How do you test your backtesting with new data?

1 Choose your data

The first step to test your backtesting with new data is to choose the data source and the time period that you want to use for out-of-sample testing. Ideally, you should use data that is similar to the data that you used for backtesting, but from a different time period or market segment. For example, if you backtested your strategy on the S&P 500 index from 2010 to 2020, you could use the S&P 500 index from 2021 or another major index like the Nasdaq 100 or the Dow Jones for out-of-sample testing. You should also make sure that the data quality is high and consistent, and that it covers all the relevant factors that affect your strategy, such as price, volume, indicators, and signals.

2 Run your strategy

The next step is to run your strategy on the out-of-sample data and compare the results with the backtesting results. You can use the same software or platform that you used for backtesting, or you can use a different one that supports out-of-sample testing. You should apply the same parameters, rules, and settings that you used for backtesting, and avoid making any changes or adjustments based on the new data. You should also use the same performance metrics and criteria that you used for backtesting, such as return, risk, drawdown, Sharpe ratio, and win rate. The goal is to see how your strategy performs on unseen data and whether it maintains its profitability and robustness.

3 Evaluate your results

The final step is to evaluate your results and determine whether your strategy passes or fails the out-of-sample test. There is no definitive answer to this question, as it depends on your expectations, preferences, and risk tolerance. However, a general rule of thumb is that your strategy should have similar or better performance on the out-of-sample data than on the backtesting data, or at least within a reasonable margin of error. If your strategy performs significantly worse on the out-of-sample data, it may indicate that your strategy is overfitted to the historical data and not adaptable to new market conditions. In that case, you may need to revise your strategy, use more data, or apply some techniques to reduce overfitting, such as cross-validation, walk-forward testing, or regularization.

Testing your backtesting with new data is a crucial step to validate your trading strategy and increase your confidence in its future performance. By using out-of-sample data, you can avoid overfitting and verify the robustness and consistency of your strategy. You can also use the out-of-sample test as a feedback loop to improve your strategy and optimize your parameters. However, you should also remember that testing your backtesting with new data is not a guarantee of success, and that you should always monitor your strategy and update it as the market evolves.

4 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?