High-frequency trading

High-frequency trading (HFT) is a short-term trading strategy that aims to capture small profits with large position sizes. It affects all market participants, whether they themselves are high-frequency traders or not. Many of the orders that are executed in a marketplace, plus the bid-ask spreads​​ that are seen, are the result of high-frequency traders.

Placing several orders a second, let alone hundreds or thousands, is outside of manual human capabilities. Yet a computer, instructed through a program to place orders or buy and sell in various markets, such as stocks, commodities, currencies and bonds, can easily accomplish this task. That is what HFT represents: primarily computer-driven bids, offers, and transactions with the programmed intent of capturing small profits from the financial markets​.

HFT firms are companies that specialise in this form of trading. Since they trade so frequently, all traders have likely transacted with a HFT firm at some point.

Traders cannot usually detect HFT because it happens at such a high speed, where the algorithms can pick up on trading signals and execute multiple orders within a fraction of a second. Some high-frequency trades you spot are often illegal. For example, you may see large orders being posted on the bid or ask in an attempt to manipulate the market​ price, yet these orders are cancelled before they are filled.

You may also get a better price on an order than expected. This tends to happen more in the stock market​ than other markets. This could also be high-frequency traders trying to step ahead of other market participants.

HFT is algorithm based. The algorithm is designed by a programmer to take advantage of profit opportunities in the market. Here are several of the profit opportunities that it may look for:

Dark pools are private exchanges where market orders are not posted publicly, unlike typical orders that appear on the order book of any market. Dark pools allow institutional traders to transact in large quantities of securities without affecting the orders on the book. Orders on the book control the price, but there is often a limited quantity of securities on the order book at each price level.

High-frequency traders can use dark pools to attain or dispose of their financial instruments when possible. For example, if an algorithm can buy on the bid and then sell to a dark pool at the midpoint, they net half the bid-ask spread, less their fees. If an algorithm has accumulated a position, a dark pool may provide an easy exit that does not affect the price.

What is referred to as high-frequency trading in the media involves large amounts of capital, direct access to exchanges, vast computing power, often co-location, fibre-optic internet connections, and PhD-level programmers and researchers. These requirements are similar to quantitative trading​.

HFT is usually reserved for institutional investors, such as our CMC Connect platform. Retail investors can execute automated trading strategies. An automated strategy places trades quicker than a human and can be programmed based on any rule-based strategy. A trader can write code to create an Expert Advisor (EA) or application programming interface (API) that connects to their trading platform and trades on their behalf. This is not on a similar scale to high-frequency firms, but it is a similar alternative.

Artificial intelligence (AI) is a disputed part of HFT. While learning algorithms are prevalent in many sectors of the economy, the HFT community is split on whether this is beneficial. Traditionally, HFT firms have made money based on defined computations and strategies, often winning small profits with well-defined rule-based strategies.

AI is about learning and acting more human in a way. While this may work, it also brings with it the need for increased computer power and unknown risks when it comes to how the AI will learn and act. How will it act when it interprets something it has never “seen” before? This creates risks, as well as opportunities.

High-frequency trading is highly debated and charges have been levelled against many HFT firms for illegal activities. The argument for HFT is that, in most cases, it provides substantial trading volume and liquidity to the market. This means that retail traders are more likely to have someone to buy from or sell to when needed.

Despite this advantage, high-frequency traders often profit from providing trading volume. They can execute orders quicker than others, providing what some view as an unfair advantage. At the same time, HFT helps to keep markets in line by exploiting small price differences and bringing disconnected assets back into equilibrium.

A retail trader that is not interested in high-frequency trading simply needs to develop a trading strategy that looks at a slightly longer timeframe. Consult our article to the most common and effective trading strategies​. Traders can deploy scalping strategies where trades last several minutes, or trend following strategies where trades last minutes or weeks. These strategies are not typically in direct conflict with HFT. Momentum trading involves jumping into assets that are moving strongly. If the price is soaring, it does so regardless of the HFT; therefore, retails traders can capitalise.

With all strategies, there is also risk. No matter what method of trading is employed, stop-loss orders​ can be placed to help control this risk and manage position size, so that if the stop-loss is triggered, it only results in losing a small portion of the trading account. However, stop-losses are not always effective, even GSLOs, therefore other risk management practises should also be taken into consideration.