We Love Sports. What About the Scientific Data Analysis Behind It?
Do you ever wonder where targeted practice comes from? Do you ever think about why athletes perform differently against different teams? Do you ever figure out why your favorite player transfer to another professional club?
Sport Analysts can give you the answer.
How many of you here have been to a sports event and cheered for your favorite players? It can be the expensive professional one, or the relatively cheap NCAA one, or even the free ones at high school. If your answer is “No” to all the above, I would argue that you watched at least one on live. You know the names, numbers, and outstanding statistics for some players that you have been watching. Probably even the coaches, referees, and streamers if you pay close attention to them.
But do you know the data analysis person? Or I probably should ask, do you realize that there are positions like that on the team? Here I am not talking about how many wins or how many touchdowns. I am pointing at data like psychological data and biomechanical data that are factors of practice plans and therefore performances during the game.
There is a group of people that apply scientific data analysis into sports to pursue the most efficient way of training that coming along with the highest level of performance during the game. I want to use the analogy of the film crew, but their names are always presented somewhere to the audience. Going to the team’s official websites, there is no information about this crew. However, working fully behind the scenes should never be the reason to wipe off any of their contribution to the team, to the game, and the field of data analysis in sports.
Sport Analysts Since Last Decade
Data analysis of sports performance is not a newly emerged field; since the early 2000s, a great amount of research has been done in varieties of sports to examine different dimensions of performance. Like many other early researchers in sports analysis, Sève, Nordez, Poizat, and Saury, early experts in applied sports, also focused on athletes’ performances in competition in 2011. Rowing’s highly automatic cyclical movement fits the data needed for analysis, so the researchers mainly concentrated on the coordination of rowing of a coxless pair crew. Coxless pair, a special competitive rowing with only one pair of rowers, requires accuracy in both coordination and timing.
To further analyze the dysfunction that existed, especially in crew coordination, researchers identified biomechanical characteristics of the rowers’ movements. They brought up both analysis approaches of the performance: athletes’ experiences during their performance and the biomechanical characteristic of their movement. The audiovisual data recorded the communication during the performance; the verbalization data was the interview right after the performance with their perception of how they feel, and the biomechanical with coordination was captured by cameras. Even without technology, it turns out that the “characteristics of the rowers’ coordination that were compatible with their perceptions but unsuspected by them and their coaches.” Perceptions hardly change during daily training unconsciously. Fortunately, related analysis brings awareness to the coaches and athletes themselves so that targeted training can be implemented to improve overall performance.
Scientific data analysis was known for its preciseness because it is based heavily on the mathematics behind it. This research was one of the best demonstrations of integrating humanistic into data analysis. Along with what the result suggested, it brought up the interest in “indexing an objective performance analysis to a prior analysis of athletes’ courses of experience.” At a first glance, analysis like this may violates the traditional understanding on scientific data analysis. No doubt that taking humanistic taking into consideration increase the uncertainty. But at the same time, we have to admit that it does raise the preciseness because unconscious psychological motivation is one of the factors of the athletes’ performance. Therefore most of the time, sport analysts analyzing comprehensive data is essential for both trainings and games (Sève et al, 2011).
It is not hard to see how much more competitive sports have became in this past decade. More creative practices, plays, and strategies contributed to those remarkable games. But there have to be people behind the stage doing all these works: find key places that need improvement, help coaches plan for efficient targeted practice, compare overall performance over time…Similar to rowing girls, sports analysts are also “cyclically” doing the analysis. This group of people care more than the physical statistics because they do care more about the psychological perceptions that lead to the performance.
Fast Developing Technology. What About Sport Analysts?
Later the field of sports analysis integrated with one of the most discussed topics in this era: informational technologies. Fast forward to the beginning of this decade, research was conducted to reveal the use of intelligent data analysis behind smart sports training. Rajšp and Fister, researchers from the University of Maribor, Slovenia, entered the field of sports with a high-level background in computational scientific studies. Different from other studies that concentrated on a specific athlete’s performances. They used a collaboration of 109 individual studies: from practices to actual games, from individual sports to team sports, from basketball to triathlon.
The purpose of this data analysis is eventually smart sports training that requires a huge amount of use of technology. Wearable devices, recorded sensors, and intelligent data analysis methods are all put in place during daily practice to track performance. Researchers broke the process of practice into four general phases: planning exercise units, conduct evaluation, compare actual and expected performance and evaluate overall performance in both the short-term and long-term. And virtual technology may come into play at each cycle of these four phrases.
The result for one of the most popular sports — basketball — reveals the application of the Apriori method. It is widely utilized for “identification of commonly used technical actions in basketball games.” This method is not limited to basketball, but also soccer and a variety of other sports. Falling into the category of the evaluation phrase in practice, it provides a crucial reference for the training.
However, technological analysis is also commonly used in live streams to provide an easy-to-interpret visual for audiences. In fact, if you watch basketball and volleyball for a fair amount, you probably have seen that several times without realization. Right after each quarter or set, the streamer would present a map with all attacking position on the court made by the top scorer. No doubt that’s the contribution of those informational technologies on the court. This research approach truly opened up the door to unlimited opportunities with data analysis in the field of sports.
