Elite women's soccer match demand can be described using complexity-based analyses and multifractals

The present study aimed to investigate the variability of the displacements in elite women soccer players during official matches according to the position of the players and to the match progression using non-linear analyses (entropy and multifractal spectrum). Sixty-three velocity profiles from global positioning devices of elite professional women soccer players were analyzed. The mean, coefficient of variation, and maximum velocity were calculated during the match using four equal segments. Sample entropy and seven parameters derived from the multifractal detrended fluctuation analysis were also calculated. The players were split into five field playing positions: forwards (FW), offensive midfielders (OM), defensive midfielders (DM), central defenders (CD), and lateral defenders (LD). No significant playing position x period interaction was found for the different parameters analyzed. However, most parameters showed a significant main effect for playing position (p < .01) and period (p < .01). Compared to the other players, FW and OM have the lowest entropy. All players showed a dominance of small velocity fluctuations but FW players showed the widest multifractal spectrum primarily. In contrast, CD had a narrow spectrum and reduced amount of small fluctuations. As the game progressed, mean velocity and entropy significantly decreased. In contrast, the multifractal peak value significantly alongside with changes in asymmetry. From the present results, velocity profiles during elite women's soccer games depicted multifractal features sensitive to player position and match segment, which can help to understand the dynamics and variability of players' responses during a match.