Careers in Sports Analytics—Highlights of an ASA Webinar

sportswebinar

The American Statistical Association recently hosted a webinar focusing on the role of statistics in sports and career opportunities in that niche. CHANCE editor Scott Evans moderated the session. This article provides a transcript of the webinar, with some editing for space and magazine style and audience.

Scott Evans: Good afternoon, everyone. … welcome to the Careers in Sports Analytics webinar. I’d like to begin by thanking the ASA for organizing and sponsoring this event. The ASA is the world’s largest community of statisticians and is often known as the big tent for statistics.

I’d also like to thank Stephanie Kovalchik at Tennis Australia and Dennis Lock of the Miami Dolphins for participating in this webinar. They tell their personal stories about how they became sports statisticians and how they use statistics in their respective sports and jobs. They will also share insights regarding the evolving world of sports statistics.

Finally, I’d like to thank all of you for your interest in statistics and sports. I think you’ll enjoy listening to today’s discussion.

The availability of data and use of statistical methods in sports are growing rapidly. Sports teams use statistical analysis to evaluate players and determine game strategies, sports associations develop ranking and rating systems for players and teams, and sports analysts evaluate concepts such as streakiness and a hot hand. Professional athletes now use statistics to help them identify the strengths and weaknesses of their games so they can direct their training and improve their performance.

The evolution of the application of statistics to sports has created demand from professional and other sports organizations for well-trained statistical experts who can apply cutting-edge statistical tools to analyze sports data. We have two such statistical experts joining us today.

This webinar was recorded and is available at ThisIsStatistics.org.

Dr. Stephanie Kovalchik is the lead data scientist in the Game Insight Group at Tennis Australia, which is the governing body of tennis in Australia, and a research fellow and sports analytics at the Institute of Sports, Exercise, and Active Living at Victoria University. Her research focus is on the use of statistical methods to understand performance, game strategy, and mentality in high-performance tennis. She’s also the creator of a tennis analytics blog called On The T and regularly writes about tennis there, and on Twitter at @StatsOnTheT. Stephanie joins us from Australia. Stephanie, could you tell us a little more about yourself and your main responsibilities at Tennis Australia?

Stephanie Kovalchik: My role at Tennis Australia is primarily research, so in that sense, it’s not that different from what a traditional statistician would do at a university or at our research institution. What makes my role in sports unique is the different audiences for the research that I’m doing.

One area—one audience—is the research community at large, other statisticians, other sports scientists. In addition, a lot of projects that I’m working on are intended for coaches or athletes, and those would involve questions more specific to game strategy or improving the training or helping players prevent injury, for example.

A set of projects that I work on is directed more toward fans and broadcasting. There, the goals are often to promote analytical thinking to a wider community. I think those are some of the general categories and ways in which a role in sports statistics can be unique.

Evans: Our second speaker is Dennis Lock, who is entering his third season with the Miami Dolphins and his second season in his current role there as director of Analytics after serving as the head analyst in 2013.

In his (current) role, he supports football operations for statistical analysis and research. He’s currently finishing his doctorate degree from Iowa State University with a dissertation on using statistics and sports. While at Iowa State, he served as a consultant for the Iowa State University men’s basketball team.

Dennis comes from a family of statisticians and has published a prominent textbook with several members of his family: Statistics: Unlocking the Power of Data.

Dennis Lock: In general, I’m in charge of everything data—data analyses, analysts, and data/analytical prospects—(in all) football operations for the Miami Dolphins. My perspective will be a little bit different from Stephanie’s in that everything I do is singularly focused on the Miami Dolphins and directly for the Miami Dolphins.

To speak broadly, I break up the main parts of my job into three equally crucial components. The first is the work we do for player personnel, and by that, I mean projects in scouting, team building, and resource allocation.

Second, the work we do with the coaches, such as game planning, and opponent and self-evaluation. Essentially, I want our coaches to have more and better information than the opposition every week.

