With a flood of stock orders hitting the tape in the last few minutes of trading each day, Deutsche Bank’s Autobahn Equity business has rolled out a new algorithm in US markets to help institutional clients and other market participants that seek to match closing prices.
The new algorithm TargetClose has also been designed to keep up with heavy market data volumes that occur in the last few minutes of the day as well as help the buy side manage their workflow challenges.
“When you look at market volume profiles, a lot of volume is concentrated at the end of the day or even in just the last few seconds,” said Jose Marques, Global head of Electronic Equity Trading at Deutsche Bank in an interview with Advanced Trading. “It’s become very hard for a human trader to participate in the last few minutes because the quote rates and amount of market data is enormous. If you are manually managing several market orders at once at the close, workflow issues are significant,” according to Marques.
The bank said that TargetClose is designed to project each day’s closing volume based on real-time and historical market data. It continuously adapts to market conditions in order to minimize the trade’s deviation from the closing price on the date of the trade. The algo is underpinned by Deutsche Bank’s technology, including integration with its ultra-fast smart order router, and takes advantage of low-latency models, and participates in dark pools to improve performance, the bank added.
Buy-side traders know that significant volume is done at the end of the day, so a lot of these traders may execute via an algorithm all day long holding back 10-to 20 percent for manual execution near the close as a way to add human value, said Marques. “Having tools for them that are optimized to deal with the problems in the last few minutes of the day are in demand,” he said.
There is demand for end-of-day prices because a lot of institutional traders are explicitly benchmarked to the closing prices and they execute their baskets to match the closing prices. There are also players who need the closing prices or who do index rebalancing and issuers of ETFs who are hedging trades that need the closing prints.
While other Wall Street firms offer this type of functionality, Marques said the unique part of this particular algorithm is the implementation and how it handles the technology issues created by high volume and high data rates just a few minutes before the market close.
“Most of our value add is around the quantitative analytics and those factors are certainly built into this algorithm, but more than anything dealing with the high market data and order entry rates around the close is the biggest challenge,” Marques maintains.
“We’re looking at the depth of the order book, how quickly orders are being completed, and applying HFT style (high frequency trading) models to orders and executions, “ he said. “Being able to do that while keeping our systems and platforms scaled properly for an institutional product is quite challenging,” Marques said, noting that the firm has thousands of institutional clients who want to trade at the same time. “For an institutional algorithmic tech platform, it’s a much more challenging exercise.
On top of launching the new algorithm, Deutsche Bank said it hired three sales traders underscoring its commitment to the electronic trading business. In the first week of September, Ryan Lee will join the bank from ITG where he has about 10 years of experience. Rob Lippert, who is starting in August, is joining from Goldman Sachs with 14 years industry experience from both the buy side and sell side. In June, Scott Chinitz, came onboard from RBS, with 12 years industry experience on both the buy side and sell side All three report to Jay Fraser, Head of Autobahn Equity Sales & Execution, Americas. In spite of the six billion shares a day volumes, said Marques, who joined Deutsche Bank two-and-a-half years ago to rebuild its electronic trading platform, said the bank is continuing to invest in the platform. After investing in the algorithmic products, and the underlying technology Marques said, “We think the next stage of the investment will be to expand our bespoke, high-value coverage for electronic trading.” The expansion also reflects growth in the firm’s business. “We need more capacity because our market share is growing despite the low volume.”
