While electronic trading has become a fact of life, many in the trading community are struggling to find a simple way to measure the effectiveness of their electronic execution strategies.

Increasingly, they have come to rely on smart-order routing (SOR) technology, which seeks to add a layer of intelligence to Direct Market Access (DMA), taking into account factors such as liquidity and execution costs.

“As the use of smart-order routers has increased, there has been a growing dissatisfaction with their performance,” said Paul Daley, head of product development at SunGard’s Fox River Execution Solutions. “They are looking for ways to measure performance in an elegant, simple way.”

Using DMA on a trading desk without the use of a smart order router is challenging. “Without an intelligent smart order router, operating effectively while dealing with a steadily increasing variety of liquidity pools and execution venues, incongruous execution pricing across venues and stringent execution requirements is almost impossible,” said Daley.

The superficial view of electronic execution as replacements for human traders has since given way to a more nuanced view.

“Initially, the attitude toward algorithms was to use them in a passive way to automate trading,” said Daley. “As buy-side traders have grown more sophisticated with their use of electronic trading, they are recognizing the need for more intelligent algorithms and smart order routers.”

Transaction cost analysis (TCA) is assuming a greater role among institutional head traders as they become more accustomed to the world of electronic trading, but standard TCA metrics such as benchmark performance and market impact are ineffective yardsticks for measuring the performance of smart order routers. Another consideration is the average execution price relative to the limit price of the order.

For example, when using a smart order router to buy stock at a price beyond the current inside offer, one would want to maximize executions at the current bid price before moving up in price. By measuring execution price versus limit price a trader can evaluate how effectively his SOR sources liquidity at each price level – a good SOR will consistently find liquidity closer to his arrival price, where a less effective router may consistently push the stock, resulting in an average execution price closer to his limit.

“There’s a convergence between smart-order routing and TCA,” said James Bryson, chief executive at Elkins/McSherry, a TCA vendor that’s a subsidiary of State Street Corp.

Elkins/McSherry covers over 24 million transactions, $7.2 trillion of principal and 375 billion shares of trading from 1,500 investment managers and 2,000 brokers worldwide.

“Our universe is a collection of trading data from institutional investors,” said Bryson. “We use this data to assess institutional trading for commissions, fees and market impact costs in 47 countries.”

The two critical metrics of an SOR’s performance are the average fill size relative to the displayed size, and the overall percentage of fills that meet or exceed the displayed size.

“Those two metrics have come to summarize the SOR experience,” Daley said.

For example, filling 1,000 shares of a 2,000 share order looks superficially bad based on the order size, but when evaluating SOR performance the market conditions need to be taken into context. If there were only 100 shares on the quote, then filling 1,000 using a SOR is impressive. On the other hand, if there were 10,000 shares on the quote, then only filling 1,000 is not good.

Demand for algorithmic execution will continue to grow as the rise in market fragmentation makes sourcing liquidity an increasingly difficult challenge.

“The movement from eighths to pennies, and from a single exchange to multiple exchanges, has resulted in a variety of venues and business models,” said Daley.

SunGard’s Fox River SOR is a customizable suite of smart order routing strategies that help firms source the most liquidity at the lowest cost in a highly fragmented marketplace.

By employing advanced technology that optimizes routing and low-latency connectivity, Fox River’s SOR dynamically sources liquidity across both displayed and non-displayed venues.

“The asset management community, especially large buy-side head traders, are in need of or have an interest in determining how and where their trades get executed,” said Bryson. “The way we address this is by providing a service that uses and focuses on FIX protocol data. FIX data fields, such as Tags 29, 31, and 851, are normalized and have granularity.”

In evaluating fill rates, traders need to take into account the “types of venues they are trading on, what percentage of executions are getting filled, and what type of consumption of the spread they’re experiencing,” Bryson said.