Evolution of Analytical Data in Sports
Analytics in sports hasn’t always been pretty. Long ago, if you were interested in seeing how your favorite athlete was doing this season, you’d probably have to open up the newspaper to the ‘Sports’ section or peruse the most recent edition of some Sports magazine. Heck, you might even have to buy a pack of baseball cards.
But now, things are different. Way different. Statistics and data in sports have become commonplace. The average fan can visit websites pull data from databases in almost every sport or even view analysis done by professionals on websites such as FiveThirtyEight.com. We can view Mike Trout’s on-base percentage against right-handed pitchers in a home game. We can see Stephen Curry’s three-point percentage in the final moments of a playoff game. We can even find a New York Times Bot that makes recommendations on when to go for it on 4th down.
Franchises such as the Houston Rockets and the Oakland Athletics have been at the forefront of bringing advanced analytics into the Sports world. Billy Beane of the Oakland Athletics focused on sabermetrics (just a fancy term for baseball statistics) to obtain the one of the most cost-effective teams in baseball. Daryl Morey of the Houston Rockets spearheaded the analytics movement in basketball and completely evolved the way the game is played. As shown in the chart below, the Rockets were way ahead of the league when it came to understanding the efficiency and value of a 3 pt. shot. Whether it’s ‘Moneyball’ or ‘Moreyball’, a lot of these concepts seem obvious now that we look back a few years. However, at the time, they were considered different and others were years off from adoption.
Look at the chart below. In the past twenty years, teams across all of the NBA have adopted widely adopted the use of analytics to better develop more efficient game styles. The midrange shot that we all grew up watching in Michael Jordan and Kobe Bryant are now considered “bad shots” unless taken by those that are of top efficiency levels there. In baseball, teams have begun to employ the shift in baseball. In football, teams have decided to go for it on fourth down conversions with far more frequency.
Scouting younger talent has also become much easier as a result of modern AI. Scouts now have access to data that show how young basketball players perform in certain in-game situations or how fast the swing speed is of a young baseball player. What had to be done in the past with instincts, the eye-test, and some pen and paper is now done with algorithms. Now, we no longer need the a video coordinator to comb through hours and hours of video footage. The development of AI and machine learning allows us to gather advanced statistics that the human eye would have otherwise missed. This new dataset also assists in better forecasting the future projections of the players as well.
Injury prevention may be one of the biggest benefits from recent trends of growing data. Teams are partnering up with tech companies in order to track player movements and analyze their tendencies that may lead to injuries. Technology, like force plates, can show if a play has a tendency to lean or twist in certain directions that might pose as risk in game situations. Organizations are also looking into monitoring the stress levels on the bodies of players and tracking the stresses of practice and gameplay. Preventing injuries not only helps to better team performance but also increases revenue as teams sell more tickets.
When it comes to sports, there is a fine line between the analytical and the intangibles. Traditionalists say that there was a lot more purity in the way sports were played back then and that with all this new-age data, the “nerds” have ruined the game. A lot of the traditional athletes tend to believe that the use of “computers” is killing the game. The biggest issue with pressing athletes with data-driven recommendations is that the recommendations are most likely coming from those who haven’t been in their shoes (more specifically their sneakers or cleats). Recommendations are easy to make when you’re not staring at whizzing 98 mph fastball or a 250 lb. super-athletic defensive end running full speed in your direction. That is why perhaps, it is of utmost importance that these two groups work in tandem to make decisions with quantitative and qualitative reasoning.
Regardless of what traditionalists desire, the expansion of data and analytics in sports has taken place and will continue to take place in the coming years. Also with the adoption of better technology such as wearables, analysts are collecting more and more data, even expanding out to the certain psychological behaviors of players in certain situations. However, data is only useful when coupled with the proper analytics. It is only then that we can make the proper decisions.
Scouting. Injury prevention. Coaching decisions. Team composition. Individual player management. The list goes on. The amount of data being collected on sports grows day by day. The future of data analytics in sports will not only help to increase the profits of owners and lengthen the careers of players but also help to grow the sport itself.