Preparing for a Career as a Sports Statistician: Two Interviews with People in the Field

Jim Albert is professor of statistics in the department of mathematics and statistics at Bowling Green State University. His interests include Bayesian modeling, the statistical analysis of sports data, and statistical education. He is the editor of The American Statistician and writing a text on data analysis, probability, and statistics for prospective teachers.

Many students are fascinated with statistics and sports and ask about a possible career working as a sports statistician. To help them understand this career, Jim Albert contacted Ben Alamar and Keith Woolner, who are actively working as statisticians with professional sports teams. Each was asked a similar set of questions, and their answers shed light on the background, academic skills, statistical methods, and people skills that are important for a person in this vocation.

 
 
 
 
 
 
 
 

Ben Alamar earned a BS in economics from the University of Minnesota and an MA and PhD in economics from the University of California at Santa Barbara. Currently, he is a professor of sports management at Menlo College and works part time as a sports statistician for the Oklahoma City Thunder. Alamar founded the Journal of Quantitative Analysis of Sports—an ASA journal dedicated to the statistical analysis of sports data—in 2005.

 

JA: When were you first interested in statistics applications in sports?

BA: I wrote my first paper on modeling the probability that an NFL team would make the playoffs during my third quarter in graduate school. I liked the paper, but never got it published. I didn’t return to stats in sports until I was a post doc at UCSF. I did some consulting work for a start-up company in the field and decided that it was the area I wanted to concentrate on.

JA: What are your professional duties as a sports statistician?

BA: I provide analysis on players and team strategy, as well as tackle larger research projects.

JA: Is your position as a sports statistician a full-time or part-time position?

BA: I work part time with the Oklahoma City Thunder and do a variety of other consulting work in the field, but I am also a professor of sports management at Menlo College.

JA: What skills and academic training (e.g., college courses) are valuable to sports statisticians?

BA: High-level statistics courses of all types are valuable, as are acquiring advanced data management skills such as SQL.

JA: Are there specific statistical tools or topics that you find especially helpful in your work? If a person had to take, say, three courses in statistics to help them in your work, what courses would they be?

BA: Some important tools include basic regression analysis, logistic regression, Monte Carlo simulation, classification, and hierarchical regression. Just as important as the technical tools though is the skill of effectively communicating the analysis to nontechnical audiences.

JA: Is there specific statistics or data-management software that you find helpful?

BA: R and SQL are very useful.

JA: What are the first steps in entering the sports industry as a statistician?

BA: There is no clear path. I recommend that aspiring sports analysts try to answer a question they think a general manager or coach would find interesting, then find a way to get that work into the hands of people who might be interested. There is no shortage of people interested in working in the field, but there is a shortage of people who have actually done good work in the field.

JA: In “Moneyball,” there was some resistance to the use of statistical methods to learn about players, especially by people who were not part of the baseball establishment. Do you think there is a similar resistance to the use of statistical methods in basketball?

BA: I would not classify it as resistance, but I think there is a natural skepticism of employing any new tool when you have had success previously without it. As decisionmakers gain more exposure to information that analysis can provide, they tend to become more interested.

JA: Do you think there will be an increasing demand for statisticians in your particular sport?

BA: Yes, I do. Data sets are becoming more complex (motion capture technology is being used to track everything that moves on the court 25 times a second) and the general concept of using statistics is gaining more acceptance; these factors will lead to more teams employing large analytics groups.

JA: Are there particular websites for the interested student to visit to learn about the current work in basketball analytics?

BA: Hoopdata.com, basketballvalue.com, and apbr.org.

JA: Do you have other general advice for high-school or college students who are interested in a career as a sports statistician?

BA: The advice I give to anyone interested in basketball statistics is to read Basketball on Paper, by Dean Oliver, try to start thinking like a GM/coach instead of a fan (think about how analysis can actually be used to inform decisionmaking), and work on communicating complex analyses to people who do not know linear algebra.

 

Keith Woolner earned two bachelor’s degrees from MIT—one in mathematics with computer science and one in management from the MIT Sloan School of Management. He then earned a master’s degree in decision analysis from Stanford University. While he was employed as a software developer, he regularly contributed to Baseball Prospectus (www.baseballprospectus.com). In 2007, Woolner left Baseball Prospectus to join the front office of the Cleveland Indians professional baseball team.

 

JA: When were you first interested in statistics applications in sports?

KW: I’ve always had an affinity for both baseball and math, and I memorized many of the statistics on the backs of baseball cards as I child, but it wasn’t until I was an undergrad at MIT in the late 1980s that I discovered the Usenet newsgroup rec.sport.baseball on the Internet and learned about the emerging field of sabermetrics and the work researchers like Pete Palmer and Bill James were doing. I was fascinated by it, eventually started tinkering around with my own methods, and began publishing my own invented stats in the mid to late 1990s. That led to my involvement with Baseball Prospectus and set me on the path to an eventual career in sports statistics. But, I didn’t seriously think a career in sports was an option for me until just a couple of years before I joined the Indians.

JA: What are your professional duties as a sports statistician?

KW: My title is director of baseball analytics, and in that role, I manage a team of analysts and programmers who support baseball decisionmaking by organizing, analyzing, and presenting information. We help baseball operations put the best possible team on the field. In many cases, that involves developing statistical models to measure player value, forecast future performance, and answer questions about game strategy and tactics.

JA: Is your position as a sports statistician a full-time or part-time position?

KW: My current role with the Indians is a full-time position. When I was with Baseball Prospectus, it was part-time and mostly a hobby.

JA: What skills and academic training (e.g., college courses) are valuable to sports statisticians?

