you're wrong, the losses can be severe. A mean-reversion strategy that's right 70% of the time can still lose money if the 30% of failures involve catastrophic losses.

Position sizing becomes paramount. Mean reversion trades should never represent concentrated bets, precisely because the strategy assumes you're fighting the current price direction. Scaling in—adding to positions as they move against you—only works if you're certain about the reversion thesis. More often, it compounds mistakes.

The most robust approach combines multiple confirmation signals. A stretched valuation alone isn't a trade. Add oversold technical indicators, sentiment extremes measured through surveys or options positioning, and fundamental stability—then you have something worth acting on. Each additional confirmation increases the probability that you've found genuine mean reversion rather than trend continuation in disguise.

Time limits provide essential discipline. Mean reversion should happen within a reasonable timeframe if the thesis is correct. If a 'stretched' condition persists for months, the market may be telling you that what looked extreme was actually the new normal. Set explicit deadlines for your thesis to play out. When they expire without confirmation, exit—regardless of whether you're at a profit or loss.

Takeaway

Mean reversion trading isn't about catching falling knives—it's about constructing positions with multiple confirmations, disciplined sizing, and predetermined exit criteria that protect you when the thesis fails.

Mean reversion is real, but it's more limited than most investors believe. It applies to stationary series like valuations and spreads, not to price levels themselves. It works in range-bound regimes, not trending ones. And it requires careful position management because the exceptions can be devastating.

The most dangerous phrase in mean reversion trading is 'it has to come back.' Nothing has to do anything. Markets can remain irrational longer than you can remain solvent, as the saying goes.

Successful mean reversion trading combines statistical literacy, regime awareness, and disciplined risk management. When all three align, you can profit from the market's tendency to overcorrect. When they don't, the wisest move is to wait.