“The eye test” is the principle of judging athletes’ talents and potential based on watching their performances alone. This assessment was accompanied by looking at basic stats: how many assists a player accrued during the playoffs, or how many walks a pitcher threw in the postseason. Stats like these provided a running count of little successes or shortcomings of players’ performances. Now consider this equation:
The image above shows a formula which outputs a basketball player’s efficiency. It takes into account points, assists, rebounds, blocks, turnovers, field goals made and missed. It differentiates between 2-point shots, 3-point shots, and free throws. It compares a players’ performance against the team’s performance, as well as with the league’s average. This equation, now ubiquitous in the front desk of general managers and coaches, describes a player’s efficacy with reproducible, and non-arbitrary assessment. Nikola Jokic, the frontrunner to win the NBA’s Most Valuable Player award used to surprise casual fans. How he was able to produce stellar performances relatively quietly, fans only realizing the amount of points, rebounds, and assists he was able to get after the game had already finished. From his earthbound leaping ability to his unassuming physique, his type of play attracted few in the beginning of his career. The eye test would not have been able to predict his current contention to be the MVP of the league. The cognitive dissonance his rise to superstardom produces has been a widely popular meme. His incredible PER is now a common argument in support of his MVP campaign.
The Effects of Embracing Data
If you’ve ever seen Moneyball, then you’re familiar with the story of the Oakland As 2002 successes due to the team’s application of SABRmetrics. In short, the As could not compete with big market teams’ budgets, so they opted to implement an empirical analysis approach to acquire cheap talented players. The results of which saw the As surprise the league by improving their record even after the departure of their three star players. Since then, the proliferation of data science occupation has not only accelerated in baseball, but has since spread to other sports.
A popular example is Stephen Curry’s impact on how the game of basketball is played today. The image below shows how traditional basketball is structured on the court. The mean height of players in the NBA is 6 foot 7 inches. Point guards are usually the shortest of the players on the court, whose primary responsibility is to distribute the ball to his teammates. Shooting guards shoot from the outside, small forwards continually cut for openings towards the basket, and power forwards play closer to the basket. The tallest player on the team, the center, occupies the area closest to the basket. This archetypal model relied on presumed abilities players should have based primarily on size. Along came Stephen Curry.
His prolific scoring from behind the three point line combined with his incredible PER was something of an epiphany for the rest of the NBA. The team he played for, the Warriors, won three championships in 2015, 2018, and 2019. Prior to his success, coaches dismissed three point shooting as the model organizations can build championship teams around. The NBA introduced the three point line in 1979. Because players’ efficiency from the three point line seldom rose above 33%, many teams considered the play style to be inconsistent and unreliable. Between 2008 and 2012, it appeared that the amount of three point shots to regular two point shots had finally reached an equilibrium. For five straight seasons, 1 out of 5 shots was a three on average. Currently, that ratio now stands at 1 out 3 shots. The quick adoption of the three is due to Stephen Curry’s display of efficiency with a play style considered to be inefficient. Through the work of data scientists in NBA organizations, teams began to realize that the most efficient shots in the game were in fact three point shots and lay ups. Moreover, teams began to see the impact of someone like Stephen Curry on a team beyond just putting up numbers. By quantifying his “gravity” or on-court presence, data scientist were able to concretely show how the three point play style improved spacing and pace. Basketball teams are now using data analytics to not only pick players and construct play schemas, but who and when to rest players.
There are those who hold disdain towards the application of data analytics in their favorite sport. Some may argue that “it takes the fun out of it”. There are also some who find ire in the way fans and players are able to compare current players with past ones using said analytics. Arguments often rely on discrediting the analyses by pointing to the dissimilarities between different eras of play. Ultimately, these criticisms do not adequately describe the effects of data analytics on both player performance and team winningness.