Last Updated on 8 January, 2022 by Samuelsson
It is becoming increasingly more difficult to trade profitably with the discretionary trading approach in recent times because the markets have become more efficient, with very few trading edges. So, it makes sense to want to try algorithmic trading. But how do you become an algorithmic trader?
Becoming an algorithmic trader is not something that can happen overnight. It requires a lot of planning and learning, but it is something you can do if you are determined to do it. Being a multi-step process, you have to start with the basics and gradually progress, but enrolling in an algorithmic trading course can facilitate things.
In this post, we will discuss the following:
- Who an algorithmic trader is
- Why you may want to become an algorithmic trader
- What you need for algorithmic trading
- The steps to becoming an algorithmic trader
- The psychology of algorithmic trading
- Some algorithmic trading tips
Who is an algorithmic trader?
Algorithmic trading, also known as algo trading, is a form of trading where the computer is given a script with trading rules to execute for you. So, it is an automated trading method that requires a hands-off approach when implemented. An algorithmic trader is someone who uses this method of trading.
But it doesn’t mean that an algorithmic trader doesn’t do much. In fact, an algo trader does some serious research in the markets, finds a trading edge, writes a code for the edge, thoroughly tests it for robustness, and sets it up to trade. Moreover, the trader will still have to regularly check in on your algorithmic trading system to be sure that everything is running smoothly.
Nevertheless, algorithmic trading still gives you a lot of freedom. With algorithmic trading, you are free to do any other thing you want while the computer takes care of the trading for you. Once it is set up and functioning smoothly, you only check it periodically; the rest of the time you are free to do whatever you want.
For us at the Robust Trader, we think algorithmic trading is much better than any other form of trading in many respects. For example, algorithmic trading makes it possible to trade an almost limitless amount of strategies at once — we trade over 100 strategies ourselves in many different markets, and those strategies range from day trading to longer-term position trading since algorithmic trading is not limited to any trading style and it is beneficial to diversify into many different strategies in different markets.
As an algorithmic trader, your major work will be to research and develop new trading strategies and write codes for them. That is the most difficult aspect of the game. To succeed in this, you must have a passion for constant research and also learn how to code in the language of the trading platform you are to trade with.
Trading ideas can come from anywhere — fellow traders, things you read online, spur of the moment, etc. It is your job to develop this idea and code it, after which you backtest it to know if it has any merit. Next, test its robustness before bringing it to live trading.
Why become an algorithmic trader
The truth is that in today’s financial trading world, it is becoming increasingly more difficult to make money from discretionary trading because the markets are getting more efficient. Gone are those times (early and mid 20th century) when you could trade a simple discretionary strategy or use your instincts and make a huge amount of money consistently.
In an increasingly AI-driven world of ours, the markets have become so efficient, and trading edges so few and far-fetched that you will surely need something more accurate than your basic ability to observe patterns in the market. Hence, the need for an algorithmic trading method that can constantly monitor the market for trade setups.
With the algorithmic trading approach, you will be designing a trading strategy that is tested on historical data, which makes it possible to uncover and notice market behavior (trading edges) that you cannot spot by observing the markets visually. But note that this does not mean that the things that work in algorithmic trading are very complex. On the contrary, the simpler the better. Actually, the things that work best tend to be the simplest.
Becoming an algorithmic trader enables you to take full advantage of the few available trading edges. The benefits include the following:
- 24/7 trading: Algo trading enables you to trade even when you are asleep. This is very suitable for those markets that open almost 24/7. The ability to trade 24/7 allows you to take every trade setup that forms, which means more money for you.
- Risk management: The computer takes care of the order execution, you can trade many markets at the same time. Being diversified across many markets is a popular way to manage portfolio risk.
- A quantifiable edge: Since algorithmic trading strategies are backtested rigorously before employed and traded live, you know your odds before you start trading live, and you can manage your position size accordingly.
- Emotional control: Since the trade execution is carried out by a computer, your trading emotions will have less impact on the trading process.
- More money: Algo trading allows you to trade many markets with different uncorrelated strategies at the same time, so when one strategy is in drawdown, another is making profits. In the end, more money.
- Fewer mistakes: Trading is very difficult, with high chances of making mistakes. Automating your trading with algos ensure that mistakes are kept to a minimum.
- More free time: As an algorithmic trader, you do not need to watch your trading screen all day, even if you are implementing an intraday trading strategy. The computer will take care of the trading for you, giving you more free time to do other things you want.
- Networking gets easier: Algo trading affords you more time to interact with others. Moreover, with strict and quantifiable rules, algo trading makes it possible to share non-cumbersome ideas with fellow traders.
- Easy start for a beginner: By buying a robust trading strategy, beginners can easily start with algorithmic trading.
