In modern football, various variables as, for example, the distance a team runs or its percentage of ball possession, are collected throughout a match. However, there is a lack of methods to make use of these on-field variables simultaneously and to connect them with the final result of the match. We consider data from the latest season of the German Bundesliga. The objective is to identify the on-field variables that are connected to the sportive success or failure of the single teams. A paired comparison model for football matches is proposed that is able to take into account on-field covariates. The model extends the classical Bradley-Terry model  to ordinal response variables and to include so-called subject-object-specific covariates. Specific L₁ penalty terms for fusion and selection  are used to reduce the complexity of the model and to find clusters of teams with equal covariate effects. The proposed model is a very general one and can easily be applied to other sports data or
to data from different research fields.