The third and final aspect, which is I think one of the big futures of sports analytics and isn’t as prevalent right now, is sports performance and training evaluation. These projects center around injury prevention, peak performance, training analyses, and just how can we get the most out of players who are in our building.

Evans: … Some of the things we’ll talk about with both Dennis and Stephanie today (include) how they got where they are, and what education and training they received along the way.

There’s a perception in sports statistics that statistics is about memorizing batting averages for all of your favorite players and that you become a data warehouse of sports trivia. However, it’s really important to note that being a sports statistician goes beyond trivia and beyond the basic statistical calculations of batting averages or yards per game or first-serve percentages.

Statistics is really the science of learning from data, and measuring and controlling for and communicating how much uncertainty there might be. Statistics is a foundation for how to think about problems and provides sort of a compass for helping us make better decisions.

One definition might be that sports statistics is a scientific discipline that yields powerful insights, provides a competitive advantage, and solves interesting problems.

I think there’s a perception that statistics is all about analysis, but statistics could also involve designing studies, collecting data, and using adaptations to ever-evolving data collection technologies. Then there’s a great deal of critical thinking regarding how to use these data from these studies to make better decisions that would improve player or team performance or maximize the probability of winning.

I would like to ask, Dennis and Stephanie, when you meet people, do you encounter this phenomenon? Do they perceive you as a data warehouse—somebody who knows a lot of statistics trivia? How do you describe to people, perhaps in a more-accurate way, what you do?

Kovalchik: I do think that that tendency is there, to sort of expect us to know records, have various facts and figures at the tip of our fingers. There is some truth in that, in the sense that we do have to, in our role, have familiarity with what data is out there. Perhaps we haven’t memorized all of the figures, but I think we have to be savvy—knowledgeable about the different data sources that are there—particularly now that so much data is available online and from different sources.

Often, that can be a value to an organization as a sort of complement to what they might collect internally and understanding what is useful for addressing a different question or what isn’t, or what potential biases there might be in those data sources. That is all a critical part of our roles.

There’s an opportunity when we’re in a situation where we have to talk about our role with giving the data, to define what our responsibilities are in our organizations and what we imagine that should be. I often see my role as being one who one provides a new perspective to an organization, since the organization largely doesn’t have much statistical training among its current staff.

That’s something I bring that’s unique. In doing that, I see a lot of my responsibility as being to address some of the biases that are common—the everyday problems that we can get into when we think about probability and statistics, because it’s often not intuitive. That can lead, perhaps, to biases in the way we’re understanding what’s happening in performance in sports. There are common ones that we come across. I see a lot of my goal as to try to reduce those tendencies.

Evans: Thank you. I agree that much of the time, trying to communicate with non-statisticians, particularly about biases, is such a critical role for us. Dennis, how about from your perspective? How would you describe what you do, particularly to those who view you as a library of statistics?

Lock: I’d say that probably on the bottom level, that’s a large part of what I am: a library of statistics. That’s the very base level of sports statistics. The interesting thing is that the sports statistics you see in the media is primarily these base-level statistics: passing yards, how they did on third down, quarterback rating, etc. Generally, just pretty much basic hindsight statistics. Working for a team, we want to go beyond the “what the statistic is” to the “why the statistic is.” The why is what really matters.

Suppose we were 2 for 13 on third down this past week. It’s not that fact that’s important to us, although we have to identify that initially, but what’s really important is why we were 2 for 13 on third down and how we can change that this week to perform significantly better than 2 for 13 on third down.

Related to that is the whole predictive aspect—using statistics, using hindsight to create foresight about what’s going to happen in the future and determining how we can affect that future. The hindsight matters, but what’s truly important in my role is predicting what’s going to happen next and determining ways that we can tip that in our favor.

Evans: Do you see the media getting any better at its use of statistics and understanding of statistics at that deeper level?

Lock: I think the media is getting better at understanding the important hindsight statistics, but that they’ll kind of always be stuck in that hindsight mold, because that’s simply what’s interesting to the fans. The casual fan is much more interested in how the team did on third down last week than on how the team may do on third down this week.

