Algorithmic trading might seem confusing and complicated at first sight. Many new traders find algorithmic trading hard to grasp, and therefore wonder if algorithmic trading does work.
Yes, algorithmic trading does work. If you have a robust trading strategy that has been vetted through rigorous robustness testing, you have a great chance of making money. In fact, for most traders, algorithmic trading works better than discretionary trading since the rules are quantifiable and retestable.
Here at The Robust Trader we have been trading algorithmically for a long time, and know with certainty that it DOES work
And not only that. We KNOW that algorithmic trading, provided that you have a properly backtested strategy, will work for many more people than, say, day trading.
Why is that?
Well, in today’s markets, it is becoming increasingly more difficult to trade discretionarily. Those times when you could trade a simple strategy discretionarily and make seemingly endless amounts of money, are gone. The markets have become so efficient, and the edges so small that most people need something more accurate than their own capability to observe patterns in the market. This is where algorithmic trading comes in very handy!
By designing a trading strategy that is tested on historical data, it is possible to uncover and notice market behavior, or edges, that you cannot spot by observing the markets visually. However, this does not mean that the things that work in algorithmic trading are very complex. Rather the opposite, the things that work the best tend to be the simplest!
So, if algorithmic trading works, how should you go about to make it work for you? Let’s find out!
Table of contents:
The Steps to Make Algorithmic Trading Work for You!
There are several steps involved in making algorithmic trading work. If you are serious about it, we recommend that you read our complete guide to algorithmic trading. There we cover ALL aspects of algorithmic trading that a beginner needs to know in order to become profitable in the markets!
You could say that making algorithmic trading work includes doing the following things:
- Finding a strategy
- Robustness Testing
- Creating a Portfolio
- Monitoring the strategies
So, let us go through these briefly, one by one!
The Trading Strategy
The trading strategy is the core of every algorithmic trader’s business. Without it, you basically have nothing. This is also reflected in how much time an algorithmic trader typically spends on finding trading strategies. In our case, if we had 10 hours to spend on our trading, we would probably allott 8 or 9 hours to finding strategies. That is how important it is!
In order to find a trading strategy, you need to go through a few steps:
Find An Idea
The first step is to define a trading idea that will form the base of your trading strategy. What you want is to come up with a pattern that could have predictive potential in the market. It could be to buy once the market has performed two consecutive lower lows, or the opposite. It really could be anything, and the only limitation is your own fantasy.
It really pays off to be creative in this phase. The most successful strategies are those that most other market participants have not discovered, and you can only find them by thinking in unconventional ways, or in other words, being creative.
This means that the best strategies are unlikely to come from a trading forum or be complete strategies you find online. Having that said, trading forums and other online resources are great resources to spark new ideas, that could become new algorithmic trading strategies!
Backtest It
Once you have your trading idea, it is time to go on to what is called “backtesting. Backtesting is the process of testing your strategy on historical data, to validate or disprove an idea. You simply program the strategy into your platform of choice, and see what results show up.
Now, if the equity curve is upward sloping, you may have found and quantified some market behavior. In other words, you might have found an edge!
Now you may go on to add filters and additional conditions to increase the accuracy of the signal, to get a better strategy!
Robustness Testing
Did you think that you can start trading something that you find through backtesting, right away? Well, that is not the case.
The reason is curve fitting, which is one of the biggest hurdles traders face!
Curve fitting means that your trading strategy was fit to random market noise, rather than true market behavior. Trading a curve-fit strategy is like gambling, since the strategy lacks an edge.
If the right robustness testing methods are not employed, curve fitting has the potential of destroying your trading career. This is because a curvefit strategy does not continue to make profits, but most times plunges right after it has been developed!
In our massive article on Algorithmic trading, we look closer at how to build a trading strategy, and some methods to avoid curve fitting!
Once the strategies are in place, one very fun part remains, namely creating a portfolio of strategies!
Creating the Portfolio
A lot of traders believe that they just need one strategy to be profitable in the markets. They spend all their time looking for the perfect trading strategy and forget about the bigger picture. All trading strategies will fail eventually, and as such you must make sure that one single strategy does not stand for all your profits.
As an algorithmic trader, you can trade an almost limitless amount of strategies at once, since they are executed by a computer. The more strategies you have, diversified across different markets and timeframes, the smaller the risk!
What you also will find when you design your portfolio of strategies, is how much of a difference having the right strategies can make to your overall profit potential. When one strategy goes down, you want another go up, so that you can increase your position size, and make more money in the long term. The less correlated the strategies are, the more money you will be able to make, and as such, you will benefit greatly from expanding into different markets and timeframes. This is because finding uncorrelated strategies in different markets is much easier than finding uncorrelated strategies in the same market!
Monitoring the Strategies
How alluring care-free trading even might be, algorithmic trading requires you to monitor the order execution at least a few times each day. There are many things that could go wrong, and even if trading software is becoming better and better, there will be hiccups. Your orders might be rejected, or the software could lose its connection to the server.
The only way to mitigate the risk of suffering severe losses due to hardware or software failure, is to monitor the order execution a few times every day. Since you probably want to run your trading on a remote server, you could be on the run and check in on your systems through your smartphone!
In other words, your freedom does not have to suffer at all because of this! And to be honest, checking in on your servers once in a while is a very little price to pay to reap the benefits that algorithmic trading offers!
Summary
Algorithmic trading indeed works, and done correctly your chances of success might be higher than if you were to try out discretionary trading.
The trading strategy really is what algorithmic trading is mostly about, and most of the work you put in will be in your search for new, fresh logics that work in the markets.
Here you can read more about algotrading in our archives.
FAQ
How Do I Find a Trading Strategy for Algorithmic Trading?
To find a trading strategy, start by defining a unique trading idea or pattern with predictive potential. Be creative in generating ideas, as the most successful strategies are often unconventional. Once you have an idea, backtest it on historical data to validate its performance and quantifiable market behavior.
Why Is Robustness Testing Important in Algorithmic Trading?
Robustness testing is crucial to avoid curve fitting, where a strategy is fit to random market noise rather than genuine market behavior. Trading a curve-fit strategy is akin to gambling, as it lacks a genuine edge. Robustness testing ensures the strategy can withstand various market conditions.
Why is Monitoring Algorithmic Trading Strategies Important?
Monitoring is essential to identify and address issues that may arise during order execution. While algorithmic trading offers automation, regular checks are necessary to mitigate risks of rejected orders or software hiccups. Monitoring ensures the overall health and performance of the algorithmic trading system.
