Interpersonal coordination in tennis: Assessing the Positional Advantage Index with Australian Open Hawkeye data

Van Meurs Edda, Buszard Tim, Kovalchik Stephanie, Farrow Damian, Reid Machar

Abstract

Dynamical systems theory provides researchers with statistical tools to explore nonlinear relationships that evolve based on spatio-temporal relationships between performers. In tennis, the Positional Advantage (PA) Index identifies a player who has an advantage to win the point as a consequence of their position on the court relative to their opponent. However, the PA Index was initially developed based on a small sample of professional tennis matches coupled with intuitive reasoning. Additionally, we believed the PA Index could be improved by applying weightings to each player’s lateral and longitudinal displacements, and by taking into account contextual factors (player orientation and identification of the player impacting the ball). Therefore, we assessed and sought to improve the PA Index by using a larger sample of empirical data from the 2017 Australian Open. The best equation was determined by assessing the correspondence between PA Index values and the outcome of a point. Despite our best efforts to improve the PA Index, results confirmed that the original equation was the best measure of positional advantage. Our discussion focuses on the value of the PA Index for understanding tennis performance.