Automated trading refers to algorithm-based trading using a computer. How does this work exactly and what are the chances of success?
Automated trading in day trading? Admittedly, I was not a friend of computer trading and so-called expert advisors for a long time. My mentors at that time didn’t think much of them and so I adopted this attitude without really critically questioning this view. If you want to be a successful trader, you have to master the chart yourself, right?
Yes and no. Because, as is well known, many roads lead to Rome and so there are many different trading setups and approaches in trading that are profitable in the long term. So there are also strategies in automated trading that are successful. Uwe Wagner has published the statistics of an algorithmic trading setup in a comprehensive article and writes as a former trader of Deutsche Bank that especially large investment banks rely on algotrading.
A statistic of Eurex says that about 30% of the total turnover is generated by automated trading and this share is growing by about 20% annually. How automated trading works and how you can also use it as a private trader, I will explain to you in the following.
Many retail traders also complain about algorithmic trading having conquered the cryptocurrency markets as well, above all Bitcoin as this is the largest market in that field. Learn more about trading in the Bitcoin market.
What is automated trading?
Automated trading enables trading with securities or derivatives through computer programs. The trader no longer presses the buy button himself, but leaves the order placement to his computer. He always has the possibility to intervene in the trading process himself, but theoretically he does not have to.
The trader gives the computer a command when and how he should open or close a trade. The computer forwards the order to the broker and the broker executes the order. The advantage is that this code only needs to be written once and not for each new order.
The logic of the algorithms is constantly evolving. Today, there are computer programs that generate their own trading ideas and then implement them at lightning speed. Another example are algorithms that search for current news or trends via common search engines and generate buy or sell orders in a stock, forex pair or index that matches the topic.
Once you have found a working, codable strategy, you have the opportunity to make sustainable profits, just like with manual trading.
With automated trading we describe only one main topic so far, which can be divided into different categories (e.g. high frequency trading). We will now focus on the possibility of simple algorithm trading for private traders.
How can an algorithmic strategy look like?
An algorithm is the unique, executable sequence of instructions of finite length to solve a problem. In principle, an algorithm is a logical sequence of given instructions. In CFD or future trading, for example, the instruction could be: “If the S&P500 is trading below its previous day’s closing price at 10:00 a.m., open a short position”. In this example, the computer would execute exactly one order each day. The next step would be to add a line to the code that commands the closing of the trade, e.g: If book loss -20 points, then close the trade (stop loss). If book profit +20 points, then close the trade (Take Profit).
With this simple strategy, the computer would now check every day at 10:00 a.m. whether to execute the command or not. It does this until the trade feeds it with new commands, or the account balance does not allow a trade to continue because the strategy was not profitable. Learn more strategies on this site.
What are the disadvantages of automated trading?
Computers are becoming more and more intelligent. Researchers are working flat out on various forms of artificial intelligence and intelligent machines are also being used in trading. In addition to all the intelligence and discipline a computer can bring to the table, one must also bear in mind that a certain degree of inaccuracy in trading makes sense. This inaccuracy cannot be operated by a computer at all or only to some extent.
Furthermore, errors are not always immediately detected. If a trader is not at the trading desk and the computer orders incorrectly, it will continue to do so until the order is cancelled or corrected. Such imponderables can destroy the account.
Besides the (calculable) disadvantages, there are also a lot of advantages.
What are the advantages of automated trading?
Admittedly, the example just shown is really no reason to have this simple trading decision executed by a computer. You could look at the chart at 10:00 a.m. every trading day and decide whether to trade or not.
This brings us to the first advantage of automated trading. The trader can always reconsider his decision to place a trade and make an emotional decision. If the trades have already ended 3 days in a row with a loss, some traders will doubt the set parameters and distrust their own setup.
For example, set marks (SL and TP) are moved to prevent fishing or the short is turned into a long trade because super indicator XYZ suddenly predicts a long scenario. Man is basically weak in spirit and as we all know, the stock market punishes this mercilessly.
Unlike humans, a computer does not know emotions. It stubbornly and stupefully follows the commands it receives. This gives it an enormous advantage in trading.
Another advantage is the possible leisure value for the trader. If I do not have to sit in front of the screen all the time, but let someone else trade according to my ideas, I can use the time sensibly for other projects or leisure activities. And especially chic is that I don’t even have to pay this “assistant” a salary 🙂
Furthermore, you can incorporate all the indicators you like into the strategy. You can write as many constraints as you like, but here too the basic idea of trading counts: Keep it simple, stupid.
Another positive aspect is the fact that each strategy can be checked on the basis of historical data on a demo account before using your own money. If you want to have a strategy tested, you first need someone to write the code for you (unless you are a programmer yourself). You can simply hire a service provider for this.
The next step is to open an account or demo account with a broker of your choice. Make sure that this broker offers software that allows you to backtest strategies. This is possible, for example, with brokers who offer Meta Trader 4 or Meta Trader 5. The Meta Trader is an ingenious, free trading platform with which you can trade fully automated, semi-automated or manually with many connected brokers.
With Trading Brokers you can get to know the platform by means of a free demo account and of course you can backtest the strategies.
How can I check a strategy?
It is particularly gratifying that every strategy you design can be checked using historical data (backtesting). Although this is no guarantee for the future, markets generally behave according to the same pattern, consisting of trendy and non-trendy phases with high or low volatility. This gives you a first orientation whether your trading plan can work or not.
In the Meta Trader you can test it retrospectively for approx. 5 years (depending on the broker) and according to our research so far, not only opening and closing prices are shown there, but every tick! So there is a lot of data that needs to be evaluated. In addition there is the possibility to buy and import historical data.
You also have to pay attention to the integration of the spread set by the broker when backtesting. Depending on the strategy, 10 – 20 trades per day can be made quickly, with the broker calculating the spread each time. Even if, for example, he only calculates 1 point for the S&P500, this one point can significantly influence the result. With active trades, the spread is fortunately calculated automatically.
If you have included the spread in the result of the backtesting, pay attention to the payment of the flat rate tax. If you trade with a broker who does not automatically deduct the tax, it must be deducted from the result at the end of the year.
Finally, in addition to a long period, it is also advisable to examine individual, shorter periods. All market phases should be tested once to achieve the highest possible security.
Last important note:
We have noticed that although a written code is implemented cleanly in backtesting and delivers precise results, it does not implement all commands in live trading. It may be that a stop loss is not followed up or the take profit is not executed. Pay attention to this and test, test, test before you start with real money.
Automated trading is used by both institutional investors and private traders. As it is well known that 90% of traders lose money permanently and this circumstance must be attributed to undisciplined, emotional trading, the biggest advantage of automated trading is certainly the computer’s emotionlessness factor. Anyone who has come up with an idea for a strategy must program this idea himself or with the help of service providers. Then this code can be checked with the help of (free) trading platforms like the Metatrader.
Automated trading does not mean that one is at the computer’s mercy for better or worse, because one always has the possibility to intervene. It also does not mean that a strategy that works today will still be successful next year. Markets change over time and this should always be taken into account.