Equities trading focus : Venue analysis : Ian Domowitz

“I’m shocked, shocked to find that [TCA] is going on in here!”¹

By Ian Domowitz*

When gambling is relabelled as just card playing or dice rolling, TCA may finally become Transaction Research. TCA has penetrated portfolio construction, fund capacity analysis, liquidation studies, fund NAV determination, trading strategy, and real-time decision support. It was a big surprise when I heard the comment, ‘venue analysis is not TCA.’

Venue analysis is an old theme. Exchanges fought over relative costs of execution, motivating academic work making the comparisons. With the arrival of electronic venues, efforts concentrated on explaining the mechanics of execution, followed by contrasts of execution quality on electronic and floor systems. Venue quality reporting via regulatory mandate appeared. The interaction of dark pools with routing schemes generated work on information leakage. Comparisons of lit and dark markets began. Flash forward to present day, and the furore surrounding Flashboys has generated collective amnesia with respect to our knowledge of venues over the past twenty years.2

Viewed through the lens of TCA, venue performance cannot be separated from trading strategy, market conditions, or even constraints on workflows. Venue analysis is most useful when set in the larger context of overall performance. Understood in that light, a better grasp of routing practices and performance permit traders to use the suite of tools at their disposal effectively, as well as generating more useful conversations with their brokers.

Trading strategy and venues

Consideration of trading strategy is an essential component in assessing venue performance. Without controlling for strategy, it is impossible to construct comparisons of individual venues, or contrast dark versus lit markets in the aggregate. Yet, the work we see on quality of execution by venue appears to ignore this simple TCA-inspired insight.

Trading strategies can affect venue analysis in several ways, given the profile of fills flowing back to the algorithm from routing information, and disparities in how strategies cycle orders into the markets. Routers may be tuned to favour certain venues depending on strategy. Venue types are characterised by different mixes of strategies which feed them.3

Some disparities across strategies are intuitive. Scheduled strategies favour lit markets, while dark aggregator strategies dominate in dark pools. The main point is that the distribution of strategy usage across venues differs, which in turn colours performance.

Fill size is a performance metric, but may also be a characteristic stemming from segmentation of participant types in any given venue. We cannot observe segmentation directly. Fill size is traceable to the nature of strategies, however, given the interaction of fills with sizes determined by the strategy itself. The disparity across venue types is illustrated in Figure 2.

The two boxes contain the top ten venues for each strategy by volume traded in our sample. Consistent with the evidence in Figure 1, not all venues appear in both boxes, and the ordering is different. The distribution of fills depends on strategy. A comparison of the same venue across strategies also is different. The distribution of fill size within the same venue depends on strategy. One cannot simply compare fill sizes across venues without taking the trading strategy into account.

Common performance metrics

If venue characteristics vary across strategies, performance comparisons also must be affected. Even the simplest performance statistic, average spread capture, varies across venues by strategy. Differences in spread capture for lit markets vary by as much as 45 percent in the US, for example. Dark market spread capture differs by 46 percent across strategy types.

In terms of the “classic” implementation shortfall measure, variation across venues is traceable to trading strategies, as opposed to distinct properties of the pools themselves.

In fact, the use of implementation shortfall for venue analysis, aggregating over strategies, is generally incorrect. Given the idiosyncratic nature of time stamps in trading systems, calculation of implementation shortfall cost is typically possible only when a broker strategy is considered. For this reason, many have resorted to reversion metrics, which depend only on a snapshot of price, at and after, an individual fill.

Reversion measures differ by the time after the fill. The idea is that if a great deal of reversion is exhibited in price over very short time horizons, then the fill is bad relative to what might have happened in the absence of toxicity in a venue. While price reversion may simply be representative of liquidity conditions, this metric is still widely used as a proxy for toxicity.

The differences across strategies, within and between types of venue, are evident. Within lit markets the reversion due to scheduled and opportunistic strategies is different by a factor of five. The difference in cost for dark pools across strategy types is as much as 200 percent. Most of such deviations jump right off the chart.

A last look

Space constraints prevent the illustration of results by venue. The differences in comparisons by strategy across individual venues are evident in an examination of the tails of the price reversion distribution, for example. There are large percentages of poor outcomes, in the case of implementation shortfall strategies, which do not appear in results aggregated across strategy types. The tail of the distribution for reversion in the dark, using dark aggregation strategies, is much smaller. Performance around zero reversion improves in that case.

Consideration of trading strategy is an essential component in assessing venue performance. This is true when attempting to compare and contrast different market structures, and also for assessing venue quality on a disaggregated basis, within market structure categories. The results suggest that proper “tuning” of routing functionality to strategy may improve performance. A proper treatment of this suggestion is a natural next, and constructive, step in venue research.

*Dr Ian Domowitz is Managing Director, Investment Technology Group, Inc. This article is based on joint work with Kristi Reitnauer and Colleen Ruane.

Footnotes

1. Apologies to Casablanca and the back room of Rick’s Café.
2. See “Garbage In, Garbage Out: An Optical Tour of the Role of Strategy in Venue Analysis.”
3. Algorithms are aggregated across brokers into several categories. Scheduled strategies include VWAP, TWAP, and Participation strategies. Dark denotes a liquidity-seeking strategy concentrating on dark pools, while opportunistic is general liquidity-seeking in nature. IS is shorthand for implementation shortfall, creating execution patterns based on cost and risk minimisation. Non-algo in the charts is generally desk trading, while Other is dominated by direct market access. The data used here is sourced by ITG, covering a spectrum of buy-side firms and their brokers, for the period 2013:Q2 through 2014:Q2.

©BestExecution 2015