Negative Skewness: The Hidden Risk of Mean Reversion

Negative Skewness: The Hidden Risk of Mean Reversion

Mean reversion strategies often exhibit negative skewness. This means they produce many small gains and few large losses. The distribution of returns has a long left tail. This characteristic can surprise traders unfamiliar with it.

Consider a simple mean reversion strategy on SPY. This strategy buys SPY when its 2-day Relative Strength Index (RSI) falls below 20. It sells when the 2-day RSI rises above 80. From January 1, 2000, to December 31, 2023, this strategy generated 1,250 trades. 1,000 trades were profitable, showing average gains of 0.5%. 250 trades were unprofitable, showing average losses of 2.0%. The strategy had an 80% win rate. However, the largest single loss was -15.0% during the 2008 financial crisis. The largest single gain was 3.0%.

This return profile creates a specific psychological challenge. Traders become accustomed to frequent small wins. They may underestimate the potential for significant drawdowns. These large losses can erase months of accumulated small gains.

Understanding Skewness in Returns

Skewness measures the asymmetry of a probability distribution. A normal distribution has zero skewness. Positive skewness indicates a long right tail. Negative skewness indicates a long left tail.

Mean reversion strategies profit from temporary deviations from an average. They typically bet against momentum. When the market moves strongly against the position, the strategy incurs losses. If the market continues to trend, these losses accumulate. The strategy waits for a reversal. If the reversal does not occur quickly, losses can become substantial.

For example, a short-term mean reversion strategy on the EUR/USD currency pair might sell when the price moves two standard deviations above its 20-period moving average. It buys when the price moves two standard deviations below. Most of the time, the price reverts quickly. The strategy captures small profits. However, during a strong trend, like the EUR/USD decline from 1.35 to 1.20 in mid-2014, the strategy would suffer. It would repeatedly short the rally attempts, then get stopped out as the trend continued. Each small loss adds up. A large, sustained trend produces many small losses or one very large loss if the position size is not managed.

Consider a portfolio of 10 mean-reverting strategies. Each strategy has an expected daily return of 0.05% and a standard deviation of 0.5%. The strategies are negatively skewed. This means 90% of days show small gains, and 10% show larger losses. Over 200 trading days, 180 days show small profits. 20 days show larger losses. One or two of these loss days might be extreme outliers.

Quantifying Negative Skewness

Traders quantify skewness using the third moment of the return distribution. The skewness coefficient is calculated as:

$Skewness = \frac{E[(R - \mu)^3]}{\sigma^3}$

Where: $R$ = Return $\mu$ = Mean of returns $\sigma$ = Standard deviation of returns $E$ = Expectation operator

A skewness value below zero indicates negative skewness. A value of -0.5 is moderately skewed. A value of -1.0 or less is highly skewed.

Let's analyze the SPY mean reversion strategy mentioned earlier. From January 1, 2000, to December 31, 2023, the daily returns of this strategy had: Mean daily return: 0.02% Standard deviation of daily returns: 0.8% Skewness coefficient: -1.25

This value of -1.25 confirms significant negative skewness. This means the strategy produced more frequent small gains and fewer, but larger, losses.

Compare this to a long-only S&P 500 strategy. The S&P 500 (SPY) itself from January 1, 2000, to December 31, 2023, had: Mean daily return: 0.03% Standard deviation of daily returns: 1.2% Skewness coefficient: -0.28

The S&P 500 also exhibits some negative skewness, but significantly less than the mean reversion strategy. This difference highlights the increased tail risk in mean reversion.

Mitigating Negative Skewness Risk

Traders must address negative skewness. Ignoring it leads to unexpected drawdowns.

1. Position Sizing: Implement robust position sizing rules. Do not increase position size after a string of small wins. Use fixed fractional position sizing or volatility-adjusted sizing. For example, risk a fixed percentage of capital, like 0.5%, per trade. If the stop loss is 2%, the position size is 0.5% / 2% = 25% of trading capital. This prevents overleveraging into a potential large loss.

2. Stop-Loss Orders: Strict stop-loss orders are non-negotiable. Define the maximum acceptable loss per trade before entry. Execute the stop-loss without hesitation. Do not widen stops. For instance, if a mean reversion trade on AAPL is entered at $170, set a stop at $165. Adhere to it.

3. Portfolio Diversification: Combine multiple mean reversion strategies across different asset classes. Diversify across uncorrelated assets. A mean reversion strategy on equities may have low correlation with a mean reversion strategy on commodities. This reduces the impact of a large loss in one strategy. A portfolio of 10 strategies, each with a -1.0 skewness, will have a lower overall portfolio skewness if the strategies are uncorrelated.

4. Trend-Following Overlay: Incorporate a trend-following component. This helps identify strong trends where mean reversion strategies perform poorly. For example, only execute mean reversion trades if the longer-term trend (e.g., 200-day moving average) is flat. Avoid mean reversion trades against a strong trend. If SPY is above its 200-day moving average, only consider long mean reversion trades. If SPY is below its 200-day moving average, avoid short mean reversion trades.

5. Option Hedges: Purchase out-of-the-money put options to hedge against extreme downside moves. This reduces the left tail risk. For example, if holding a portfolio of mean-reverting equity strategies, buy SPY puts with a strike price 10% below the current market price. This provides protection during market crashes.

6. Stress Testing: Conduct rigorous stress testing. Simulate extreme market conditions. Analyze how the strategy performs during historical crises (e.g., 2008, 2020). Determine the maximum drawdown under these scenarios. This prepares for the large, infrequent losses.

Mean reversion strategies offer attractive characteristics, but their negative skewness demands careful management. Implement robust risk controls.