Testing Quantum Advantage in High-Frequency Trading with Market-Derived Nonlocal Games
This paper proposes a framework for testing quantum advantage in high-frequency trading by deriving nonlocal games directly from historical market data rather than theoretical models, finding no quantum-classical gap in the initial two-agent, three-stock arbitrage scenario but demonstrating the method's ability to identify and validate potential advantages under specific market conditions.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the author. For technical accuracy, refer to the original paper. Read full disclaimer
In the high-stakes world of finance, speed is everything. Traders use automated systems to buy and sell stocks in fractions of a second, racing to capture tiny price differences before they vanish. This is the realm of high-frequency trading, where computers act as the primary agents, reacting to market data faster than any human could. For decades, scientists have wondered if the strange laws of quantum physics could give these machines an edge. Quantum physics allows particles to become "entangled," a connection where the state of one particle instantly influences another, no matter how far apart they are. This phenomenon, known as nonlocality, has long been a subject of deep theoretical study. Researchers have proposed that if two trading agents could share this kind of quantum connection, they might coordinate their actions better than any pair of classical computers could, even if they cannot talk to each other during the trade. The idea is tantalizing: could the universe's most fundamental weirdness be the key to making money in the most competitive market on Earth?
A team of researchers from IonQ has taken this question out of the realm of pure theory and into the messy reality of actual market data. Instead of inventing a fake market to test their ideas, they built a new kind of test using real historical records of stock trades. They focused on a specific scenario involving two automated traders and three different stocks, where the traders compete to buy and sell at slightly different prices across three different exchanges. In this setup, the traders must decide whether to act on a price difference without knowing what the other trader is doing at that exact moment. The researchers used twenty specific trading days of real order-book data to create a "game" that mimics the actual conditions these traders face, including the fees, the speed of data transmission, and the way liquidity dries up when multiple traders try to buy the same thing at once.
The first step was to see if a quantum advantage was even possible in this real-world environment. The team calculated the best possible outcome a pair of classical computers could achieve by trying every possible combination of decisions. They then compared this to the theoretical maximum a quantum system could reach. When they ran this analysis on the market data from the twenty highest-VIX days spanning August 2024 to March 2026, the result was clear: there was no gap. The best classical strategy performed just as well as the best possible quantum strategy could have. The market conditions as they naturally occurred did not offer a situation where quantum coordination would provide a benefit. The researchers found that the specific interactions in the real market, such as the cost of competing for the same stock shares or the way prices shift between different stocks, were not strong enough to create an opening for quantum mechanics to exploit.
However, the story did not end with a dead end. The researchers realized that while the real market didn't show a gap, it might be because the interactions were too weak or too rare. To understand what would be required for quantum advantage to exist, they created a controlled version of the game. They took the same market data but artificially strengthened the effects of competition and price changes, essentially turning up the volume on the market's friction. In this modified environment, they searched for a specific quantum strategy that could outperform the classical one. They found that when the penalties for competing for the same stock were high enough, and the rewards for coordinating on different stocks were significant, a quantum strategy did emerge. This strategy, which relies on entangled particles to coordinate decisions, beat the best classical approach in the simulation.
The final and most rigorous test was to see if this theoretical win could survive the transition back to a realistic trading simulation. The team took the quantum strategy they had discovered and ran it through a full end-to-end backtest, simulating the exact process of sending orders to the market, dealing with partial fills, and managing the timing of trades. They compared the quantum strategy's performance against the classical one on the same modified data. The results showed that the quantum strategy maintained its advantage. It earned more profit than the classical strategy, proving that the theoretical gain was not just a mathematical artifact but something that could persist through the complexities of actual trade execution. The researchers noted that on the most active days, the realized profits were slightly lower than the theoretical predictions, a discrepancy caused by the simulation's rules regarding stale data, but the quantum edge remained clear.
This work establishes a new method for testing the potential of quantum technology in finance. By grounding the test in real market data rather than idealized models, the team showed that quantum advantage is not a guaranteed feature of all trading scenarios. In the specific cross-venue arbitrage setting they studied, the natural market dynamics were too subtle to allow a quantum edge to emerge. Yet, by systematically adjusting the conditions, they demonstrated that there are regimes where quantum coordination could be superior. The study does not claim that quantum trading is ready to be deployed tomorrow, nor does it suggest that current markets are ripe for quantum disruption. Instead, it provides a roadmap: a way to identify exactly which market conditions and trading strategies might one day benefit from the unique properties of entangled particles, turning a theoretical curiosity into a practical tool for the future of finance.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.