Build it better and they will come – trades that is.
Trading is now a highly complex and data-driven activity. U.S. equity markets comprise dozens of execution venues, each with different liquidity characteristics, order types, fee structures and latency differentials. Trading algorithms are an essential tool for navigating these seas of liquidity, but execution performance can vary significantly from one algo to the next.
The responsibility for Best Execution falls squarely on the shoulders of traders, who are being given substantial flexibility from portfolio managers. For 82% of orders, traders have full discretion on execution strategy. To determine appropriate execution strategies and algos, traders are relying on increasingly sophisticated execution data, which is now a vital tool on trading desks.
“Trading in this new environment both requires and produces vast amounts of data,” says Richard Johnson, Principal for Greenwich Associates Market Structure and Technology and author of Peak Performance: What the Buy Side Expects from their Algos. “The key to achieving peak performance with algos is knowing which data points matter most.”
The new Greenwich Report represents the most comprehensive study of algo key performance indicators (KPIs) to date and helps traders optimize algo performance by identifying and analyzing the most important components of execution quality. Nearly 100 buy-side traders and market structure experts reflecting over $6 trillion in AUM and over $800 million in commissions participated. It reveals how traders rate the importance of factors like impact, reversion, information leakage and adverse selection for liquidity-seeking, implementation shortfall, VWAP, and other types of algos.
Customized Algos
The increase in execution venues, order types and tactics, coupled with advancements in data analysis and TCA, has driven many brokers to customize the algos they offer. The buy side is keen to engage the sell side in more algo experimentation and customization – Approximately 60% of buy-side traders in the U.S. are involved in algo customization in some capacity. They are increasingly focused on how to generate better performance, with an appetite to dive into the details of venue selection, algo KPIs and algo customization. The goal: to gain more control over their order flow and optimize trading performance.
Measuring Success
The buy side is clearly focused on slippage vs. arrival price across different types of algo strategies, with Price Impact being the most meaningful KPI for better algo performance. Nearly one quarter of traders perform venue preferencing as part of their algo customization, based on the results of cost analysis, or other qualitative reasons.
“Brokers should feel more comfortable discussing algo enhancements and A/B testing,” says Richard Johnson. “After all, the objective is to improve performance for client orders, and this type of discussion and partnership could enhance the overall relationship whether or not clients are open to experimenting.”
