Exploring the feasibility of a novel experimental method to study talent selection and decision making in high performance sport

Abstract

Despite athlete selection occurring from youth to professional levels of sport, identifying and selecting athlete potential is often difficult and inaccurate (Jones et al., 2020; Koz et al., 2012). The aim of this study was to test a novel experimental method for studying talent decision making in sport. Using blinded historical data from twenty National Basketball Association (NBA) draftees, participants were asked to make talent predictions based on three outcomes: draft position (lottery pick, first round non-lottery, second round pick, or undrafted), length of career in years (0-1, 1-3, 4-7, 8-11, 12-15, and 16+), and whether players would achieve an All-Star/All-NBA selection. Participants (n=11) from various levels of basketball coaching in Canada completed the task. Overall, participants were less likely to correctly identify lottery picks (OR: 0.38, 95% CI: 0.22-0.65), non-lottery first round pick (OR: 0.27, 95% CI: 0.14-0.53), and second round picks (OR: 0.22, 95% CI: 0.11-0.42), but were more accurate at identifying undrafted players (OR: 0.59, 95% CI: 0.38-0.92). Odds ratios for length of career predictions also reflected a lower likelihood of correct predictions (OR range: 0.15 to 0.32). Comparatively, participants were more accurate at identifying athletes that would not be selected for All-Star/All-NBA teams (OR: 0.82; 95% CI: 0.62-1.08). This pilot study suggests that this method can provide insights into talent selection decision making processes, such as estimations probability, decision-making accuracy, inter/intra individual variability, and cognitive associations between variables.