How much backtesting is enough?
You’re testing some strategy on the asset you’re trading and the results seem strong. But what is the point at which you can confidently enter the market with your trading strategy and deem that you’ve backtested enough?
The answer is when you’ve managed to avoid overfitting. The results of your backtesting strategy must not work only in singular circumstances. To do this easily you want to be testing on a platform where you can easily:
1. Have all historical data available
2. Switch between timeframes, assets and trading periods
3. Can easily make changes to the strategy
If doing the above takes you little time you can be done with this in a single day. So, pick your tools wisely like this, but you can choose whatever you like.
How to make sure you’re backtesting enough:
1. Test your strategy on a period that gives you at least 30 positions — let’s say 2020
2. Test your strategy on a different period that again gives you at least 30 positions — 2021
3. If you’re getting results in the same direction change the timeframe
4. Lastly try a different asset
The 30 positions minimum is to achieve statistical significance. Without it your results may be a fluke. Aiming for higher is even better and many traders prefer at least 100.
Checking different market conditions is crucial. You’re not after the same success rate, but you want to avoid a change in the market being disastrous to your trading.
Different assets give you confidence that you’re not just trading an outlier in the market that any strategy would profit on.
What about forward testing and your execution?
This is an excellent way to really gain faith in your trading strategy. Any platform that has paper trading or even better a Manual Backtester can help you run a live market trial with 0 risk.
Manual Backtesting is essential for strategies that have a discretionary element to them. This means you don’t always make the exact same decision. There are now excellent tools out there that really allow you to examine your strategy with plenty of statistics.
Another aspect that makes Manual Backtesting important is drilling your execution. If you are not doing this there might be a gap between your plan and execution.
Unless you automate your strategy, you may be opening your trading up to this mistake that’s keeping you from being a winner just before the finish line. Manual Backtesting is the final touch that crystalizes not only your backtesting, but your actions when live trading.
