Theoretical physics wouldn't exist without experiments that back it up. In particular, for a theory to be relevant, it has to correctly predict the outcomes of all past experiments. In contrast with math, where you can easily make up "worlds" (models) that have no basis in reality whatsoever.