More on Erkut's "More on Morrison and Wheat's 'Pulling the Goalie Revisited' "

Through the use of the well-known Morrison/Wheat/Erkut ice hockey model, we demonstrate that even optimal solutions may be open to question. Based on the assumptions which are made to estimate data values, optimal solutions may vary among decision makers. We also note that under any reasonable set of assumptions, hockey coaches wait too long before pulling their goalies. tion made by Erkut is a major contribution to ice hockey as we dread the lumping of our Broad Street Bullies with other teams. While we very much endorse the generalization of the formula I n ancient times, Morrison [1976] preXsented the seminal analysis on the optimal time to pull a goalie in an ice hockey game. More recently, Morrison and Wheat [1986] realized that the original paper contained a major error and published a revised, corrected paper. Most recently, Erkut [1987] generalized the Morrison/Wheat (MW) model so that his students would not feel "uneasy" about the Morrison and Wheat assumption that essentially lumped Erkut's NHL Champion Edmonton Oilers with Morrison's beloved and improved Detroit Red Wings, We, too, feel that the generalizafor the optimal time to pull the goalie as given by