Raising the Bar on Venue Analysis
A player with a .300 batting average seems like an attractive pinch-hit option, late in a tie game with two on and two out.
But consider who’s sitting next to him. That player is a .260 hitter, but he hits .320 against left-handed pitchers, his on-base plus slugging (OPS) is .800, and he has four hits including a home run in 10 lifetime at-bats against the ace southpaw on the mound.
It’s an easy decision for the manager. Player B gets fewer hits per at-bat overall, but because he has a better chance of driving in runs in this specific situation, he gets the call.
Such granular, sabermetric-type analysis is becoming available to institutional equity traders, who need help deciding which exchange or dark pool to route orders to. The end goal — trading’s version of runs scored — is efficient execution with minimal slippage; as in baseball dugouts, aggregate data is no longer good enough on trading desks either.
“Traditional venue analysis ‘buckets’ all the executions on a particular venue and grades that as a whole,” said Joe Wald, chief executive officer at trading-technology provider Clearpool Group. “That yields inconclusive and imprecise information. A better, and differentiated, approach is to look at the intention of the trader, and then give a score based on what that intent was.”
Wald said trade performance on venues is best measured from a ‘phase’ perspective, where each phase represents a unique trader intention. Examples of phases include taking liquidity from lit markets, taking liquidity from an inverted market, executing at the midpoint, resting passively, or using a conditional order. A venue that shows superior executions at the bid-ask midpoint would be best for a trader with that intent, even if that venue’s overall execution data are ordinary.
Aside from measuring by phase, the new iteration of venue analysis factors in a broad set of data that includes peer executions. This goes beyond the current standard of analyzing trades only against your own firm’s results.
“You need a significant amount of data to be able to really understand what’s happening with your execution,” said Clearpool President Brian Schaeffer. “Being able to compare what you’re doing to what the peer universe is doing gives you a differentiated level of context, which helps you make better decisions.”
As algorithms are ubiquitous in today’s electronic markets, algo intent is a critical input for venue and routing analysis, according to Linda Giordano, CEO of Babelfish Analytics. Clearpool is one of 25 brokers whose data Babelfish uses to provide routing, venue and transaction-cost analysis for its buy-side and sell-side clients.
“There are a lot of moving pieces that lead to the taking or posting of liquidity. It’s a very intricate process,” Giordano said. “All of those pieces need to be measured, monitored and incorporated back into the algorithm, because it will have an impact on the ultimate trading cost.”
For the buy side, “one of the challenges can be getting brokers to implement changes based on analysis,” said Giordano. “A ‘self-serve’ model shortens that cycle and allows firms to proceed directly to hypothesis testing.”
MiFID II Tie-In
Another advantage to measuring venue performance by phase is that the rigor of the approach clears the bar for more stringent best-execution mandates, such as what is called for under MiFID II, the European rule set that will take effect in January 2018. “It gives you enough order-trail data to understand and justify where and how your orders were executed,” Schaeffer said.
Assessing venues by trade intent is helpful for the assessees as well as the assessors. “Just saying a venue is ‘good’, or ‘bad’, isn’t fair,” Schaeffer said. “It’s better to say that venue’s not great when you try to take liquidity, but it’s pretty good when you’re resting, or when you’re in the midpoint, or whatever you’re doing. That’s what’s important.”
Venue analysis is an emerging field with no true industry standard, noted Richard Johnson, vice president of market structure and technology at Greenwich Associates.
Time-pressed buy-side traders might review venue analysis only quarterly, Johnson noted. But as the methodology evolves and its utility expands, it needs to be a more regular consult.
“As we’ve seen in ‘flash crashes’ and so forth, liquidity can move from venue to venue in the blink of an eye,” Johnson said. “Something may show great venue-analysis statistics for the last week, or the last day, or even the last hour, but that doesn’t necessarily mean that same liquidity is going to be there for the trade you’re trying to do right now.”
Factoring in measures of trade intent, such as how active or passive a given order is, would be a helpful bake-in to venue analysis, Johnson indicated. More broadly, venue analysis needs more clarity around metrics to move forward in terms of its usefulness for trading desks.
“A term that’s often thrown around is toxicity — obviously the less toxicity, the better,” Johnson said. “But toxicity is not a well-defined term, so people look at other metrics such as fill rates and reversion. I think in general, the industry needs to come up with better standards and better definitions around venue analysis.”
The new offering sports 11 liquidity-seeking algorithms with more to come.
GMAG says vacation season is a good time for design, implementation and testing of machine-based trading.
A bank will be able to recommend specific changes to clients to improve their access to liquidity.
Tech vendor will support Canadian equities trading and interlisted securities trading via its AMS.
A target-close model allows traders to mitigate market impact, Deutsche Bank's Andrew Royal writes.