Evans: I want to (mention) the New England Symposium on Statistics in Sports, also known as NESSIS. It’s run here at Harvard every other year. The next one will be in September of 2017. It’s a research symposium where people submit research that they’ve been working on in statistics in sports. Both Dennis and Stephanie have presented at this symposium.

I mention it because it’s a good resource for looking at video presentations and slidesets from people who present their work there. If you go to www.NESSIS.org, there are videos from past symposia, and you can view people presenting their research on statistics and sports there. You could find both Stephanie and Dennis there.

I also want to mention that many schools are organizing clubs and groups that talk about sports statistics. Harvard has a new sports analytics laboratory that is organized by Mark Glickman. He’s also the person who organizes NESSIS with me. The American Statistical Association, which is sponsoring and organizing this webinar, has a section on statistics and sports for people who are interested in this area. You might be interested in joining the American Statistical Association, joining the statistics in sports section, and becoming part of a community of statisticians who work and are interested in this area.

There are student rates for joining the American Statistical Association. They make it very affordable. If you’re interested in that, you can gather more information on www.amstat.org.

Let’s talk about what Dennis and Stephanie do in a little more detail. Perhaps each of you could describe some of the more interesting problems that you’re working on, and why are they important to your respective sports and important to the organizations.

Kovalchik: I’ll mention three current projects that I think are interesting in that they each represent broader problems in the sport where a lot of further research can be done.

One of these is looking at how players handle pressure. One of the areas that I’m particularly interested in is the mental aspect of the game, which traditionally has been thought to be particularly crucial to staying in an individual sport. There also is so much time that the players aren’t actually actively in play—during a professional tennis match, actually about 80% of the time is spent in-between points.

A lot of our interest has been in how the thinking and the processes, the routines, that a player does in that period influence their performance?

One way that we’re starting to look at that is by looking at patterns and routines in between points, and specifically in tennis, developing a model for the time that players take to prepare to serve. That’s nice, because it’s an event in a match that’s almost entirely under the control of the server.

We can use that to understand an individual player and what kind of patterns are being based on the duration of the preparation time. We look at that from one point to the next and, in doing so, we can then look at variation in that preparation time and develop a measure of consistency in those patterns.

I’ve been working on this with Jim Albert, who is a statistician at Bowling Green State and former editor of the Journal of Quantitative Analysis in Sports. We have done a lot of work in baseball, so I was able to recruit him for this tennis project and we developed a Bayesian model that would allow us to measure individual, player-specific average times and consistency, which are essential for the variation in their time to prepare. That can give us insight into how players are influenced by different aspects of the game, so we can incorporate the game context; for example, what is the score and how does that influence the average time or the consistency in a player’s routine, which can give us more direct insight to the mental aspect of the game. That’s one area.

We’re also interested in terms of the course of play—how a player is responding to pressure in actual active play. That’s involved developing measures for bigger points in a match.

You often will hear stand commentators talk about big points or this idea of clutch performance. We’ve tried to quantify those moments. In doing that, we can then look at how a player’s performance varies under those higher pressure situations compared to ones that may be more-typical or less-important points.

That can give you insight into how players respond to pressure, and that’s a particular interest for coaches, because it can identify potential weaknesses that would be of more impact for the game outcome, so it would highlight some aspects of a player’s performance that would be particularly crucial to success and potentially could influence training decisions for those players.

The third one is sort of a general area of advanced metrics that could be used both for research and understanding performance and the sport in general, but also as a potential product for fans. One that we’re developing is a shot difficulty measure.

This is a particular interest these days because it involves tracking data that we have more and more available now, with systems like the Hawk Eye technologies that underlie the player challenge system. This is a multi-camera system that tracks the ball and player position, essentially continuously throughout the course of a match. Now more of that’s becoming available to our research group and others, (providing) richer data to assess performance.

One of the things that we’d like to do with that is to actually look at the shot level and start to say, “What shots are more effective than others?” and develop metrics based on that.

