Paul Crowley comments on Evidential Decision Theory, Selection Bias, and Reference Classes