It was mentioned in the research that “a coach is replaced by a smart agent which manages all the aspects of training, except for actually performing the proposed exercises for the trainee.” It seems stupendous that replacement would happen. But from another perspective, “the workload reduction can apply either to the athlete or his trainer. For an athlete, an improved training plan means he can achieve better results with the same or even less, amount of training, and for his trainer, this means the assistance of IT technologies can automate parts of his coaching routine.”
Nevertheless, what about sports analysts? If a coach can be “replaced”, wouldn’t an analyst also be “replaced” easily? Realizing that those informational technologies are applied in all four phrases of practice and even the game. The authors acknowledged that this research has not been widely transferring into the real-world along with its cooperation with trainers and athletes. Further down the road, integrating all data that was analyzed and put together the pieces together in the puzzle should always be a crucial role (Rajšp et al, 2020).
Technologies developed a lot in the last decade, they can do a lot more. But at the same time, limitations still applied. It was programmed like a computer, so it can understand the data recorded and analyze it as needs.
Sports analysts are working based upon the analyzed result. Technology reduces the workload and leaves more time for analysts to interpret the results, find possibilities, and draw a sound response to this result. In “Big data and tactical analysis in elite soccer: future challenges and opportunities for sports science,” Rein and Memmert, researchers from German Sport University Cologne, approach sports analysis with all sets of data. Except for the commonly known physiological data, they included psychological data, coaching data, and even crowd data to analyze performance from a comprehensive perspective. To make sense of the psychological data and crowd data, humanistic is required. No machine can better analyze humanistic than humans themselves (Rein et al, 2016).
Informational technologies and analysts are complements, no substitutions.
There Are Always More For Sport Analysts.
And of course, with the goal of pursuing better performances, factors to take into consideration has widened through years. Algorithmic bias along with social issues all have an influence on performance subtlety. Do you ever wonder why one team recruits a certain player? Reasons can be very straightforward: that player is one of the best players in the league, or the team needs a player on that position, or even because of their limited budget… Reasons does vary a lot, and all of the reasons affect performance more or less, especially for team sport. Analysts cannot ignore any of these to interpret the results and give conclusions.
Experience, language, and salary difference are three top ones that I pay very close attention to while watching volleyball. The United States did not have any professional volleyball league before 2021; that being said, all the professional volleyball players have to play overseas all the time except for international games. Thus, it is very common that an experienced U.S. volleyball are playing in a club with almost no experience in international games. Having them on the team is an honor because everyone else can learn so much, but what about the pressure they bring to the team at the same time? Sport analysts also need model with data to “predict” how to balance this out before making the transfer official.
At the same time, a lot of U.S. players end up playing in different countries, mostly in Turkey, Italy, Russia, and China. And no matter how good they are already; they still need to bond with other team members by communications. Being on a team with no one to talk to is almost no different from being in a cell in the prison by yourself. It would be necessary for analysts to evaluate players’ performance along with mental health and the efficiency of communication. According to experiences, a lot of teams tend to have two or more foreigners so at least there’s a bubble.
The main salary difference is based on skills, but position player is also another major one. In volleyball, a top setter earns only one-fifths of a top outside hitter. People obviously can argue that you cannot compare salary without clear measure their contribution to the team. With no authorized perspective to answer the salary difference, it is absolutely hard for analysts to interpret its effects on performance. However, this is unavoidable.
It is not hard to see that what sport analysts have to take into consideration in their data analysis are deeper down the surface that we can see. This group of people not only work fully behind the scenes, but also put tremendous efforts into all possible aspects in their analysis — from players themselves to the world around them.
Sports became popular to the general public 100 years ago, but the data analysis approach in sports just started to develop in the past two decades with minimal attention from the public. However, there are researchers from all fields who come together to share different perspectives and integrate all of them. And very fortunately we have a community of people who work as data analysts in the field of sports. A huge amount of work has been done by everyone behind the stage to present those unforgettable games. The next time you sit in a stadium and support your favorite team, please also cheer for this community and all their hard work.
Works Cited
Sève, C., Nordez, A., Poizat, G., Saury, J. (2011). Performance analysis in sport: Contributions from a joint analysis of athletes’ experience and biomechanical indicators. Scandinavian Journal of Medicine and Science in Sport, 23(5). https://doi-org.libproxy.lib.unc.edu/10.1111/j.1600-0838.2011.01421.x
Rajšp, A., Fister, I., & Jr. (2020). A systematic literature review of intelligent data analysis methods for smart sport training. Applied Sciences, 10(9), 3013. doi: http://dx.doi.org.libproxy.lib.unc.edu/10.3390/app10093013
Rein, R., Memmert, D. Big data and tactical analysis in elite soccer: future challenges and opportunities for sports science. SpringerPlus 5, 1410 (2016). https://doi-org.libproxy.lib.unc.edu/10.1186/s40064-016-3108-2