For example, you could start to think about errors in an entirely different way. We don’t have to be constrained by this idea of unforced and forced errors. You can start to give errors in relation to the difficulty of the shot that those errors occurred on. Those are some of the things that we’re hoping to advance.

Evans: That’s very interesting. Some of those things are so hard to quantify. Working on methods to do that is important. I know when I’ve played racket sports in particular, or watched them, you note that some players don’t play as well when they’re ahead because they lose focus. Then there are players who don’t play well from behind because they give up, and lastly there are the people who don’t play well when it’s close because they get nervous. It is a very mental game in that sense. You have to get tough from all angles.

Evans: Dennis, do you have a couple of in-depth stories about projects you’re working on?

Lock: Sure. One of my personal favorites is resource allocation in the player procurement process—how we’re building a team and how we’re getting the players in here. This one’s fascinating to me because it’s a level playing field, since the NFL is a salary cap league where everyone has the same amount of money to spend on the players and everyone has a comparable set of draft picks every year, so everyone has the same set of resources and the team that uses those resources the best is the team that’s ultimately going to have success.

Another interesting project is the prevention of soft tissue injuries while producing peak performance in our athletes. In the NFL, once the season gets going, the guys you’ve got are the guys you’ve got, primarily. For that part, we want to help our players not just play each week, but play at their best each week. To aid us, we have piles of data regarding each week of training on each individual, so it’s our job to use these data to determine what leads to peak performance without leading to a high incidence of injury.

Evans: That is exciting. One question for both of you is how interesting and exciting do you find your jobs?

Lock: I love game day; I love draft day; I love the high-intensity, high-adrenaline, high-anxiety moments—being at a stadium and feeling the electricity, then watching the game and seeing some of our department’s ideas come to fruition out there on the field. It’s especially nice when they’re working, not quite as nice when things aren’t going well, but either way, it’s a lot of fun and excitement.

Kovalchik: I definitely share those sentiments. The two things I’m most passionate about are statistics and tennis, so it’s really the perfect marriage. I used to work in the more traditional role in rich health science applications and I was quite happy doing that, but I would spend my evenings and weekends grabbing data from the Web about tennis stats and looking into questions that I hadn’t really seen tested in any way, never thinking that I could do this work full-time.

When this opportunity came along, it was kind of a no-brainer to do it even though it involved moving across the globe. I was happy to do it because I knew I’d be so excited each day when I was working on the projects I would have available.

Evans: Who are your main collaborators? With whom do you have a lot of interaction? Dennis, do you interact a lot with various members of the coaching staff, do you interact with players much? Stephanie, how about you?

Kovalchik: It’s a combination, partly because of my role being split between a university and industry organization through Tennis Australia. At Tennis Australia, I share my office with our performance team, which includes primarily coaches and trainers. They’re going to be the main people I’m collaborating with there, but I also am involved with more academic work and have colleagues at Victoria University and some of the other universities in the area that are addressing questions sometimes specific to tennis, sometimes just to sport science in general.

One of the nice things about that is that I’m able to work with a lot of students, which I didn’t know was going to be a possibility until I had been in this role for some time. There’s been a lot of opportunity to get students, from the undergrad level to a more-advanced degree, involved in projects. It’s been a lot of fun to see the ideas that they have and how motivating it can be for them to combine an interest in sports and an interest in analysis.

Lock: My number-one collaborator is Tom Pasquali, another statistician and the other half of our analytics department. The two of us collaborate on pretty much all of our projects to come to a joint conclusion.

It sounds like Stephanie is doing a lot of things in academia and collaborating with universities. We’re not doing as much of that, but we’re collaborating with pretty much everyone in football operations within the Dolphins—the GM, the executive vice president, all the coaches, strength and conditioning coaches, the sports performance individuals, even as far as the training staff, and really pretty much anyone who has data that we can get our hands on to help them do what they do.

