Survivorship Bias in Trading: Avoiding the Illusion of Success

Survivorship Bias in Trading: Avoiding the Illusion of Success

INTRO: Imagine you're trying to learn how to play a game. You only study the people who have already won, ignoring all the players who lost. You might think you understand the game, but you're missing a crucial piece of the puzzle: the reasons why people fail. Survivorship bias in trading works the same way. It's the tendency to focus only on the successful assets or strategies, while overlooking the ones that didn't make it. This can lead to a distorted view of the market and potentially, costly trading decisions.

Definition

Survivorship bias is a type of cognitive bias where we only consider the successes in a dataset, while ignoring the failures. In trading, this means focusing on assets or strategies that have survived over a period, without accounting for those that have been delisted, liquidated, or otherwise failed.

Key Takeaway

Survivorship bias leads to an inaccurate assessment of trading strategies and market performance by ignoring the failures, leading to potentially inflated returns and underestimated risks.

Mechanics

Survivorship bias skews data in several ways:

1. Selection of Data: When analyzing historical data, traders often use databases that only include currently active assets. This automatically excludes assets that have failed, such as delisted stocks or bankrupt companies. This is particularly relevant in the crypto space, where projects fail frequently.
2. Performance Inflation: By excluding the losers, performance metrics are artificially inflated. For example, if you backtest a strategy using only surviving stocks, the results will likely be better than if you included the performance of stocks that went bankrupt.
3. Risk Underestimation: Survivorship bias can lead to an underestimation of risk. Because failing assets are excluded, the analysis may not accurately reflect the potential for losses in the market. The volatility of strategies using only successful assets will typically be lower than a strategy that is tested against a comprehensive dataset.
4. Misleading Comparisons: Peer group comparisons and relative performance evaluations can be distorted. If fund managers who manage underperforming funds are shut down, the remaining funds will appear to perform better relative to their peers, even if they are not necessarily exceptional.

Trading Relevance

Survivorship bias affects trading decisions in several ways:

• Overestimation of Returns: Traders may overestimate the potential returns of a strategy or asset class if they only consider the successful examples. This can lead to unrealistic expectations and poor risk management.
• Poor Risk Management: The underestimation of risk can lead to larger positions and a greater exposure to potential losses. If a trader believes that a strategy has a lower risk profile than it actually does, they may not adequately protect their capital.
• Ineffective Strategy Development: When backtesting, if survivorship bias is present, a strategy may appear to be profitable when it is not. This can lead to the implementation of flawed strategies that ultimately lose money.
• Misguided Investment Decisions: Investors may be drawn to funds or strategies that have shown strong historical performance, without realizing that this performance may be inflated by survivorship bias.

Risks

• Overconfidence: Overestimating the potential for returns can lead to overconfidence in a trading strategy. This can make traders more likely to take unnecessary risks.
• Poor Capital Allocation: Misleading performance data can lead to poor capital allocation decisions. Traders may allocate too much capital to strategies that are not truly profitable.
• Emotional Trading: A false sense of security can lead to emotional trading decisions, as traders may become less cautious and more susceptible to market fluctuations.
• Failure to Adapt: Strategies that appear successful due to survivorship bias may fail in different market conditions. Traders may be slow to adapt to changing market dynamics if they are relying on inaccurate historical data.

History/Examples

• Mutual Fund Performance: Consider mutual fund performance. If you only look at the funds that are still in existence, you will likely see a higher average return than if you included the performance of funds that have been closed or merged due to poor performance. This is because underperforming funds are often shut down, and their poor performance is not included in the historical data.
• Venture Capital: Venture capital is another prime example. Many startups fail. If you only analyze the successful startups, you get a skewed view of the overall investment landscape. Including the failures is crucial for a realistic assessment of the risk and returns.
• Early Crypto Projects: Think back to the early days of Bitcoin and altcoins. Many projects launched with great fanfare but ultimately failed to gain traction and disappeared. Analyzing only the surviving projects like Bitcoin and Ethereum gives an incomplete picture of the market and its risks.
• Backtesting Software: Many backtesting platforms may not automatically account for survivorship bias. Traders must actively incorporate delisted assets or adjust their data to avoid this bias.
• Market Indices: Market indices like the S&P 500 can also be affected. The index is constantly rebalanced, and failing companies are removed. This means the index's historical performance may be inflated by the exclusion of these failures.

Mitigation Strategies

• Comprehensive Data: Use datasets that include both surviving and non-surviving assets. This is the most effective way to address survivorship bias. Databases that provide historical data on delisted stocks and failed projects are essential.
• Backtesting Adjustments: Adjust backtesting methods to account for survivorship bias. This can include incorporating delisted assets, simulating market conditions that reflect the potential for failures, and using statistical techniques like bootstrapping and Monte Carlo simulations to create more robust trading strategies.
• Realistic Expectations: Understand that past performance is not indicative of future results. Be skeptical of strategies or assets that show consistently high returns.
• Risk Management: Develop robust risk management strategies that account for the potential for losses. Diversify your portfolio, use stop-loss orders, and manage position sizes appropriately.
• Focus on the Process: Focus on developing a sound trading process rather than chasing high returns. A well-defined process includes thorough research, risk management, and the ability to adapt to changing market conditions.
• Third-party Verification: Consider using third-party verification services to validate your trading strategies. These services can help to identify potential biases and ensure that your strategies are robust.
• Use Systematic Trading: Systematic trading offers the best defense against survivorship bias by ensuring decisions are based on comprehensive, historical datasets and not just the winners. This approach helps to remove emotional biases and provide a more objective view of the market.

Conclusion

Survivorship bias is a significant risk in trading that can lead to inaccurate assessments of strategies, inflated expectations, and poor trading decisions. By understanding the mechanics of survivorship bias and implementing mitigation strategies, traders can create more realistic and reliable strategies, improve risk management, and ultimately, increase their chances of long-term success. Always remember to analyze the failures, not just the successes, to get a clear and accurate picture of the market.