Data Message Rates
As predicted, quote message traffic in U.S. stocks has surged as a result of Reg NMS, which forced the exchanges to automate their market centers and link to one another. This is leading market participants to conduct more quantitative trading in listed stocks, just as they've been able to do in Nasdaq-listed stocks, generating more electronic order, cancel and replace messages, and quote updates.
After the New York Stock Exchange launched its Hybrid platform, "all the regionals came out with the new trading systems and new players entered the market," such as the CBOE Stock Exchange (CBSX) and the ISE, notes Manisha Kimmel, executive director at the Financial Information Forum. "And then Reg NMS happened. That means those exchanges had to route to each other. And wow -- you see the slope of the line [skyrocket]," Kimmel adds, referring to a chart of listed message traffic.
According to the FIF, from May 2006 to February 2007, the one-minute peak message rates on the Consolidated Quotation System (CQS) -- the feed that disseminates data for AMEX- and NYSE-listed securities -- increased 79 percent over the nine-month period. Once the exchanges rolled out their automated trading systems on March 5, there was a more dramatic increase in quote message traffic. During the four-month period between February 2007 to June 2007, one-minute peak quote message rates increased 172 percent. (see chart, page 16).
In May 2006, the rate for the CQS feed was 3,860 messages per second (mps); by June 2007 it was up to 18,823 mps, relates Kimmel. "It's a paradigm shift," she remarks.
Listed stocks generated 24,049 mps in June, as compared to 6,900 mps in January of this year. "Everyone for a long time predicted explosions in listed trading. That's not happened," asserts Mike Traynor, chief strategy officer at the National Stock Exchange (NSX). "Our volumes are in the 40-million-share range for all stocks, but we are processing an inordinate amount of quote traffic," he says. Quotes have "increased at a rate more than 10 times the actual trading volumes."
While exchanges have invested in excess capacity to keep up with the message volumes, brokers and hedge funds are concerned about distributing the tidal wave of data messages into their internal applications and trader desktops. "We are all grappling with the number of messages," says a sell-side institutional trader who requested anonymity. "What's happened on the NYSE is that all the stat arb funds are able to trade stocks electronically, and you have many more people changing their bids and offers."
Clogged Pipes
The bottom line is that with so much data flowing through the pipes, quotes could be old by the time they reach internal applications, creating latency in making trading decisions. "By the time you see it, it could be a second old, it could be a minute old," suggests the sell-side trader.
"The increase in data tends to mean more latency," notes Peter Lankford, director of the Securities Technology Analysis Center (STAC), which operates testing labs in Chicago and New York.
On top of this, quotes are updating so quickly that additional capacity is needed to process them. For example, in July IBM's quote was updating 60 times a second or 3,600 times a minute, observes the sell-side trader. That equates to more than 200,000 updates per hour in just one stock. "We buffer the data -- though we have to process all the data, we don't display it all to the desktop," he explains. "The numbers are astronomical, and as more players come in that want to do stat arb and quantitative trading, you can get to the point at which you're dealing with 100,000 updates a second."
Meanwhile, the equities message traffic is dwarfed by the explosive growth in options data coming through the Options Pricing Reporting Authority, or OPRA, feed. On July 10, OPRA revised its traffic projections to take into account the addition of Nasdaq's options exchange to the feed late in the third quarter or early in the fourth quarter of 2007. Now OPRA is telling firms to prepare for nearly 1 million mps by July 2008.
Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio
Tweet This