Evans: It’s so important for a statistician to be a good collaborator in many ways. When you’re collaborating with coaches and players and people in front offices and so forth, they help make sure that you understand what the problem is. It’s also important to make sure you understand the questions before you try to find out what the answers are.

I want to back up a little bit and talk about the statistics profession. The statistics profession is really a very promising career. Google’s chief economist, Hal Varian, said about statistics, “I keep saying the sexy job in the next 10 years will be statisticians.”

When you look at the sports world, are you finding that teams are hiring more statisticians? Are they appreciating statistical expertise? What are your thoughts about that?

Kovalchik: I think this one may differ depending on whether we’re talking about a team sport or an individual sport, just because often the way the league or equivalent is organized could look quite different.

For example, if you have a fractured organizational structure, there isn’t one centralized body that owns all professional activity. It’s split into these different groups. I think the opportunity is a bit more difficult to assess there because each organization could go with some direction.

However, from the conversations that I’ve had, I do think that there’s a sense of excitement about a lot of possibilities, so I do think all sports probably benefit from the success that’s perceived in sports like baseball. You hear people throwing around “money ball” as kind of a general term to refer to some idea about statistics having the potential to make huge impacts in sports.

I think that has raised a general sense of awareness in some organizations. For tennis, which has generally been seen as being in the dark ages of statistical analysis, there’s even a sort of sense of urgency growing in that area to help with the competitive advantage. Right now, the absolute numbers are kind of small, but the growth is high.

Evans: Great. I agree. Dennis, what’s your perspective?

Lock: I’ve got a little story that can paint what I think is a good picture of how the field has grown.

I graduated from undergrad in 2008 and had just completed a project on evaluating NHL players with an adjusted plus-minus system with Dr. Michael Shuckers (who at the time had not done much with NHL data, but is now one of the leaders of NHL analytics). I presented the results in multiple places, eventually to an assistant head coach for an NHL team, and then his GM later that fall—the fall of 2008.

At that time, this individual and most of the league had not even heard of or considered statistics as a way to improve their process, so naturally, that didn’t work out.

I went back to grad school to learn more about statistics, improve my réésumé, and try again. In 2013, just five years later, when I started speaking with teams again, most either had some form of an analytics presence or were looking to start one. Over the course of five years, it moved from most teams not knowing what analytics was to most teams having some form of analytics.

Now, years after that, the NHL has partnered with SAP in making advanced stats public and available to everyone. Terms like Corsi and Fenwick, which are statistics based on shot attempts, are being discussed regularly during NHL broadcasts.

The point of the story is eight years ago, no one in the NHL was even considering analytics; three years ago, all teams were aware and some teams were doing it; and now, pretty much all teams are doing it and everyone knows about it.

This boom has been comparable in other non-baseball North American sports in the past decade. I know multiple NBA teams that are looking right now for individuals to hire, and we’re going to be looking for an individual at the end of this season. It’s boomed, and it’s still booming.

Evans: A growing market—I agree.

Dennis, how much of a say does your analytics group have in personnel decisions?

Lock: I wouldn’t call it a say. What we’re doing is providing information for the decision-makers as part of their process toward making an informed decision. We’re just providing as much information as we can so they can do their jobs to the best of their abilities. If we do our jobs well, our information will be used whenever an important decision has to be made.

Evans: So they can make more-informed decisions.

Lock: Exactly.

Evans: Another question that came in for both of you regarding when you’re discussing your findings with decision-makers and your collaborators—how much granularity do you go into? Realizing they may not know high-level statistics, or at least to the level of your understanding, is that communication difficult?

Lock: It can be. The ability to communicate findings is just as important as the statistical analysis, and sometimes it can be quite a bit more difficult. Essentially, even if you do the perfect statistical analysis, if you can’t then explain that analysis to a decision-maker in a way that he can understand, it’s not a very good analysis.

Evans: Stephanie, how about you? Challenges in describing the results of your work?

