Algorithmic trading: tips, tricks, and how-to’s
When it comes to cryptocurrency trading, and trading in general, algorithmic trading plays a vitally important role. Within the world of trading, everyone from manual traders to exchange owners are affected by trading algorithms. With crypto-trading, the same applies. Exchange owners, crypto startups and projects, and of course crypto investors are all heavily influenced by trading algorithms and their impacts on the market.
In this article, we’ll dive deeper into the concept of algorithmic trading, exploring the place that it holds within the cryptocurrency trading world. More importantly, we’ll provide some practical insights into algorithmic trading – such as the different types of algorithmic trading in crypto, and how to spot a good algorithmic trading fund.
So, let’s start with the basics.
What is algorithmic trading?
Algorithmic trading is the use of process- and rules-based algorithms to employ strategies for executing trades. Algorithmic trading lets the trader define a rule set based on what trading should happen and with that the algorithm will monitor the markets, check whether the rules apply and send orders to the market for execution if the defined conditions are met. By automating trades, traders are freed from constantly needing to watch markets and executing orders manually. Because of the vast benefits of algorithmic trading at observing the markets and acting rapidly, it has completely transformed trading in the last decades.
Today, algorithmic trading is widely used by both retail and institutional traders, from investment bankers to pension funds and hedge funds. Astonishingly, up to 80% of daily trade volume in the U.S. is driven by machines and algorithmic trading engines, according to Guy de Bloney, a fund manager at Jupiter Asset Management.
Why are algorithms superior at trading?
When looking at trading in general, it’s best to see it as a 3 step-process. That way, we can better understand and evaluate the pros and cons of algorithmic trading.
Step 1: Collecting data
Since the financial world has transformed to be an almost entirely digital industry, it is obviously beneficial to use automated systems to collect and store large amounts of market data or any other data necessary to support the trading decisions. Without machines, the data collection would be slow, error-prone and manually operated.
Step 2: Making trading decisions
Decision-making about which trades to pursue form the most difficult part of the three-step process, since many types of trading strategies can be based on a large amount of input factors and variables. As a result, deciding on a trade is not something which is necessarily quantifiable, or codifiable in a way that a computer could understand. Because of the more abstract nature of this step in the trading process, we generally see this as a step in which the best traders truly excel, and where they can sustain a larger trading edge over their competitors. The question is, how does one quantify gut feeling?
By reading a trading book or taking a trading course, you’ll likely find a list of trading rules which come across as the accumulation of experience and wisdom by a particular trader or group of traders. Often, you’ll come across lists like “10 rules of trading”, which are filled with maxims like “Do not listen to your emotions”, “Don’t follow the herd”, or “Always stick to your trading plan”. As humans, we’re often prone to follow our instinct and our impulses, which makes it all the more harder to truly follow such advice. A machine, on the other hand, can be programmed to obey such rules, and will do so by default.
Therefore, a properly quantified and correctly-implemented trading strategy is the best version and closest thing to a trader who would rigidly stick to his own trading plan, and to the “universal” laws of trading. Essentially, an algorithmic trading bot acts as a pro-trading replica that operates around the clock, does not make mistakes, and is lightning fast in terms of execution. The downside of this, however, is that such an algorithm is not able to point out a potential error in a trading plan, like a typing error, or a corrupted data entry. This can sometimes lead to catastrophic mishaps, or losses that could otherwise have been avoided through timely human intervention.
Step 3: Executing the trade
Once a trading decision has been made, either by a human or machine, a corresponding order is sent to the exchange on which the order will be executed. With digital exchanges, this step is also highly automated, and the advantage of machines are not questionable here. When the latencies (the time that elapses from the moment a signal is sent to its receipt) are under a second, or when dozens of exchanges should be used or thousands of orders sent out as quickly as possible, machines are the only way to execute trading orders efficiently and at scale.
It’s important to keep in mind at this point that many algorithmic trading strategies exist, which will be explored below. Some types are more efficient, modernised versions of older trading strategies, such as arbitrage. Others, on the other hand are highly quantitative and either impossible to be executed without computers, or highly manual and labour-intensive.
Main algorithmic trading strategy types
Here’s a simplified overview of the main algorithmic trading strategies that are commonly used:
At its most basic, arbitrage can be defined as the purchase of an asset in one market, and the selling thereof in a different market, in order to take advantage of a difference in price.When a trader uses arbitrage, they are essentially buying a cheaper asset and selling it at a higher price in a different market, thereby taking a profit without any net cash flow.
