During 2020-2021, many hockey leagues had shortened seasons or no seasons due to the restrictions on play caused by the COVID-19 pandemic. The Ontario Hockey League did not play any games during this “COVID” season, which raises the question of whether not playing games during the 2020-2021 COVID season negatively impacted player development (or even caused players to get worse). Statistical and causal inference methods were employed to assess the impact of not playing during the COVID season, including Ordinary Least Squares, Gamma, Mixed Effects, and matched regression models, as well as a Bayesian Additive Regression Trees Model. All models pointed to no significant difference in post-COVID performance for treated (played during COVID season) vs control (did not play during COVID season) groups. Future research should focus on studying other populations, experimenting with alternative variable definitions, and studying player development curves to more thoroughly assess treatment impact.
Check out the journal article for the full story!