Kovalchik: I agree with Dennis that effective communication about our work is really an essential part of what we do. It can’t have an impact if our team of collaborators can’t understand what we’ve done or what the implications are of the results. That can be challenging because there is such a diverse range of disciplines in the teams that we work with, and also sometimes some resistance to change.

When you are in an area that is growing so rapidly, you are kind of the new kid on the block, and there can be many challenges in that role as well. Being conscientious of the audience often can mean reminding myself that the person I need to communicate to may be in more of decision-making role than I ever am.

I’m usually the information provider, as Dennis was describing, but often the people I’m providing the information to have to make the decisions on a daily basis, so their perspective is often, “What does this mean in terms of any action that I need to take?”

Being able to put yourself in that perspective might help you see what’s the key message here and then think of how that key message can be explained in plain terms. It does make sense that, if you’re finding that it’s difficult to (relay) a plain language message, you should rethink why that is the case, because it may suggest something about the analysis that should be used in a way that should potentially improve.

Evans: Certainly a valuable skill for a statistician to have is being able to explain things in non-technical language. If you can’t do that, then people don’t use your work as much. Learning to explain things is part of the process.

Let’s highlight the breadth of the statistics profession in general. CHANCE magazine is a publication of the American Statistical Association. It publishes articles about various topics, with special issues, including ecology and forensic statistics. There’s a special issue on sports from a couple years ago. One of the latest ones highlights a Zika virus article on the cover.

Statistics is applied in many different settings. Statisticians apply statistical thinking and methods to a variety of scientific, social, and business endeavors; everything from astronomy and biology, education, economics, engineering, genetics, and so forth.

A well-known statistician, John Tukey, is often cited as saying, “The best thing about being a statistician is that you get to play in everyone’s backyard.” Certainly the fun world of sports is one of those.

Dennis, you talked a little bit about how you got here. Some might have questions about what education and skills are needed to become a sports statistician. For example, do you need a PhD or a master’s degree or a bachelor’s degree? What are employers looking for? Remind us about how you got to where you are.

And, if you were advising a younger person who was interested in a job like the one you have, what would you suggest in terms of their education?

Lock: I don’t think I’m the norm as a statistician who has a PhD education in statistics and is working with a team. I think that most of the analytics individuals out there working with teams primarily have master’s degrees, and several have bachelor’s degrees.

I see a wide variety of fields (using statistics), from economics to statistics itself; computer science is also a big one right now, as well as predictive analytics and other subsets of statistics.

I would advise that even those who won’t be computer science minors or majors take a few computer science courses because there are a lot of times on my job where I feel more like a computer scientist than a statistician. We’re always looking for that mix of computer scientist and statistician.

I would also advise students to take advantage of the opportunities they have at the university or high school they are attending to perform some kind of statistical analyses for a team at that university or school, offer to be a volunteer, and help them look at what they’re doing in a unique way. Working as a graduate student volunteer for the Iowa State basketball team gave me valuable experience.

Evans: Stephanie, your thoughts?

Kovalchik: There are opportunities at any degree level, and there isn’t one particular path to getting into a role as a sports statisticians. Some of them are going to make those options more readily available than others.

The more advanced degree you have in a statistically low-rated field, the more options you will have, but I don’t think that it’s necessarily possible to demonstrate a skill or interest in sport and the statistical skills for adjusting problems with data and analysis in other ways. It just requires a bit more initiative, so there’s a long history of people who have gotten degrees in other areas—or maybe not at all—but through blogging and writing on their own have demonstrated a new way of looking quantitatively at problems in sport that has helped them get important roles with these other sports organizations.

I just think that’s probably a harder path because it requires a lot of self-initiative and time outside the primary work.

In terms of the skillset needed, in addition to computing, which I agree with Dennis is more and more essential these days, there’s also a lot of interest in having people who are not only familiar with statistical modeling approaches but also machine learning. More and more organizations are working with people in these skills and their methods as well.

Demonstrating an interest in sports at some level definitely is going to set you apart from other people who may be competing for the same role that you’re interested in. Whether that’s through an internship with an organization at an early stage or through presentations at events like NESSIS, showing an interest in the sport of knowledge in terms of subject matter is going to be a big help in getting a role.

