What Is Algorithmic Trading?

Hello and welcome to the Amplify Trading YouTube channel. My name is Milan and I'm a technology analyst for Amplify Trading. I'm sure many of you have never heard of me before, and that's perfectly understandable. I only started working for Amplify Trading back in October 2019, which feels like a lifetime ago. And a little bit about myself, I studied computer science at Royal Holloway University of London and I currently work alongside the technology team and the business side of AmplifyTrading on a variety of different tasks. So today on this video, we're going to be discussing what is algorithmic trading, and this is going to be the start of a new series that relates to tech and finance, and we're going to be exploring a couple different concepts over the next coming weeks. So you can find my videos on the AmplifyTrading YouTube channel every other week. So let's get straight into what is algorithmic trading. So the simplest way to understand what is algorithmic trading is to think of it as code and technical analysis that allows you to automate entering and exiting trades. These trades can then be entered and exited according to a predetermined set of parameters, which are your triggers essentially, for entering that trade. These triggers or predetermined parameters could be based off of volatility levels, for example, or price movement. Then once these predetermined criteria are hit, you'll be entered into a short or long position on your behalf by the trading algorithm. So that raises the question, why do people use algorithmic trading? Well, one of the first things you can think about is that this algorithm will essentially enter into positions when you're awake or asleep, and that can definitely be seen as a benefit. So that basically means you can let the algorithm do what it wants no matter where you are or what you are doing. This then gives you increased market opportunities and maximizes your exposure to the market itself. Another benefit of algorithmic trading is that you're not plagued with one of the key issues that a lot of traders face these days. The hardest obstacle a trader is potentially faced with, arguably, is human emotion, and using an algorithm deployed on a server can take this human emotion completely out of the equation, you know. An algorithm is not going to hesitate if a hit— a target or predetermined level is hit. It will always execute the trade on your behalf. I'm sure a lot of traders can agree about that feeling when you have a great trade running and you decide to book your profits early, and then you look back at it the next day, or, you know, 10 minutes later even, and you see if you would just let it run that little extra mile, it would have been a fantastic trade. And this is something that the algorithm is designed to overcome, essentially. And that's exactly what algorithmic trading aims to improve. You're not having that human emotion get in the way of you extending your profits, and, you know, cutting your losses early. The algorithm, at the end of the day, is just a bunch of zeros and ones, and you can tell it exactly what to do, and it will stick exactly to what you've told it to do. Now, that all sounds lovely and great, you you know, for algorithmic trading, and you, I'm sure you're all sitting there right now thinking, I want to get straight into algorithmic trading. There are a few things that algorithms still struggle with today, and one of these being optimization and overfitting the algorithm to the market. This idea of overfitting or optimizing way too much for the market is one of the key issues that plagues a lot of algorithmic traders, and it eventually even leads to some giving up on their strategies. This idea of overfitting or overanalyzing is when you're doing backtesting and you look back at 5 or 10 years worth of data and you make tweaks to the algorithm, you know, adjusting the simple moving averages or exponential or whatever strategy is you're employing to the point where in the past it may perform and excel exceedingly above expectations, but when this is deployed on a live market, you don't get the same results as you were from the past, and that's simply because when you're backtesting, you have that bias of knowing what's coming. Well, when you're trading live, obviously you don't know the future. And if anyone does know the future, please do reach out to me. I'm sure we could work on a great partnership. But what happens here is that your algorithm then struggles to deal with the real market, and the real market is something that you can't predict a lot of the time, and it's something that can do anything it wants to, you know. It's not set in stone that price moves from A to B, as it may be when you're looking at backtesting. And this is one of the things I think a lot of algorithmic traders really do struggle to grasp, the idea of creating an algorithm that excels in backtesting but equally excels in the real market itself. So this idea of overfitting is something that many algorithmic traders have faced, and it's something that they spend hundreds and thousands of hours even trying to overcome. And unfortunately a lot of traders do at the end of it find that they're unsuccessful with their strategies and they have to go back to the thinking board and devise a new approach to the market itself. So you've heard the pros and cons, I guess, of algorithmic trading and, and the struggles that you can go through when trying to create a strategy, and it really just does raise the question: is algorithmic trading for you? It really depends. I wouldn't recommend algorithmic trading to anyone who's new or not very familiar with coding. Alongside coding, you should also know how to work with network systems, since you'll have to maintain and, you know, keep updating and optimizing this algorithmic trading strategy to the point where you'll be working with a variety of different technologies and tools. As with anything in technology, there's a lot of things that can go wrong. You could have a server go off, your internet could cut out while you're in a position, and these are the things that you need to be able to maintain and monitor and constantly look after and keep updating and, you know, maintaining this algorithmic trading robot or whatever it is, because at the end of the day, it is a piece of software, and software needs updates and needs someone to look after it, essentially. On top of that, you need to have a good understanding of how markets work and trading strategies to find the best and optimal fit for the algorithm you're planning to create. It really does give you that edge over other algorithmic traders. And it's definitely something I think a lot of computer science students or, you know, anyone with a coding background who might be trying to create an algorithmic strategy might be overlooking. So that's definitely something to consider as well. So that was a very quick look at algorithmic trading, just discussing some pros and cons about what it is and the current state of it in markets today. If algorithmic trading is definitely something that interests you, please do reach out to Amplify Trading. To see our new algorithmic trading simulation and definitely find out some details on how you could potentially get involved with a simulation like this. We usually use Python in the simulation, and we go through testing 3 different stocks over a 1-year period to see if you can beat the AmplifyTrading benchmark, which is set by myself. So apart from that, guys, thank you very much for watching the video today. I really hope you have enjoyed it and learned something today. As usual, please do like, subscribe, and comment on the video. I'll definitely be around in the comments answering some questions, um, related to algorithmic trading. So please, we do appreciate all the support as usual, and thank you for watching this video today.