A comparison of running intensity during professional football training and matches and the impact of running intensity on the expected goals performance outcome

The two main aims of this investigation were to identify whether running intensity in training compares to the running intensity in matches for a professional football team and whether running intensities impacted the expected goals (xG) outcome of the team during a professional football match. The participants, 27 professional football players at Gillingham Football Club, were recorded during training and matches using GPS during the 2022/23 season. Training distance per minute and high-speed distance (> 18km/h) from medium- and large-sided games were collected, these metrics were also collected during competitive matches alongside xG. Following data collection in-depth statistical analysis was concluded, distance per minute covered presented no significant difference between training or matches when considering positions (P = 0.57). Match distance per minute was significantly higher than training when all playing positions were aggregated (P = 0.03). Defenders covered significantly less distance per minute than midfielders (-8.43m.min-1, P = <0.01), and forwards (7.33m.min-1, P = 0.02). High-speed distance was significantly lower in training than matches by position (P = <0.01), and significantly lower when positions were aggregated (P = <0.01), and when positions were compared against each other midfielders cover significantly more than defenders (P = 0.04) and forwards (0.01). Mean distance per minute aggregated across position showed no significant correlation with xG scored (P = 0.97), this was the same with high-speed distance (P = 0.48). Forwards, midfielders, and defenders showed no significant correlation for distance per minute and xG (P = 0.59, P = 0.51, P = 0.88, respectively). Forwards and defenders highspeed distance had no significant correlation with xG (P = 0.35, P = 0.41, respectively). However, for midfielders high-speed distance there was a significant correlation with xG (P =<0.01), although an R2 value of 0.34 suggests a weak correlation. In summary, training distance per minute produces similar values to match data when considering position, implying it is an effective training tool to prepare for the demands of a match. There was no impact of running intensity on xG performance of the team, although there was a weak positive correlation identified between high-speed distance of midfielders and xG performance.