What you need for algorithmic trading
There some hardware and software you will need to kickstart your algorithmic trading journey. These are the main ones:
- A starting trading capital: The size of your starting trading capital will depend on what markets you want to trade and how many markets you want to trade simultaneously.
- Your trading infrastructure: These include your PCs, remote server hosting, cloud storage backups, reliable internet, and possibly emergency power supply.
- A trading platform: This may be provided by your broker or you may have to subscribe to one
- Market data: Some trading platforms may come with market data, but it doesn’t, you need to purchase the service.
The steps to becoming an algorithmic trader
Becoming an algorithmic trader is not just about wishful thinking; it requires you to take some practical steps to learn the process, develop your strategies, and implement them. Here are the steps to take:
1. Get the right information
The first step in starting anything is always to seek knowledge about the thing. In this case, you can read books about algorithmic trading, such as “Trading and Exchanges: Market Microstructure for Practitioners”, “Algorithmic Trading: Winning Strategies and Their Rationale”, and “Algorithmic Trading and DMA: An Introduction to Direct Access Trading Strategies.” However, the best way to gain the right information in the shortest possible time is to enroll in an algorithmic trading course.
2. Make available the necessary hardware you will need
You will need some basic equipment to start algo trading. One of them is a computer or laptop. If you can get two monitors, it is preferable, but you can start with one. You do not necessarily need an expensive computer — a PC with 4 or more cores and 8GB RAM is just fine. Ensure you have a high-speed internet connection, and you may have to subscribe to a remote server to host your algorithmic trading. Also, you should get storage disks for backing up your trading files and codes. Cloud storing services are also necessary.
3. Register with the right broker
You need to register with a broker. It is your broker that links your trade orders to the markets. There are many of them you can choose to register with — TradeStation, Interactive Brokers, Amibroker, etc. Some of them provide you with a trading platform and market data, but others don’t. TradeStation is known to provide its registered clients with its proprietary TradeStation trading platform, as well as market data.
4. Ensure that all your trading software is ready
You will need a trading platform and market data, which include both historical and real-time market data. For us, the best trading platforms are TradeStation, Amibroker, and MultiCharts. TradeStation offers a brokerage service and provides market data. If you choose any of the other two, you will need to source for market data: make sure the historical data is up to 10 years because you will need it when backtesting your strategy.
5. Setup up your trading screen
Install your trading software and set up your trading screen. Open the trading platform and load the charts and other trading tools from whichever source you are getting them. When your charts are fully set and ready, with both real-time and historical market data in various timeframes, you can now start the main work of developing a trading algorithm.
6. Research some trading ideas to formulate trading strategies
At this stage, you may already have some trading ideas, but if you don’t, do a bit of research to find trading ideas that worth checking out. You can check some financial journals where academics publish their theories and findings. An example of a trading idea is mean reversion. Formulate specific rules for each idea to turn them into trading strategies.
7. Code and backtest the trading strategies
Write the codes for the trading strategies and run your script in the strategy tester to backtest them one after another. With the Amibroker platform, you can backtest multiple strategies simultaneously. Ensure you use up to 10 years’ data in your backtesting. The result would determine if you go ahead with forward-testing or modify the strategy and backtest again.
8. Forward-test the strategies
When your backtesting results are good enough, you should go ahead and forward-test the strategies for robustness. This tells you whether the strategies can be profitable in a real market environment or not because backtesting results may be hampered by curve fitting.
9. Implement your algo system and monitor your portfolio
With everything looking great, you may now implement your algo system to execute your trades for you. However, you need to keep monitoring the system from time to time to be sure that it is performing as expected. Just checking it each day may be enough. But more importantly, set your risk management parameters so as to protect your trading capital. Depending on how your algo is coded, it may have stop-loss features, but more importantly, ensure to set your maximum acceptable drawdown.
You might be interested in the following article.
The psychology of algorithmic trading
Algorithmic trading will surely help if you cannot cope with the intense psychological pressure that comes with trade execution. However, while algorithmic trading relieves you from the burden of placing the orders manually, that does not mean that it relieves you from all the psychological pressures that are associated with trading. Even when the execution is automated, you can still have psychological issues with algo trading for any of the following reasons:
- If you frequently check your profit and loss levels in your ongoing trades, you will get stressed out by the changes in the account equity.
- Frustrations may be associated with developing your trading strategies since finding good trading edges are hard.
- If you are in a huge drawdown, you may find yourself getting seriously worried and confused, especially if it is a new strategy.
While these issues can come up, there are things you can do to reduce the psychological stress associated with algorithmic trading. Here are some things you can do:
- Reconfirm your goals and set parameters
- Make sure you know your drawdown tolerance and set it there; then try to stay calm knowing that everything is going according to the plan
- Avoid checking your daily profit/loss and monitoring your daily equity