This is done by estimating the effect of both chemistryand COVID on match results and analysing the performance of prediction models where both effects are included and threshold intervals are usedfor classification. Four different chemistry measures are introduced and all are evaluated.
Chemistry has the expected positive effect on performance only for the top teams in the estimations where interaction effects are included for two different chemistry measures. COVID has the expected mitigating effect on home advantage.
The inclusion of both effects in prediction models does not increase prediction accuracy consistently, although for various symmetric threshold intervals the prediction models with chemistry and COVID included outperforming the baseline models.
See how this article has been cited at scite.ai
scite shows how a scientific paper has been cited by providing the context of the citation, a classification describing whether it supports, mentions, or contrasts the cited claim, and a label indicating in which section the citation was made.