
This episode discusses the pitfalls of overfitting in trading strategies and the importance of robust backtesting.
When it comes to systematic trading trading, one of the most common objections concerns the reliability of backtesting itself. Many people believe that trading systems only work on past market data and that the results achieved through historical testing have no real predictive value. And to some extent, that criticism makes sense. Achieving excellent backtest results is not particularly difficult: all you need to do is keep adapting rules, filters, and parameters to past market data until you create a strategy that appears almost perfect. The problem is that, in many cases, you are not actually measuring the true statistical effectiveness of the strategy, but simply how well it fits the historical data used to build it. And this is exactly where one of the most common mistakes in systematic strategy development comes into play: overfitting. Overfitting refers to the tendency to create systems that are too closely "tailored" to historical data and therefore tend to lose effectiveness once applied to real markets. So does this mean that backtesting is useless? Absolutely not. When used correctly, historical testing is an essential tool for evaluating the robustness of a strategy…
Organizations: Unger Academy
Explore listener stats, chart rankings, contacts and more on the The Unger Games - Trading Tips by the 4-Time World Champion podcast page.