Skating for Gold ... as a Statistics Major

Nathan Chen’s gold-medal-winning performances on February 7 and 9 were stunning exhibitions of grace and athleticism. The post-ceremony interview featured the usual questions asked at such events: “You’re the defending world champion and crowned Olympic champion … how does that feel?” and “Can you explain the pressure you feel to perform well at the Olympics?”

I wish, though, that a journalist would ask him about an aspect of his career that sets him apart from many of the other athletes: Chen, a junior at Yale, is majoring in statistics and data science. What led him to choose to major in statistics? And how does he apply statistical methods to improve his skating?

How a Statistics Major Can Help College Athletes

When you think about it, statistics (or mathematics with a concentration in statistics) is an ideal choice of major for an athlete.

Athletes are constantly processing data. A tennis player has information about the velocity of an approaching ball from sensory input (she can see it) and from prior information (she has seen other approaching tennis balls and has a rough idea of the probability distribution of their velocities). Both of these inputs have variability. Bayesian models have been used to explain how “our minds make inferences that appear to go far beyond the data available,” by essentially combining the probability distribution of the data (the uncertainty about the velocity of the approaching ball from sensory input) with the prior distribution to obtain an updated probability distribution that has smaller variance than the original data distribution (because it’s incorporating additional information from the prior).
Statistical methods can be used to improve one’s own performance at any endeavor. Many businesses use designed experiments to learn how to increase the quality of their products and reduce costs, and those same statistical methods can be used to improve almost anything in life. I can’t claim any athletic prowess, but I have made many things in my everyday life easier and better with designed experiments, from cleaning the shower door to mastering a passage in a Mozart piano sonata to making coffee.
College athletes need time to train for their sports and flexible academic schedules. A mathematics or statistics major provides these. Math and stats classes typically don’t require time-consuming lab sessions (unlike chemistry and biology) or research papers (unlike humanities classes).* All you need for learning the subjects are pencil, paper, and laptop, and you can do the thinking and homework almost anywhere.
A career in statistics is fun and contributes to society. Few student athletes end up practicing their sport professionally; even those that “make it” may have short athletic careers. Why not major in a subject that is in high demand? The Occupational Outlook Handbook (put out by the statisticians at the Bureau of Labor Statistics) lists statistician as one of the 20 fastest-growing occupations, with an anticipated 35% change in employment between 2020 to 2030.
Statistics is an ideal subject for students with wide-ranging interests, since you can apply statistical knowledge in any field, from developing vaccines to monitoring climate to protecting national security. Students interested in a career in sports can use statistics there too, for activities such as building a winning roster of players (as in Moneyball), studying home-field advantage, and exploring techniques for speedskating.

How many college athletes major in mathematics or statistics? I could not find any reliable statistics about this.** But here’s an idea for a sampling class discussion. How would you design a survey to study college athletes’ choices for college major? What kind of information would you want to collect, and who is in the target population you want to study (current college students, college football players, figure skaters, … )? What information would you use for stratification? What type of clustering would you use?

Footnotes and References

*Classes in statistical theory typically do not require research papers. But I asked students in my sampling classes to analyze data from a large-scale survey and write a report on their findings. Students in my regression and design-of-experiments classes wrote brief critiques of research articles that used (or misused) statistics and wrote reports on the results of a factorial experiment they designed and carried out. These papers gave students practice and training in communicating statistical results (good communication skills are essential for statisticians!). And I learned a lot too: instead of grading the same set of problems over and over, I read about the experiments that students designed and implemented on how to make perfect-every-time crème brûlée or on what combinations of feeder, nectar color, and sugar concentration attract the most hummingbirds (measured by the amount of nectar consumed).

**I found statistics from convenience samples (which I will not repeat), and some news stories about individual athletes who have majored in mathematics or statistics (such as Frank Ryan, who worked on a doctorate in mathematics while playing for the Cleveland Browns). The Gallup Alumni Survey compared the percentage of student-athletes earning an advanced degree with the percentage of non-student-athletes, but did not provide information on field of study.

Sharon LohrFebruary 14, 2022