Figure 8
A line graph titled goaldiffpp- All leagues and Top-5. The horizontal axis represents the coefficients ranging from -0.004 to 0.001. The vertical axis lists different player categories and statistical measures, including Mean Min, Median Min, Mean Min Top 5, Median Min Top 5, Mean Min G K, Median Min G K, Mean Min Top 5 G K, Median Min Top 5 G K, Mean Min D F, Median Min D F, Mean Min Top 5 D F, Median Min Top 5 D F, Mean Min M F, Median Min M F, Mean Min Top 5 M F, Median Min Top 5 M F, Mean Min A T, Median Min A T, Mean Min Top 5 A T, and Median Min Top 5 A T. Each line represents a different category or measure, with varying trends and data points. The lines show the estimated coefficients from Poisson regressions of average goal difference per match on seasonal workload proxies, indicating an association with more balanced matches for negative coefficients.Coefficients of Poisson regressions, impact of different minutes on Total Goals Difference per match Note: The figure reports estimated coefficients from Poisson regressions of average goal difference per match on seasonal workload proxies. Workload is measured as mean or median minutes of the 50 most-utilized players in the previous season. Models include tournament-type controls. Negative coefficients indicate an association with more balanced matches (lower goal differences). Source: Authors' elaboration