KW: I would say there are three sets of skills you need to be a successful sports statistician:

• Quantitative skills—the statistical and mathematical techniques you’ll use to make sense of the data. Most kinds of coursework you’d find in an applied statistics program will be helpful. Regression methods, hypothesis testing, confidence intervals, inference, probability, ANOVA, multivariate analysis, linear and logistic models, clustering, time series, and data mining/machine learning would all be applicable. I’d include in this category designing charts, graphs, and other data visualizations to help present and communicate results.
• Technical skills—learning one or more statistical software systems such as R/S-PLUS, SAS, SPSS, Stata, Matlab, etc. will give you the tools to apply quantitative skills in practice. Beyond that, the more self-reliant you are at extracting and manipulating your data directly, the more quickly you can explore your data and test ideas. So being adept with the technology you’re likely to encounter will help tremendously. Most of the information you’d be dealing with in sports statistics would be in a database, so learning SQL or another query language is important. In addition, mastering advanced spreadsheet skills such as pivot tables, macros, scripting, and chart customization would be useful.
• Domain knowledge—truly understanding the sport you want to analyze professionally is critical to being successful. Knowing the rules of the game; studying how front offices operate; finding out how players are recruited, developed, and evaluated; and even just learning the jargon used within the industry will help you integrate into the organization. You’ll come to understand what problems are important to the GM and other decisionmakers, as well as what information is available, how it’s collected, what it means, and what its limitations are. Also, I recommend keeping up with the discussions in your sport’s analytic community so you know about the latest developments and what’s considered the state of the art in the public sphere. One of the great things about being a sports statistician is getting to follow your favorite websites and blogs as a legitimate part of your job!

JA: What are the first steps in entering the sports industry as a statistician?

KW: My path into sports was atypical, so it’s hard to use that as a basis for a strategy to get into the industry. There are more opportunities in sports now than there were years ago, but also greater competition due to the increased awareness of sports analysis as a career path from the popularity of “Moneyball” and the like.

What I usually tell people who ask me that question is that the best way to break into baseball analysis is to just start doing it on your own. Build up a base of knowledge so you’re aware of the state of the art and complement that with the technical skills you need to answer your own questions well. Write, write, and write some more. Your body of work is your résumé, and if you demonstrate your capabilities and develop expertise in a particular area, teams will notice. The more you can show that you both have the quantitative skills and the baseball knowledge to help a team, the better off you will be.

JA: What kinds of data do you collect and analyze beyond team and individual performance?

KW: There’s a tremendous amount of detailed data being collected during every game. The basic outcome data (how many hits, walks, runs, errors, homers, etc.) are the most obvious example, but we also collect information about where on the field each ball was hit, what kinds of pitches were thrown, and what situations each batter and pitcher faced.

There are cameras installed in every MLB ballpark that track the path of the pitch in flight to the plate, so we know how fast each pitch was thrown, how much each curve ball curved, where each pitch was released and crossed the plate, whether the batter swung, and how hard he hit it if he made contact.

Beyond the data collected in-game, we have many years’ worth of scouting reports, physiological tests, medical histories, psychological profiles, coaching assessments, contract data, service time records, arbitration case histories, negotiation records. We also have thousands of hours of recorded video. All these sources can be useful in analyzing and forecasting both team and player performance.

JA: How have statisticians influenced the operations of your team?

KW: Although I can’t get into specifics, I think where we have had the most influence is on improving the consistency of our decisionmaking process. The clarity that comes from performing the same analyses and running the same models every time we evaluate a different scenario provides some balance against being swayed by emotional reactions, or unduly influenced by the loudest voice in the room. At the same time, you will always have a need for the knowledge and intuition of experts like coaches, scouts, and executives, because every situation is unique. There’s no one-size-fits-all approach that works every time, and no computer can evaluate every possibility. But having that consistent evaluation process as a starting point helps ground your thinking so you avoid making mental mistakes as you work through a decision.

JA: Do you believe there will be an increasing demand for statisticians in your particular sport?

KW: I think there will be an increasing demand for statisticians who both understand their sport’s unique characteristics and can identify the right statistical methods to apply to a problem.

More and more, teams are even looking beyond their sports operations department and employing statistical techniques on their ticketing, marketing, ballpark/stadium operations, personnel scheduling, food and beverage, and merchandising data. There are many ways an analyst could help a club in these areas, too, so they shouldn’t be overlooked.

Because baseball was ahead of the curve in adopting statistical analysis, we may be closer to the saturation point than other sports. Eventually, there is an upper limit to how many jobs there could be with major league teams, as the number of clubs is pretty small and it’s unlikely that a team would need hundreds of analysts to meet their needs. The number of opportunities may be even greater in other sports than in baseball. But, even now, baseball teams are still hiring and expanding their analytic departments, so we’re not at that point yet. There’s still room for growth.

JA: Do you have some other general advice for high-school or college students who are interested in a career as a sports statistician?

KW: Don’t neglect the soft skills. Being successful in sports analysis is not just about having the most data or the best algorithm. Many brilliant analysts have struggled because they couldn’t get others to listen and buy into their ideas. It’s not enough to be right; you also have to be persuasive. Communication skills, both spoken and written, are important and under-rated. If you can’t explain what you’ve analyzed to someone who doesn’t have the same statistical training, you’ll have a hard time influencing the decisions they make.

Also realize that choosing a career in sports means spending a lot of time and long hours with your coworkers. How you are perceived by your colleagues determines how much they like you, respect you, and trust your judgment. Optimism, humility, open-mindedness, and a sense of humor go a long way toward building the foundations of good working relationships.