data science in sports management

Sports Management
The Emergence of Big Data in Sports Performance Management
Introduction
In the past twenty-five years, the amount of data available in this world has grown exponentially. In Western society in particular, we have become obsessed with the quantification of just about everything. One of the reasons is simple – there’s a lot more data out there, it’s relatively cheap to acquire, and there are just enough statisticians who know how to get value from data to make sense of it all. Some sports have always loved data – baseball, especially. The large data sets from playing 160 game seasons, and the way that each play occurs in isolation makes it easy for casual fans to understand baseball’s class stats. When people started digging deeper into baseball’s numbers, they came up with the ideas in Moneyball, rooted in the idea that there is a sufficient amount of granular data available that is not being used by everybody, and therefore can serve as a source of competitive advantage. Sports management has run with this concept, and now advanced quantitative analysis has become a core component of the management not just of entire sports, but individual athletes as well. This paper will explore the emergence of Big Data in sports performance management.
The Trend Towards Big Data
The trend towards Big Data in sport mirrors the trend in many other industries as well. The term Big Data appears to have originated in the late 1990s in Silicon Valley (Lohr, 2013), used to describe vast sets of data, newly available types of data, and the means to parse this data to derive meaningful results. The Internet was one of the key technologies in this trend, because it allowed for companies – such as Amazon – to collect data on consumer shopping and purchasing, and do so at scale. Businesses have always like information, but the cost of acquiring useful sets of quantitative data has often been prohibitively high, but when the cost of acquiring useful data in quantity began to drop, rapidly, in the late 90s and early 00s., Big Data became a trend that every industry wanted to get in on.
Sports has actually been one of the slower industries to adopt Big Data, but it is here now. Millington and Millington (2015) draw the pathway as to how Big Data arrives in the sports industry. Before Big Data, statistics were relatively simple – how many touchdowns, how many goals, what’s the shooting percentage. These easy-to-collect stats were used in contract negotiations, so it is a natural extension that if there are better numbers to support a position in a negotiation, those numbers will be entered into the negotiation.
In that sense, the use of data in sports management echoes the use of data in other forms of management. If a manager sees data as an asset, then the next natural question is what to do with that asset (Dhar, et al, 2014). Management teams in sports use data routinely in contract negotiations, to help make their marketing decisions, and increasingly as a means of governing the performance of individual athletes.
In sports, one of the new common uses of data is to provide more context and granularity to evaluations of individual athletes. Sports teams gather and use data in-house for this purpose. Fan blogs focused on data are prevalent now across many sports. An academic example of this phenomenon is a paper from 2017 on modeling scoring probability in basketball (Zuccolotto et al, 2017). We have arrived at the point where the day-to-day management of athletic performance is driven by data, and management in just about any sport would likely find it valuable to have at least a rough grounding in statistical concepts.
Applications in Sports
One of the older applications of quantitative data in sports is in contract negotiations between teams and athletes. Until recently, statistics were basically nothing more sophisticated than counting stats and averages. But these were the measures of performance that guided teams and agents in their negotiations at the professional level. Their importane at the professional level, combined with the ease of understanding, made them fun for fans, too, and adoption at amateur levels became commonplace. Children in Little League know their batting average. So this basic data not only served a functional purpose in negotiations, but is also fun and a means by which engagement with fans can be increased.
The next evolution is more of what would be known as the Moneyball approach, where a team uses quantitative analysis in order to optimize performance within the bounds of a given constraint – in that particular a player budget. Breaking down the game played into a series of contributions that each player might make provides a better understanding of how each athletc contributes to victory in a team sport.
The natural next step from there is to circle back to individual athlete performance. This is basically the application of fundamental management concepts to sports. Think of a draft pick as a new hire – the team will want to get that new hire up to speed as quickly as possible. They identified the hire on the basis of specific skills or traits, and now must develop those so that the new hire can make the best possible contribution to the organization’s goals, usually winning championships.
For their part, athletes also have a significant interest in using data to improve their individual performance. Elite athletes have high intrinsic motivation, and in most sports there are financial rewards associated with improved performance. Athletes now use metrics to determine weaknesses that they need to correct, to adopt new tactics, and especially to enhance their training. Equipment makers do the same – they use massive data sets to optimize the designs, materials and other aspects of athletic gear and footwear. This trend is relatively recent, and is still in its development stages, but it holds a fair amount of promise for sports management.
Reactions of Athletes
