Block trading ATS provider Pipeline Trading Systems has launched what it claims is the first intelligent algorithm selection engine to be made available in the institutional marketplace. Using historical and real-time trading data, Pipeline’s Algorithm Switching Engine aims to translate a trader’s execution preferences into an optimal choice of algorithms and execution styles.
When activated by a trader, the so-called Algorithm Switching Engine will work in tandem with the existing Pipeline Block Market, applying predictive analytics to correct the inefficiencies associated with the static choice of a single algorithm on an order large enough to live through significant variations in market conditions.
“Our clients know that the most effective way to trade their larger orders is to leverage The Block Market’s 20% fill rate for mega-block trades,” says Fred Federspiel, president of Pipeline. He believes large amounts of natural liquidity does not always exist within traders’ time constraints, so they are often obliged to algorithmically execute very large block orders by participating in venues that are optimised for sub-block trading.
“Our clients have requested best-of-breed access to these sub-block markets without the opportunity costs of exiting the Block Market,” says
Federspiel. “ While it would have been relatively straightforward to marry Pipeline with a list of well-regarded algorithms, the larger orders served by Pipeline are inevitably exposed to widely varying market conditions, rendering any single algorithm choice suboptimal.”
He adds, “Customer demand is much better served by the Algorithm Switching Engine, enabling institutional traders to take advantage of proprietary predictive market intelligence to guide their participation in the retail-sized markets while fully leveraging Pipeline’s existing capability to deliver large natural trades – thereby achieving the upper limits of block execution performance in any market situation.”
“If we can’t get an order done in a large block venue like Pipeline, we’ll often utilise algorithms to complete the trade,” says David Brooks, director of global equity trading at The Boston Company Asset Management. “Pipeline’s unique approach – to automate switching between different algorithms – makes sense for large blocks with longer time horizons, as well as small residual orders. And since the Algorithm Switching Engine will work in tandem with the core Pipeline block matching functionality, I’ll get the best of both worlds. I look forward to taking advantage of this new capability.”
Pipeline says the Algorithm Switching Engine was created in conjunction with Adaptive Technologies, a provider of predictive analytics co-founded by Pipeline’s director of research Henri Waelbroeck, and with J Doyne Farmer, professor at the Santa Fe Institute, and co-founder of Prediction Company.
Farmer is a researcher in the complex behaviour of markets and modelling risk aversion, volatility and price formation areas. “While executing a large order algorithmically, I have found that it is absolutely imperative to update tactics to reflect changing market conditions,” he says. “Pipeline’s dynamic switching engine is the right approach to minimize market impact costs of trading institutional-sized orders in the retail-sized markets.”
The Algorithm Switching Engine is a component of Pipeline version 4.0, which is scheduled for release in the second quarter. Pipeline is now in its third year of operation. It offers an electronic platform where block traders can enter priced orders for large blocks of stock without the fear of the predatory practices that can occur in traditional trading venues.