Each week, Daniels is featuring a researcher who conducts meaningful research that impacts their field and the wider community. Learn more about their work in Q&As with the Daniels Research team and email them to nominate yourself or a colleague for a future Q&A.
Ryan Elmore is an associate professor in the Department of Business Information and Analytics. Prior to Daniels, he worked as a senior scientist in the Computational Sciences Center at the National Renewable Energy Lab in Golden, Colorado. He has also held positions at the Australian National University, Colorado State University and Slide, Inc.
Elmore’s research interests include statistics in sports, nonparametric statistical methods, and energy efficient high-performance computing. His work in sports statistics has led to the position of Associate Editor for the Journal of Quantitative Analysis of Sports (2015–present).
What are you currently researching?
I like to look at interesting problems that arise in sports, a lot of which tend to come out of bar room-type conversations. For example, one of my recent papers related to timeouts in the NBA started as an argument over beers in Steamboat. Roughly, the problem can be stated as, does a coach calling a timeout in an NBA game affect the outcome of the game? Do the timeouts actually matter? So, my research program revolves around taking these questions and framing them as mathematical or statistical problems, and then using the tools available to me to solve it.
Another basketball problem that I’ve been working on recently, along with Ben Williams and some other colleagues around the country, is a question related to aspects of teams’ and players’ offensive playing styles. We looked at data from the past ten or twelve years and grouped teams/players on shooting characteristics. The idea is that by studying these patterns, we can use this information to develop a new offensive metric to evaluate players. Consider a player like Steph Curry. He takes a lot of shots, he makes a lot of shots and he gets more points per game than a player who doesn’t take as many shots. Can we find a player that his similar shooting characteristics as Curry who might be undervalued? The short answer is no, but you could imagine this exercise being useful.
I would like to work on something related to golf, but we no longer have access to the Shot Link database that the PGA Tour collects.
How do you share your research with the public?
I really harness social media, particularly X (formerly Twitter). The sports analytics world is fairly active on that platform, and we’re always trying to promote each other’s work. In this field, there’s an element of luck in who picks up your work, but you definitely have to put it out there online to make it accessible to begin with. My research has been quoted by the Economist, the Wall Street Journal, among other outlets, and I don’t know that that would have happened if I hadn’t been active in pushing it out on social media.
How do you integrate your research into the classroom?
In 2017 or 2018 I taught the first sports analytics class at DU. It was super popular, which I think stems from the fact that most people know at least a little something about sports and so students can more easily grasp the application of stats and analytics. I like to share my own research with the students so that they know, not only do I know about the work in sports analytics, but I’m also a participant in this world. I would also bring in external speakers to those classes from big, recognizable organizations like the Nuggets and the Rockies, which really enhanced the learning experience and motivated students.
I’ve also volunteered with the Carnegie Mellon Sports Analytics Conference, which is an annual event in Pittsburgh where you get to work with students on research problems. That’s something that I’d like to do more of, in particular helping students get their work published.
How would you like to see your work impact the public sphere?
It would be impactful to see general managers, pro teams, commissioners and other high-level folks within sports to view some of my research and understand the value that it could bring to a team and/or players. Then, upon realizing that our work adds value, they would release more data to the academic community so that we could develop additional insights. In other words, I want them to understand that the more publicly available data there is, the better it is for the sport overall.