Gillis is in the realm of multivariate statistics. Not your simple batter v. pitcher. But rather shooter vs. goalie, oppostion opponents, ice conditions, coaching strategy, # of game in the season, time remaing in the game, linemates, postion drafter of shooter vs. other draftees, mascot attire, towel waving present, is the puck frozen according to Darren Millard standards , etc….

So you see your “binary” analysis is that of a simpleton. Mike Gillis has mastered multivariate statistics and ultimately will master the NHL. Note – he has predicted a comeback for NYR vs. PIT. To the laymen this is absurd, but to Mike Gillis and his multivariate statistical model it is not. He has factored in that the loss of Sean Avery, the inexperience of the Penguins, the poor ice, the lack of adequate fan support in an small Igloo, Sidney Crosby and Malkin’s performance in previous conference semi-finals, Marc Andre Fleury’s save percentage with his old pads vs. his new and has come up with a Rangers win.

RJ- Trust the model, trust the GM. I mean if I could predict the future based on past precedent I would become a GM and not a billionaire stock investor. It just makes sense. Smarten Up.