Understanding the physiological demands placed on different positions among professional soccer players during matches can assist with individual and group training to maximize performance. Furthermore, understanding players' physiological responses each half may help coaches make in-game adjustments and substitutions. Purpose: To examine the differences of in-game physiological performance metrics between positions and halves in professional soccer players. Methods: 18 third division professional male soccer players categorized as Defender (D: n = 7), Midfielder (M: n = 6), or Forward (F: n = 5) were monitored with an individual GPS and heart rate bioharness during the course of a season. Bioharness metrics of total distance (TD), maximum speed (MS), sprint distance (SD), number of sprints (#S), average heart rate (HRavg), peak heart rate (HRpeak), and calories burned (CAL) were used for analyses. Matches were separated into first (H1) and second half (H2) for all analyses. Only complete half data, where players were not substituted off during the half, were included for analysis. A two-way MANOVA was run examining Half and Position on all physiological variables of interest. Post-hoc Bonferroni analyses were used on all significant main effects (p = 0.05). Results: There was no statistically significant interaction effect between Half and Position on the combined dependent variables, F(14, 216) = 0.79, p = 0.69; Wilks' Lambda = 0.91. However, a significant main effect was found for Half (F(7,108) = 5.40, p < 0.01; Wilks' Lambda = 0.75). Players had higher TD (0.36 ± 0.07 mi [mean difference], p < 0.01), HRavg (8 ± 2 bpm, p < 0.01), SD (0.04 ± 0.02 mi, p = 0.03), #S (3.3 ± 1.2 sprints, p = 0.01), and CAL (65.3 ± 12.2 kcal, p < 0.01) in H1 compared to H2. Neither MS (p = 0.93) nor HRpeak (p = 0.12) were significantly different between halves. A significant main effect was also found for Position (F(14,216) = 16.50, p < 0.01; Wilks' Lambda = 0.24). Post-hoc analyses revealed that M had higher TD than D (0.26 ± 0.08 mi, p = 0.01) and higher HRpeak than D (6 ± 2 bpm, p = 0.02). M had lower MS than D (-2.26 ± 0.31mph, p < 0.01) and F (-1.85 ± 0.34mph, p < 0.01). F had higher SD (0.07 ± 0.02 mi, p < 0.01; 0.08 ± 0.02 mi, p < 0.01) and #S (6.6 ± 1.4 sprints, p < 0.01; 7.4 ± 1.5 sprints, p < 0.01) compared to both D and M, respectively. Conclusions: Results indicate that players were under higher physiological stress and performed more specific tasks (TD, SD, #S) during H1 compared to H2. Additionally, the demands placed on the players during a match are unique to the position that they are playing, as F are tasked with more sprinting and M are asked to cover more distance as a result of their roles on the field. Lastly, all players, regardless of position, experienced similar declines in physiological responses and performance measures as the game progressed. Practical Applications: Coaches and training staffs should look at the unique demands of each position when designing conditioning and practice programming to best maximize player performance based on position. Staffs should also note the potential decreases in physiological stress and performance that happens regardless of position from H1 to H2. This could indicate that fatigue or tactic changes impact performance variables and subsequent physiological responses in players as games progress.
© Copyright 2021 The Journal of Strength and Conditioning Research. National Strength & Conditioning Association. All rights reserved.