OU blog

Personal Blogs

A picture of Russell Larke

High-Frequency Trading, Dark Pools, and the Fragmentation of Market Infrastructure

Visible to anyone in the world
Edited by Russell Larke, Monday 26 October 2026 at 12:43

High-Frequency Trading, Dark Pools, and the Fragmentation of Market Infrastructure

TL;DR: Key Structural Takeaways
  • HFT is an Evolution, Not a Replacement: High-Frequency Trading firms perform the same inventory management function as traditional market makers, but at microscopic speeds, introducing correlated withdrawal during stress (e.g., Flash Crash).
  • Dark Pools Solve Market Impact: Alternative Trading Systems (ATS) and dark pools allow institutions to trade large blocks without revealing their intent to the public order book.
  • Manipulation as Structural Exploitation: Practices like spoofing, layering, and quote stuffing are not legitimate inventory management; they are exploits of the market's reliance on algorithmic quote data.

Abstract

As established in my previous analysis of The Market Maker as Infrastructure, the provision of liquidity is a business of inventory management, not directional prediction. However, the structural reality of that business has been fundamentally altered by technology. This article examines the evolution of market infrastructure through the lens of systems thinking. By analysing the mechanics of High-Frequency Trading (HFT), the hidden architectures of Alternative Trading Systems (ATS) and dark pools, and the structural exploits of spoofing and layering, we can understand how the pursuit of speed and anonymity has fragmented the market and introduced new forms of systemic fragility. This analysis forms part of my broader research into market mechanics and participant constraints.

1. HFT and the Speed of Inventory Management

In modern electronic markets, the traditional human specialist has been replaced by algorithmic trading. High-Frequency Trading (HFT) firms use sophisticated algorithms and ultra-low latency infrastructure to perform the same function—inventory management—at microscopic speeds.

The academic literature, notably Hendershott, Jones, and Menkveld (2011), shows that algorithmic trading has narrowed spreads and improved liquidity under normal conditions. However, from a systems perspective, HFT introduces a new structural fragility. Because algorithms compete on speed, they often withdraw liquidity instantly when faced with rapid adverse price movements. This dynamic was starkly illustrated during the 2010 Flash Crash, when HFT firms, programmed to manage risk, pulled their quotes simultaneously, creating a sudden liquidity vacuum that exacerbated the price drop. The infrastructure is fast, but its response to extreme stress is mechanical and highly correlated. As I discuss in my existing analysis of market maker moves and inventory, reading these dynamics requires looking past the price to the structural behaviour of the algorithms.

2. Dark Pools and Alternative Trading Systems (ATS)

While the public exchanges display the order book for everyone to see, a massive amount of institutional trading happens in the dark. Dark pools and Alternative Trading Systems (ATS) are private venues where large blocks of shares can be traded without the order being publicly displayed before execution.

From a systems perspective, dark pools serve a necessary function. If a large institutional fund wants to sell a million shares, displaying that order on a public exchange would immediately drive the price down (market impact). Dark pools allow them to find a counterparty privately. However, this creates a fragmented system with two parallel structures: the visible market (lit exchanges) and the invisible market (dark pools). The price discovery primarily happens in the lit markets, but the volume is heavily skewed toward the dark. This fragmentation creates arbitrage opportunities for HFT algorithms that bridge the gap between the two, further complicating the structural picture. As I outline in my guide on how to choose the best trading platform, understanding where your order is actually routed is a critical structural decision.

3. Manipulation: When Mechanics Become Exploitation

The reliance on algorithmic execution and quote data has led to specific forms of market maker manipulation. These are not legitimate inventory management strategies; they are structural exploits of the market's reliance on order book data.

Three practices are particularly common:

  • Spoofing: Placing large, non-bona fide orders on one side of the book to create the illusion of supply or demand, then canceling them before execution. The goal is to trick other algorithms into trading against the real (smaller) order on the other side. I have broken down the mechanics of this in my guide on what spoofing is and how it moves markets.
  • Layering: Similar to spoofing, but involves placing multiple orders at different price levels to create a false "staircase" of depth.
  • Quote Stuffing: Flooding the exchange with a massive number of orders and cancellations to slow down competing algorithms, giving the manipulator a millisecond advantage.
  • Wash Trading: Executing trades with oneself to create artificial volume. See my notes on the fake volume that traps traders.

These practices exploit the structural constraints of electronic markets. Understanding these mechanics is essential for identifying when order book data is genuine and when it is a constructed illusion. We track these dynamics in real-time within my community archive on market structure.

4. Conclusion

The evolution of market infrastructure—from human specialists to algorithmic HFT and dark pools—has increased the speed of liquidity provision but fundamentally altered the structural landscape. The market is now fragmented, driven by the mechanics of speed and anonymity. Understanding the dynamics of HFT, the function of dark pools, and the structural exploits of spoofing and layering is not an academic exercise; it is the only way to distinguish between genuine liquidity and the algorithmic illusions that characterize modern market microstructure.

References

  • Hendershott, T., Jones, C. M., & Menkveld, A. J. (2011). Does algorithmic trading improve liquidity?. The Journal of Finance, 66(1), 1-33.

About the Author

Russell Larke is a financial market structure researcher and systems thinker, currently completing an MSc in Systems Thinking in Practice at the Open University. He is the founder of Trading Beyond Charts.

About Me | ORCID | Wikidata

Permalink
Share post