According to the statistician Francis Galton and his concept ‘Regression of the Mean’, he argues that extreme events are generally followed by more normal events. In other words, things tend to even out and stabilize over time. In line with this, a mean reversion trading strategy involves betting that prices will revert back towards the mean or average. A simplified example of a mean reversion strategy in action would entail, for example, buying a stock after it has had an unusually large fall in price.
Scalp trading (“or scalping”) is where a trader aims to make multiple short and small trades, with the goal of profiting off of a stock’s small movements. These trades often last from seconds to minutes, resulting in tiny profits with under a percent return. With scalping, you’re not trying to catch trend movements. Instead, you’re looking to profit from the ups and downs throughout the day as orders from large traders and institutions push prices up and down.
Trend trading is a trading strategy that attempts to leverage gains through the analysis of an asset’s momentum in a particular direction, namely up or down – which is known as a trend. Trend traders enter into a long position when a security is trending upward. An uptrend is characterized by higher “swing lows” and higher “swing highs”. Trend trading strategies assume that a security will continue to move in the same direction it is currently trending. Such strategies generally contain a take-profit or stop-loss mechanism in order to secure a profit, or avoid big losses if a trend reversal or downswing occurs.
In technical analysis, a technical indicator is a mathematical calculation based on historic price, volume, or open interest information. Traders and investors use technical indicators to analyze the past, and to predict future price trends and patterns of a certain stock or asset. A growing number of technical indicators are available for traders to utilize, such as a moving average or the stochastic oscillator, as well as commercially available, proprietary indicators. A stochastic oscillator is a momentum indicator comparing a particular closing price of a security to a range of its prices over a certain period of time.
Momentum trading is a strategy in which traders buy and sell based on the strength of the recent price trends of a stock or asset, often in line with general hype in a market. Momentum investors generally take advantage of market volatility by taking short-term positions in assets going up, and selling them as soon as they show signs of going down. A great example of this was the Bitcoin bull run at the end of 2017, in which cryptocurrencies became a global phenomenon, and the price of Bitcoin soared to almost $20 000.
Algorithms in cryptocurrency trading
If we look at algorithmic trading within the cryptocurrency market specifically, it’s first important to consider that with crypto, the market is active 24/7, 365 days per year. It never sleeps. Apart from this, it’s equally important to consider where the crypto market is in terms of maturity. Due to cryptocurrency trading being so fragmented between multiple exchanges, the market itself is (to a degree) subject to the successful operation of such exchanges (think of Binance and Coinbase, for example). If a leading exchange gets hacked, or even if such an exchange goes down due to maintenance work, the market can be affected. This can have an impact on an algorithmic trading bot, which is subject to the platform’s up-time.
Similarly, cryptocurrency exchange platforms are still far away from providing the latency provided by traditional exchanges – meaning the time required for orders to be matched is slower between crypto exchanges and Wall Street exchanges, for example. The biggest determinant of latency is the distance that the signal has to travel or the length of the physical cable (usually fiber-optic) that carries data from one point to another. This means that trading strategies based on high speed latency are less applicable within crypto trading than in traditional financial trading markets.
How to spot a good algorithmic fund
Firstly, it’s important to consider that algorithmic funds do not share the details of their strategies, considering that such funds are tied to algorithms which are constantly optimized and monitored by teams of analysts and developers. As a result, it’s not possible to know exactly what kind of market circumstances a fund has been programmed towards. Therefore, a good indicator of a strong algorithmic fund is whether or not it performs well in different market types. For example, if the market is facing a downswing, a strong algorithmically-operated index fund for crypto assets should have a smaller drawdown than the market over the same period of time.
Another important thing to consider when distinguishing a strong algorithmic fund from a weak one is the fund’s risk profile. With this in mind, you should decide on your investment size based on a fund’s risk profile and benchmark. Depending on how risk-averse you are, and how you’d like to structure your own risk portfolio, you can determine how much capital you’d like to invest into a fund.
The Icoinic Algorithmic Fund
At Icoinic, our own algorithmic fund combines many of the above-mentioned elements to bring together a fund based on constant market monitoring, and the persistent optimization of returns for our clients. This is all made possible by our team of specialists with different backgrounds, such as years of experience in both traditional markets and in digital assets. It is this unique composition that enables Icoinic to approach the volatile market from a different perspective, which results in a competitive edge.