Benjamin Alamar is a Professor of Sport Management at Menlo College, sports analytics consultant/researcher, and author of the book Sports Analytics: A Guide for Coaches, Managers, and Other Decision Makers (July, 2013). He has consulted with a variety of teams in the NFL and NBA, including five seasons with the Oklahoma City Thunder and with a variety of companies in sports analytics. He has published numerous research studies in sports analytics and has written on sports analytics for outlets such as ESPN, Analytics Magazine, and the Wall Street Journal. Additionally, Professor Alamar is the founding editor of the Journal of Quantitative Analysis in Sports, the first peer-reviewed academic journal for research in sports analytics. You can follow him on twitter @benalamar.
GSABR: How did you wind up working with Oklahoma City Thunder?
BA: I was very fortunate to have had the opportunity to work for a startup company called Protrade that was doing advanced scoring metrics for fantasy sports. Through Protrade I had the opportunity to build a draft model for the Portland Trailblazers. Through Protrade’s website, the work attracted the attention of some folks working in the NBA who I spoke with occasionally about my analysis. Through those connections I was given the opportunity to interview with the Seattle Supersonics when Sam Presti was hired. I would like to say that there was a grand plan that I executed to get the opportunities that I did, but really, I just kept saying yes to interesting projects and talking to as many interesting people as possible.
GSABR: How do you deal with situations in which you don’t have the best data, for example prospects in the draft, vs. situations in which you have lots of established data, like signing or trading for veteran NBA players?
BA: The answer starts really with the purpose of statistical analysis in decision making, which, to me, is to help give the decision maker more and better information to reduce the uncertainty around a decision. Taking that perspective, the analyst needs to gather as much data as possible and honestly assess how much uncertainty remains after they have analyzed it. There is always going to be uncertainty in decision making and no amount of analysis will ever eliminate it, but, the analysis can significantly reduce it. The key for the analyst is to be honest in their presentation of the analysis about how much uncertainty remains after the analysis. When the analyst is clear about how much uncertainty remains the decision maker can more clearly weight that information in the process and both the analyst and decision maker can try and identity areas for more thorough data collection and analysis that would assist in further reducing the uncertainty.
GSABR: How do you give context to some of the advanced metrics you use? What sport-specific experience is necessary to make sense of the models?
BA: Context for metrics is key to properly assessing what information the metric is conveying as well as presenting the metric in an effective manner for decision makers. The first step is for analysts to understand the processes that create the data that they are using. A quarterback’s passing data, for example, is the result of the efforts of 22 players on the field and the schemes outlined for them by their coaches. The analyst has to consider this carefully when building a metric that claims to measure a QB’s abilities. When working with Dean Oliver at ESPN on their QBR, we worked to first put each play in context of the situation (down, distance to go, yard-line etc.) and then, as much as possible, separate the efforts of the QB from all of the other factors that create the result of a play. Building models like those that create the QBR requires an understanding of the sport so the analyst can identify the factors that lead to any particular result so they are measuring as precisely as they can given the data available, what they really intend to measure.
GSABR: What can students do to give them the technical, communication and sport-specific skills they need to do good and actionable analysis and eventually intern and work full-time for a team?
BA: The major technical skills needed are advanced statistics and data management. Training in statistics and data management provides the key tools that any analyst will need to make a career in sports (or any analytic position). These are the prerequisites though and not sufficient to really be successful. The analyst needs to practice asking questions that decision makers care about and doing their best to answer them with the tools that they have. Doing work in the area is key because it demonstrates real interest and ability. Completing projects in a thoughtful and insightful way, even if it is not exactly analysis that a team would utilize out of the box, is a signal to teams that the student is committed and interested and able to think analytically about the sport.
GSABR: Who should read your book, Sports Analytics: A Guide for Coaches, Managers, and Other Decision Makers, and what will they get out of it? What inspired you to write the book?
BA: The book is for anyone who has questions about using analytics within an organization. It provides an overview of the various components of analytics, how to build an analytic team and how to employ resources in a strategic way to maximize the advantage that analytics can provide. It is a non-technical guide to the field and provides tools and strategies for getting the most out analytics.
Special Thanks to Benjamin Alamar for his time and insight
Interview by Nik Oza, Georgetown Class of 2016