EDIT I realized reading the comment that my first message was not precise enough. I do not have a specific distribution in mind, my initial idea is to adapt the Maher model for the number of goals in soccer matches. In this model the number of corners of team $$T$$ vs team $$T'$$ is assumed to followed a Poisson distribution with Parameter $$a_T d_{T'}$$ where $$a_T$$ and $$d_T$$ are the attack and defense scores of team $$T$$ which are estimated from the data using macimum likelihood.