What’s Next for HFT Firms?

by Alessandro Ishak

Do high frequency traders (covered in a previous blog) harm markets?

High frequency traders are responsible for a large share of activity in modern markets, often estimated at around 50–60% of equity trading volume in major markets. This scale naturally raises concerns about their impact. The main criticism is that faster firms may disadvantage slower participants and create an uneven playing field.

Supporters argue that HFT firms provide substantial benefits. By constantly offering to buy and sell, they increase liquidity and narrow bid–ask spreads, making it cheaper for others to trade. In normal market conditions, this generally improves efficiency and price discovery.

The real concern tends to arise during periods of market stress, when some firms reduce activity to protect themselves. Regulators have responded with rules designed to improve transparency and stability. Overall, while fairness remains debated, most evidence suggests that high frequency trading improves everyday market functioning rather than harming it.

Why is secrecy necessary?

In high frequency trading, even very small advantages are valuable, and they disappear quickly once others copy them. If strategies, models, or systems were made public, competitors could replicate them, removing the edge that makes the business viable. Secrecy, therefore, protects intellectual property rather than hiding wrongdoing.

This is especially important because high frequency trading is a zero-sum environment. One firm’s advantage is often another’s loss. Revealing how trades are priced, how risk is managed, or how systems respond to market conditions would directly undermine profitability.

Secrecy also helps maintain market stability. Publishing detailed strategies could encourage imitation without full understanding, increasing the risk of poorly managed systems. For these reasons, discretion is not optional; it is essential to survival.

Machine Learning, and how it’s helping HFTs

Despite the attention machine learning receives, most high frequency trading decisions still rely on relatively simple, well understood models. Classical approaches such as linear models, probabilistic frameworks, and optimisation techniques are transparent and predictable, which is critical when decisions are made in milliseconds. Firms need to understand exactly why a system behaves the way it does, especially during periods of stress.

Machine learning is used, but selectively. It is often applied to tasks like pattern recognition, feature selection, or improving inputs rather than directly making trading decisions. Highly complex or opaque models can be difficult to control and risky if they behave unexpectedly.

In high frequency trading, reliability and interpretability matter more than sophistication. A slightly less accurate model that can be fully understood and trusted is often preferred to a black box that cannot be explained in real time.

The Big Picture

High frequency trading is often portrayed as mysterious or predatory, but at its core, it is a highly technical form of market infrastructure. Firms like Jane Street exist to price financial instruments accurately and provide constant liquidity, allowing markets to function smoothly at scale. They are not built around bold predictions or market opinions, but around discipline, risk control, and precision.

What makes these firms unusual is how closely they resemble applied science organisations. Ideas are tested empirically, systems are stress tested before deployment, and failure is analysed rather than ignored. Small edges are pursued carefully because even minor mistakes can overwhelm profits when trading at high volume and high speed.

This model rewards humility and consistency rather than bravado. Markets provide immediate feedback, and poor ideas are quickly eliminated. Over time, this creates cultures that value evidence over intuition and collaboration over hierarchy.

In the broader financial system, high frequency trading firms sit quietly in the background. Most investors never interact with them directly, yet benefit from tighter spreads and more liquid markets. While debates around fairness and speed will continue, these firms have become a permanent part of modern markets. Understanding them requires seeing them not as gamblers, but as engineers of pricing and risk.