One of the best basketball players of all times, Nikola Jokic was famously drafted by the Denver Nuggets during a Taco Bell commercial. Picked 41st in the second round of the 2014 draft, he has since surpassed all expectations, becoming a three-time league MVP and one of NBA’s biggest stars.
This begs the questions of how coaches, scouts and sport administrators decide which athletes might succeed and which might not? Researchers from the University of Toronto Faculty of Kinesiology and Physical Education (KPE) conducted a pilot experiment to find out.
“It could be argued that one of the most difficult predictions a person makes is what another person’s future will look like,” says Kathryn Johnston, a senior research associate at KPE, who worked on the research project with Joseph Baker, a professor at KPE and the Tanenbaum chair in sport science, data modelling and sport analytics. “To better understand this question, we designed an online approach – similar to a dating app, and tasked coaches with making mock selections.”
The coaches were asked to state their preference between athletes labeled as hard workers or natural talents, rank the most important variables for success from a provided list, and imagine a scenario in which they have an empty roster to fill.
The coaches saw 15 athlete profiles, including pictures, anthropometric information (the measurements and proportions of the athletes’ bodies), a description of the athletes’ interests, abilities and testing statistics. They were then asked how likely each athlete was to be successful and how likely they were to select them to their hypothetical team.
Findings from the pilot experiment, recently published in the journal of Sport, Science and Performance Psychology, show a slight lean in the coaches’ selection behaviour towards athletes who are labelled as hard workers. The coaches also preferred qualities such as anthropometrics, passion and commitment to sport, speed and explosivity.
When it came to defining success and talent, that’s where the researchers observed the most variety in the responses, getting nearly 18 different definitions for each of these terms from the 18 coaches in the sample. The coaches also weighted the information sources in unique ways, each valuing various aspects of the players’ information differently.
“For some, profile pictures were critical, for others, it was the birthdate, etc.” says Johnston.
According to the researchers, these findings have the potential to help illuminate preferences, biases and tendencies when making athlete selection decisions, which can help scouts, coaches and other selectors determine their alignment with organizational priorities and highlight any blind spots in the athlete assessment and selection process.
“For example, if the mandate of the organization is to select athletes who have characteristics X, Y and Z, the results in this report could indicate that coaches do - or don’t - prioritize athletes with those qualities,” says Johnston.
Following the pilot experiment, the researchers are looking forward to examining decision-making behaviours in athlete selection on a much larger scale.
“We are going to use a similar design to try and determine whether selectors are selecting in line with their stated preferences or otherwise,” says Johnston. “Beyond that, we hope to create adaptations of this experiment to better understand how subtle language changes might actually influence selection behaviour. “