So you want to get into automated trading? This article aims to share with you my experiences going through the journey of becoming an automated trader and building TradersPost.
Everything starts with an idea. However, an idea is only worth 2 cents. How do you translate that idea into an actual executable automated trading strategy?
If you don’t know how to code and aren’t up for the challenge of learning how to code, you can very easily hire a Pine Script developer to translate your idea into code. You can find Pine Script developers on sites like Fiverr or Upwork. We’ve compiled a list of some Pine Script developers here that have helped other TradersPost customers build their strategy.
Finally, you can use the TrendSpider No Code strategy builder to build your strategy and rapidly test ideas without having to write any code.
Once you have your idea converted into a strategy, you need to backtest it. Ideally you should backtest your strategy across multiple different time periods when the market conditions are wildly different. The goal is to get an idea of how robust your strategy is and how it will perform in different market conditions.
Be sure to account for slippage in your backtest. Slippage can occur due to the spread between the bid and ask but you also have to account for slippage caused by the time in between when a signal is generated and when the order is executed.
If you’ve gotten this far and your backtest is looking good, then you are ready to forward test your strategy in a paper broker. You can read more about paper trading here.
It is important to compare the fill prices in your strategy backtest to the fill prices in your paper broker. If there are big differences in fill prices, then something probably isn’t configured quite right and you may have too much or too little slippage configured in your backtest.
How long should you forward test your strategy? There is no such thing as testing too much, but usually it is good to forward test for at least a month, if not longer. The more trades you take and the more data you collect, the more you will understand about your strategy.
If you’ve forward tested your strategy and everything is still looking good, then you are ready to try it with a live broker. It is important to test with small size and monitor closely when you start testing with live money. Use a quantity of 1 to start.
Just like you did with your paper trading, compare the fill prices in your live broker to the fill prices in the backtest. If there are differences between the fill prices, then you likely need to tweak your slippage settings in your backtest to be more accurate.
Congratulations! If you got this far then you have an automated trading strategy live. Your journey doesn't end here. Automated trading requires constant tweaking, learning and evolving. The market is forever always changing. All strategies work, just not all the time. The trick is knowing which strategies to run and when. So ideally you would have multiple strategies to choose from depending on some macro market conditions. Good luck on your automated trading journey!