The use of data by athletes to enhance their performance is a relatively new idea, and there are definitely sceptics. But a lot of athletes, even ones who claim disdain for quantitative techniques, use them without knowing it. Big Data has become a significant part of coaching. A coach who works with a pro to refine her golf swing is probably drawing on a lot of data – if that golfer’s swing has been recorded hundreds or thousands of times, a program can identify key things that might have changed. The coach might do this with their eyes, too, but the data provides confirmation, and in some cases the eyes aren’t as reliable as the coach thinks. So there are many athletes who take advantage of the knowledge made available by large data sets, to enhance their training, their diets, their sleep, and their skills building.
There are certainly some athletes who resist the use of data. This is fairly common in any field – people distrust what they do not understand. Athletes are trained to think of themselves in physical ways, and may be more comfortable with that. So while there are athletes who have been more willing to embrace their quantitative selves (Baerg, 2017), especially in individual sports, that embrace has been slow to come, and is more prevalent today among younger athletes who have grown up in the data era.
The Data Gap
As the use of new forms of data, and massive data sets, has become more widespread, sports management is being transformed. But because Big Data in sports is still relatively nascent, its benefits have not accrued to all sports equally. One of the most significant drivers of the data revolution in sport has been driven by new means of gathering data – motion sensors, an increase in multi-camera coverage, smartphone apps and more. These developments, however, have often been driven by commercial considerations – either the sports industry itself or from companies basing their marketing around sports like apparel or video games. Smaller sports, women’s sports and amateur sports typically have much lower commercialization, and as such the development of measuring tools is behind in those sports.
The problem is that the rise of Big Data has created another way for fans to engagement with their favorite sports. As a result, there is a data gap between the data-rich sports and the data-poor sports. If large data sets provide an additional pathway for fan engagement, and if they provide a means for athletes and coaches to improve their performance – thereby creating more intense competitions, for example – then the gap between data-rich and data-poor sports might grow wider, the detriment of those underrepresented sports – women’s, amateur, Special Olympics, etc. (Hutchins, 2015).
Impact of Big Data on Sports Management
Sports management is benefitting from the increase use of Big Data sets. In the major professional sports, the rise of data puts management teams more on parity with their counterparts in other entertainment industries, giving them the tools to make better decisions. For athletes, Big Data opens up opportunities to improve performance, and gives their agents new tools to be used in contract negotiations.
The downside, for some at least, is that the purity of athletic completion may be eroded. Data-poor sports are likely to receive less attention in the Big Data world, to the detriment of creating opportunities for the athletes who participate in those sports. Furthermore, the increasing quantification of sports is seen by some as reducing the emphasis on the physical elements of sport – the faster, higher, stronger parts of human achievement. The advances in athletics in the past 150 years have been the result of the application of different sciences – biology, chemistry and design have all transformed different sports. Now, data is the technology that is changing the way that sports are played, and managed.
From the perspective of somebody with an interest in sports management, Big Data is part of the process of professionalizing this business. Managers on the talent side of sports are achieving parity with their marketing and finance departments, for example, but also with other data-rich entertainment competitors. All told, the use of Big Data will be a positive in sports management, one that allows sport to receive a greater share of attention, ideally a greater share of revenue, and by way of that to allow sports to continue to thrive.
Conclusions
Over the past twenty years, Big Data has entered most business realms, and the business of sport is no exception. Yet, sport has been relatively slow to adapt to Big Data, despite the obvious opportunities it presents to make sporting organizations more efficient and effective. Big Data’s impact on sport is just starting to be known, and going forward there should be substantial influences on things like individual athlete performance that are relatively unexplored areas thus far.



References

Baerg, A. (2017) Big data, sport and the digital divide: Theorizing how athletes might respond to big data monitoring. Journal of Sports and Social Issues. Vol. 41 (1) 3-20.

Dhar, V., Jarke, M. & Laartz, J. (2014) Big data. Business and Information Systems Engineering. Vol. 6 (5) 257-259.

Hutchins, B. (2015) Tales of the digital sublime: Tracing the relationship between big data and professional sport. Convergence: The International Journal of Research into New Media Technologies. Vol. 22 (5) 494-509.

Lohr, S. (2013) The origins of Big Data: An etymological detective story. New York Times. Retrieved April 14, 2019 from https://bits.blogs.nytimes.com/2013/02/01/the-origins-of-big-data-an-etymological-detective-story/

Millington, B. & Millington, R. (2015) The datafication of everything: Toward a sociology of sport and big data. Sociology of Sport Journal. Vol. 32 (2) 140-160.

Zuccolotto, P., Manisera, M. & Sandri, M. (2017) Big data analytics for modeling scoring probability in basketball: The effect of shooting under high-pressure conditions. International Journal of Sports Sciences and Coaching. Vol. 13 (4) 569-589.