Storm Copestand
Traders Magazine Online News

Conquering Fear in Trading

In this exclusive to Traders Magazine, therapist Storm Copestand examines how traders can manage expectations and conquer their fear during the entire execution process.

Traders Poll

Amid changes in builder, do you think the CAT project will be completed by 2020?

Free Site Registration

September 30, 2004

Building the Better Mousetrap?

By Peter Chapman

The ability to deposit an order in Pipeline for an entire day gives the system an edge over at least one established ATS, Berkeley claims. A trader can leave his order in this ATS all day, says the exec, but is unlikely to. That's because the users only receive progress reports following one of the periodic matches. With Pipeline, a trader always knows where he stands.

This communication aspect of Pipeline creates more of a "marketplace feel," says Federpsiel. "This is not just put an order in and pray.'"

Representing Algorithms For Chris Keith, the former

Big Board CIO and inveterate tinkerer, the past is prologue.

The man's latest invention - PDQ - is an outgrowth of his more ambitious plan of three years ago. That envisaged a virtual stock market where robotic floor brokers traded with one another. The scheme failed to gain traction, but the core technology, intelligent agents, is behind PDQ, an acronym for "Procedure Derived Quotes."

In Keith's latest vision, the agents, essentially search engines, represent algorithms in their search for liquidity. "Algorithms have become the dominant source of generating orders," Keith says. "It stands to reason that they would also be the dominant source of responding to orders if someone created a facility in which they could do that."

The use of algorithms, or computer-managed trading instructions, has exploded since decimalization. They can range from simple fill-or-kill orders to more complex order types that make decisions after analyzing reams of market data.

PDQ is essentially a very sophisticated example of the technology known as smart order routing. SORT seeks out the market center with the best liquidity and directs an order to that market center.

The difference here is that the liquidity is not an order sitting on a market center's book, it is an order en route to the book. Once the PDQ agent finds an appropriate order whizzing through cyberspace, it notifies its customer, an algorithm. The algorithm may then direct PDQ to create an order and forward it to the same destination.

The hope is that the algorithm's order reaches the destination first. That increases the chance that it will trade against the first order. PDQ does not guarantee its customers will match.

But "you could make the argument that the probability of getting filled is higher," says Andrew Shapiro, PDF's director of trading methodology, "when you are being notified of a liquidity opportunity rather than chasing quotes."

Shapiro points out that conventional smart routing systems simply notify the order sender (often an algorithm) of the location of the market's best quote - on Archipelago, for example. By the time the order reaches Arca, the quote may have disappeared. This has become increasingly the case in an age of "flickering" quotes.

Shapiro adds: "The difference here is that the best quote is about to get to Archipelago. So, why don't you go there? In all likelihood, your order is going to get there first."

The concept of using unexecuted orders as information is not new. Intelligent agents in human form are found on the floors of exchanges as well as upstairs trading desks.