Evans: Yes, I agree. I think that many people who have sports statistics jobs now have bachelor’s degrees. There may be a few with master’s degrees and PhDs. As time goes by, such positions are going to get more competitive and more advanced as the field evolves. In the future, these jobs could require higher degrees, such as master’s and doctoral degrees.

For example, if you were applying for a statistics professor position, most colleges and universities would require a doctoral degree. The sports world may not be quite there yet, but as they continue to appreciate what statistics might be able to help them with and as more and more people get interested, the advanced degrees will be appreciated and valuable.

In terms of courses, certainly any statistics class that you can take is helpful. There are a lot of emerging data science programs out there that will involve some of the computer courses that have been mentioned, as well as logic and mathematics.

Of course, learning about the sport is also important. Taking the opportunity to interact with people like Dennis and Stephanie, whether it’s at a conference like NESSIS or through the ASA and the Section on Statistics in Sports, can be valuable. You’re going to learn a lot from people like that and those types of interactions.

We have a few questions coming in online: What programming languages do you use? What is the best place to find open positions for sports statisticians? Do either of you have thoughts about that?

Lock: I use R primarily, and one of the universal data management languages you’ll find in the team sports landscape is SQL. That’s one language it is very valuable to get at least a basic level of experience with. If you have experience with any major statistical software, statistical programming languages, that’s great.

In terms of the best place to find open positions, I would recommend the Sloan Sports Analytics Conference—there are usually a lot of open positions listed there and you can also submit your résumé to the réésumé book. That’s one of the first places we check when we’re looking to contact potential employees.

Evans: Stephanie, any thoughts?

Kovalchik: I would agree in terms of the programming languages—R or an equivalent is definitely a skill to have. Also SQL for database management; that is also what our organization primarily uses. We also have people working in MATLAB, so that’s something as well. With proficiency in MATLAB or R, you can easily pick up one or the other.

In terms of careers, in addition to (making) phone (contacts), any of the sports-specific conferences, in addition to NESSIS, which is where I found out about this job at Tennis Australia, so I have to thank Scott for organizing that one.

There are also some international events. If you are able to travel overseas, there are events like Mathsport International, which is a great way to interact with sports statisticians doing work primarily in sports that may be more popular in Europe, like soccer or cricket. If you have an interest in some of those or think you might, that’s a great place to show your work. Those (events) tend to be smaller than something like the Joint Statistical Meeting, which has the advantage that it is easy to meet everyone who’s there, all of the speakers, and that’s a great way to find out about opportunities that may not necessarily be advertised.

Evans: I want to remind people that there’s also information at thisisstatistics.org about careers in statistics. You may also find more information about careers in statistics at www.amstat.org, which is maintained by the ASA.

Dennis or Stephanie, any closing thoughts from you?

Lock: I would just say that whether or not you want a job in sports statistics, statistics is a fantastic discipline because it can be applied to any field. Anything you want to work in, if you get a degree in statistics, you can probably find a job in that field.

Evans: Well said. I agree.

Stephanie, how about you?

Kovalchik: If you’re really passionate about statistics and sports statistics, you should get started today. There a lot of questions out there that we can answer with data that’s available publicly. That’s really the best way to see if it is the right fit for you, as well as to go ahead and start getting involved in the community out there.

Evans: Thank you. Again, I want to thank Stephanie and Dennis for offering their advice and telling us their personal stories and their thoughts about the future of statistics and sports. I also want to thank the ASA for organizing this event and to thank all of the attendees.

There were some questions that we didn’t get to, but there are lots of different avenues by which we can continue some of these conversations; certainly via e-mail. The ASA has mechanisms for discussions on statistics of sports, too. You could join these discussions by joining the ASA and becoming a member of the Statistics of Sports section.

Perhaps we will see some of you at symposia like NESSIS next year.

Back to Top

Tagged as:

Leave a Response

Please note: comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.