Mean Reversion Trading

What Is Mean Reversion?

Mean reversion is the statistical observation that prices tend to return to their average over time. When a stock drops sharply below its 20-day moving average, the probability of a bounce back toward the average is higher than the probability of further decline — in the absence of a fundamental catalyst.

This is the conceptual opposite of trend following. Where trend followers buy strength and sell weakness, mean-reversion traders buy weakness (oversold) and sell strength (overbought). The two strategies are naturally complementary — when one underperforms, the other tends to outperform.

Mean reversion works best in range-bound markets and on liquid, large-cap stocks where institutional activity creates natural support levels. It is less reliable on small-cap stocks, during earnings seasons, or in strongly trending markets where "oversold" can keep getting more oversold.

Core Entry Signals

RSI Extremes

The Relative Strength Index (RSI) measures how overbought or oversold a stock is relative to its recent price action.

• RSI(2) below 10: Deeply oversold on a 2-period RSI — buy signal. This is the Larry Connors system, one of the most backtested mean-reversion setups. Exit when RSI(2) crosses above 70.
• RSI(14) below 30: Standard oversold reading — longer-term signal with fewer false entries but slower response.
• RSI(14) above 70: Overbought — potential short or exit signal for long positions.

Bollinger Band Touch

Bollinger Bands plot 2 standard deviations above and below a 20-day moving average. When price touches or penetrates the lower band, it is statistically extended and likely to revert toward the middle band (the 20-day MA). The reverse applies at the upper band.

Distance from Moving Average

Simple but effective: buy when price is more than 2 standard deviations below the 50-day moving average; exit when it returns to the average. The further the stretch, the stronger the reversion signal — but also the higher the risk of a fundamental breakdown.

Pairs Trading

A market-neutral form of mean reversion. Identify two correlated stocks (e.g., Coca-Cola and Pepsi). When their price ratio diverges from the historical norm, buy the underperformer and short the outperformer. Profit when the spread reverts. This hedges out market risk entirely.

Risk Management

Mean reversion's biggest risk is buying into a falling knife — a stock that is oversold because something is fundamentally wrong, and it continues falling far beyond any statistical extreme. Defences:

• Hard stop loss: Even though the strategy bets on a bounce, set an absolute stop at 5–8% below entry. If the thesis is wrong, cut the loss before it becomes catastrophic.
• Fundamental filter: Avoid mean reversion on stocks reporting earnings within 5 days, stocks under SEC investigation, or stocks that have gapped down on fundamental news. Reversion is a statistical phenomenon — it doesn't apply when the company's value has actually changed.
• Market regime filter: Mean reversion works best when the broad market is above its 200-day moving average (bullish regime). In bear markets, oversold stocks tend to stay oversold or get worse.
• Position size: Because mean reversion trades against the prevailing move, keep individual position sizes smaller (0.5–1% risk per trade) than you would for trend-following trades.

How AI Tools Enhance Mean Reversion

• Trade Ideas — Holly AI scans the entire market for oversold setups that match historically profitable mean-reversion patterns. It can identify RSI extremes, Bollinger Band touches, and distance-from-MA conditions across thousands of stocks simultaneously. AIClarity Score: 8.5.
• TrendSpider — automated RSI and Bollinger Band scanning with multi-timeframe confirmation. Its backtesting engine can test mean-reversion rules against historical data to validate your parameters. Review →
• Kavout — uses machine learning to score stocks based on multiple factors including mean-reversion potential. Stocks with low Kavout scores that are showing technical oversold signals can be strong mean-reversion candidates. AIClarity Score: 7.6.

"Mean reversion is the most psychologically comfortable strategy — you're buying low and selling high. The danger is that comfort makes you sloppy about stops."

Key Takeaways

• Mean reversion buys oversold and sells overbought, betting that extreme moves are temporary.
• Win rate is high (60–70%) but average gain per trade is smaller than trend-following strategies.
• RSI extremes, Bollinger Band touches, and distance-from-MA are the core entry signals.
• The main risk is catching a falling knife — always use hard stops and fundamental filters.
• Works best in range-bound, bullish-regime markets on liquid, large-cap stocks.
• Naturally complementary to trend following — combining both smooths portfolio returns.