The Shot Chain

Thibaud, Mikael, Tyrel and I teamed up again to analyze the shot component of our Markov decision process as developed at the Stathletes' Big Data Cup. In this presentation we go beyond the traditional expected goals model by building out a chain of Bayesian logistic models via INLA to analyze shooter talent and the context surrounding a shot.

Beyond this, we develop an adaptable expected goals models that allows the user to easily transition between including all shot attempts, unblocked shots or shots on goal.

Additionally, we demonstrate how to apply these models by analyzing NWHL and International women's hockey players and gameplay.