Show Me The Data! Highlights from the CME Group Tech Talk 6.0

The CME Group brought their annual Tech Talk back to Chicago this year after stops in New York and London and the topic of the day was data. With presentations from the Chicago Cubs, Orbital Insights, Tellus Labs, and Google, the CME showcased the multiple ways that the data deluge is transforming our world.

Leading Off: The Most Data-Driven of All Sports

The session kicked off as Julie Armstrong of the CME conducted an one-on-one interview with the Executive Vice President and General Manager of the Chicago Cubs, Jed Hoyer. Baseball has always been stats-driven but advancements over the past ten-plus years have had a profound impact on the sport and Hoyer has had a front row seat. He got his start in baseball with the Boston Red Sox and came to the Cubs along with Theo Epstein. The rest, as they say, is history as the Cubs snapped their 108 year World Series championship drought in 2016.

Interestingly, there are many parallels between baseball and professional trading in terms of past and current developments. Specifically,

  • Early technology adopters had a distinct advantage. In the case of baseball, the Red Sox were one of only three teams that conducted in-depth, statistical analysis when Hoyer joined them and they certainly reaped the benefits. For example, they were able to sign future stars David Ortiz, Bill Mueller, Kevin Millar, and others on a $10 million budget. That simply wouldn’t be possible now as all teams have added deep analytics capabilities. Similarly, early movers to electronic trading enjoyed big gains and recent results are decidedly more mixed as former industry leaders such as KCG (Getco) and RGM have been acquired or merged with competitors.
  • Insights gained from data is increasingly fleeting. Before the advent of video replay, they used to keep track of the calls that umpires made on stolen base attempts and were more likely to attempt a steal  when certain umps were covering second. Currently, specific approaches to pitching to a player or pitcher/hitter match-ups are quickly counteracted as the other side reacts to your moves. In trading, the life of a trade is more and more short-lived as the market “learns” and adjusts to new strategies and tactics.
  • The role of people has shifted. In the old days, baseball scouts would time how fast a pitcher delivered the ball to home plate or how quickly a shortstop could turn a double play. All of that information is readily and more accurately available from digital sources these days, so the role of human scouts has changed. In baseball, this means that scouts now focus on work ethic and attitude with players while in trading the role of a “trader” has exploded into quant, developer, and trading functions.
  • The data keeps on coming. Hoyer said that the Cubs gather data that they don’t even have a use for yet because they know that they may well find a value for it in the future. The same is even more true in markets as alternate data sets are vastly expanding and trading firms are aggressively mining data sources searching for the next edge.

Hoyer had a difficult time breaking into baseball because he had only played D3 ball at Wesleyan but the Red Sox took a chance on him and hired him as an intern when he was 28 years old. In an auspicious twist, he once interviewed for a trading job with Will Hobert of WH Trading. Fortunately for Cubs fans, he didn’t get the job.

The Impact of AI on Commodities Markets

Artificial intelligence along with geospatial imaging, Internet of Things (IoT) sensor data, and the power of cloud computing are leading to a revolution in commodity market analysis. David Gresky of CME led an informative discussion with John Han from Orbital Insight and Tellus Labs’ David Potere gave concrete examples of how AI is already changing the markets. For example, Han described how satellite data has helped to see through the fog of incomplete or unverifiable data on the Chinese steel industry and Potere described how Tellus was able to estimate the 2016 U.S. soybean crop months faster and more accurately than anyone else. In these cases, better and faster information helps to decrease uncertainty, intelligently guide business decisions, and lead to better outcomes for everyone.

Google Comes to the Market

The final keynote of the day came from Ulku Rowe of Google. Rowe documented the many amazing technological feats that Google has and is accomplishing, including an amusing video on how DeepMind had an algorithm teach itself how to walk without giving it any instruction or guidance beyond simple requirements setting. But the most unique and interesting piece of news came with the announcement that Google will be rolling out a point of presence (POP) at the CME data center run by CyrusOne in Aurora, Illinois. What makes this so interesting is that Google typically has businesses come to them, not the other way around. In this case, Google is making a big commitment to the trading community by putting the full weight and expertise of their cloud right at traders fingertips. I’m sure that there are more than a few quants lining up for a shot at Google’s resources when they go live next month.

Tech Talk Delivers

CME Tech Talk is a refreshing alternative to the typical product/marketing/hype-fests that most exchanges produce. CME has consistently attempted to stay in front of the most important issues of the day and this year was no exception. Data is the order of the day and the insights provided by the Cubs, Orbital Insights, Tellus Labs, and Google illuminate what promises to be a very interesting future.