The forgotten instrument
For years, futures trading remained on the periphery of electronic execution innovation. Before the 2008 financial crisis, algorithmic development primarily focused on equities, with futures offerings often being a mere adaptation of existing equity algos. Following this, in the high-volatility post-crisis landscape, banks approached futures akin to OTC products, viewing them as an opportunity to capture bid-offer spreads through block pricing rather than as a market segment deserving dedicated execution research.
This lack of investment in futures execution tools left buy-side traders navigating an environment where traditional transaction cost analysis (TCA) models fell short.
The post-crisis shift
The period following the financial crisis saw the emergence of new players dedicated to refining execution in the futures space. Firms like Quantitative Brokers (QB) and BestEx Research disrupted the landscape by offering algorithms designed explicitly for the nuances of futures trading. Meanwhile, non-bank market makers gained a stronger foothold, contributing to improved liquidity and tighter spreads across many contracts.
Advances in futures margining efficiency, alongside persistently high OTC financing costs, further fuelled the buy side’s growing interest in listed futures. These structural changes created the ideal conditions for specialised execution tools to thrive.
Where legacy futures algos largely replicated equity-based logics, modern entrants built execution frameworks that factored in futures-specific liquidity dynamics. By integrating sophisticated order placement strategies, predictive analytics, and adaptive execution models, they significantly enhanced execution outcomes for institutional traders.
But are algos alone enough?
Despite these advancements, algo execution is only one part of the equation. The future of best execution in futures trading requires a seamless, intelligent transition between execution modes. Traders need access to automated execution processes that can shift dynamically between algos, request-for-quote (RFQ) and direct order book interactions.
A next-generation futures algo must do more than execute static strategies—it must replicate the decision-making process of a human trader. This means evaluating liquidity in
real time relative to historical data, understanding market microstructure shifts due to economic events, and leveraging relative value dynamics between correlated contracts.
Subsequently, a change in the current understanding of algos will occur, wherein they become more akin to agents— a single order may be filled across multiple modes, where a traditional algo is only one component as the agent responds to changing market conditions in real time.
At Liquidnet, the approach to execution extends beyond standalone algos to a holistic framework that integrates multiple essential components. The firm leverages BestEx Research to implement best-in-class futures algos, leveraging optimised execution logic to ensure efficiency. In addition, Liquidnet has developed pre-trade analytics in-house, allowing traders to assess real-time market conditions before execution. Experienced traders remain integral to the process, providing oversight and intervention when necessary to ensure optimal execution outcomes.
Beyond these elements, Liquidnet has built, across multiple asset classes, a robust liquidity network to offer deep and diverse market access. Equally, the firm plans to bring to bear its already establishedmachine and reinforcement learning capability to continuously refine execution logic, using historical performance data and real-time market feedback to improve decision-making. By combining cutting-edge technology with human expertise, Liquidnet aims to bridge the gap between automation and intelligent execution.
Moving towards intelligent automation
The futures market has long been underserved by traditional execution technology. While post-crisis innovation has improved algo performance, the key to the future lies in intelligent automation that enhances human decision-making rather than replacing it.
Liquidnet is committed to evolving futures execution beyond static algos, integrating real-time analytics, market-adaptive decision-making, and seamless transition capabilities to ensure traders achieve optimal execution outcomes. By shifting focus from isolated order book events within an algo execution to a comprehensive execution process, Liquidnet sees significant implications for futures TCA. They envisage audit trails close to natural language that will provide event by event commentary on an ongoing execution – the question ‘Why did you do what did at the time that you did it?’ will be answered alongside a TCA ‘tape’ that incorporates non-order book data, such as RFQ responses or implied price calculations.
In an era where liquidity is constantly shifting, execution decisions must be both data-driven and flexible. The next generation of futures execution technology must not only react to market conditions but anticipate them, facilitating the transition between automated efficiency and human expertise.