Training load comparison between small, medium, and large-sided games in professional football

This study aimed to assess if internal and external load parameters were different between sided game formats, if players' positions influenced these parameters, and if load parameters were different among sided game types (from 2vs2 to 10vs10) in professional football players. Twenty-five male players of the same club were enrolled in this study (age = 27 +/- 9 years and body mass = 78 +/- 14 kg). Sided games were categorized in formats as small-sided games (SSG, n = 145), medium-sided games (MSG, n = 431), and large-sided games (LSG, n = 204). Players were divided into roles such as center backs (CB), fullbacks (FB), center midfielders (CM), attacking midfielders (AM), and strikers (ST). STATSports 10 Hz GNSS Apex units were used to monitor external load parameters such as distance, high-speed running (HSR), sprinting distance, accelerations, and decelerations. The linear mixed model analysis found differences between formats (p < 0.001) for the rate of perceived exertion (RPE), distance, HSR, sprinting, accelerations, and decelerations. Differences were found between positions for HSR (p = 0.004), sprinting (p = 0.006), and decelerations (p < 0.001). Moreover, a significant difference was found between sided game types (p < 0.001) for RPE, distance, HSR, sprinting, accelerations, and decelerations. In conclusion, some sided games formats are more suitable for specific load-specific parameters (e.g., distance per minute, HSR, and sprinting are greater during LSG). The number of accelerations and decelerations is higher in MSG compared to other formats. Finally, players' positions influenced external load metrics, specifically HSR and decelerations but not RPE and distance.