ITG Launches Dark-Trading Algo12.07.2017
What’s old is new again in algorithms.
ITG has just rolled out the newest iteration of its Dark aggregation algorithm. The new twist is that this new version is not only designed to maximize dark block liquidity capture but also grab as much natural liquidity as it can. The firm says this upgraded Dark aggregator now snags 82% of ALL available midpoint U.S. block volume.
In a nutshell, Dark provides broad access to natural liquidity using an unbiased routing optimization across all major dark pools, and uses advanced segmentation strategies to find quality fills that minimize information leakage.
In speaking with Traders Magazine, Ben Polidore, Head of Algorithmic Trading Products at ITG, said Dark is built to help institutions trade large blocks of stock quicker and more efficiently. Citing his own data, Dark had client order sizes in the third quarter of 2017 that were 5.9% of average daily volume—64% higher than the industry average for dark aggregators.
Available now in North America, Dark now offers an enhanced workflow that provides traders with a new one-click block mode and control over order aggressiveness, order type usage, and minimum-fill thresholds with an intuitive urgency parameter. Traders have full transparency into real-time child order placement via Triton EMS or ITG’s web-based Prism tool.
“One of the most important issues facing global investors is the transaction cost created by unnecessary intermediation in our fragmented marketplace,” Polidore said. “Our goal with Dark is to consolidate the agency block market with an efficient and unbiased product designed to link buyside traders wherever they are.”
Also commenting on the rollout, ITG’s Global Head of Product, Brian Pomraning, added “the enhanced Dark algorithm demonstrates our commitment to investing in best-in-class execution, liquidity, analytics and workflow technology products to help our clients improve investment performance and increase operational efficiency.”
How long did it take to revamp Dark?
Polidore said ITG spent about a year working on this new iteration, which included new research surrounding information leakage and how to minimize it as well as new software development. For the research aspect of it, ITG wrote a new investigative paper that gauged leakage, looked at better ways to minimize it and how to incorporate the findings into Dark’s logic.
In its research paper, “Put a Lid on It,” ITG uncovered that Information leakage in the trading process is a major concern for traders. In a poll, it was revealed that 37% of buy-side traders attributed more than half of their trading costs to information leakage. The paper explained It is difficult to attribute information leakage to a single venue because most modern routers, ITG’s included, create routes to many venues proximate to any price move. Since it is possible to have information leakage without a fill—the most obvious example being lit quotes on exchange—any of the venues used before a price move could have been the cause.
So, what do you do?
ITG built a feature into its dark router to disable a random dark pool for the first 30 seconds of a parent order 70% of the time. This creates a control group (30% of orders are unchanged) and several test groups, one for each dark pool accessed. The study selects only orders that are large enough to be eligible to trade to every dark pool it routes to. Using the data from ITG’s randomized controlled trial, then a logistical model is created to estimate the probability that the price would move away from the parent order arrival midpoint price by more than one-half of the spread in the first 10 seconds of the order. After fitting the model, ITG can compare the probability of the control group with that of each test group and look for statistically significant differences.
“We think randomized controlled trials are an important tool for measuring complex strategies,” ITG wrote in the paper. “This type of analysis allows us to find more meaningful relationships between tactical choices and parent order performance. In some cases, such as venue-based adverse selection analysis, the results of a tactical study can even conflict with the results at the parent order due to bias or second-order effects. We plan to continue running this experiment and others like it to rigorously search for improvement in our own strategies and how we interact with our counterparties.”
Dark focuses on interacting with quality sources of liquidity by continuously monitoring and limiting information leakage through large scale micro experiments which inform the algorithm’s routing logic, Polidore explained.
And Dark isn’t just all about the block, Polidore added. It also is designed to help traders execute smaller fill sizes. The central tenet is to find natural liquidity, he said.
This latest version of Dark could also be deployedin Europe and Asia, ITG said. For the European version, Dark will be tailored and comply with all MiFID II market structure requirements.
Instinet's Anushree Laturkar says investing in research and analysis is key to improving client outcomes.
AI can optimize algorithms to better recognize and adapt to real-time market conditions.
Due diligence process has not kept pace with the proliferation of algorithmic trading.
By Greta Zhou and Andy Cheung, APAC AES, Credit Suisse
Broker's new platform generates execution strategies that leverage client-specific trading data.