Last summer, readers Eric Isaacson and Catey Tarbell proposed this simple prediction algorithm: Best Record Wins; If Records Equal, Home Team Wins. Their idea, which TMQ dubbed the Isaacson-Tarbell Postulate, ruled the landscape of NFL predictions, finishing the season 183-84, or 69 percent correct. Of the dozens of NFL predictors TMQ tracked, John Czarnecki of Fox Sports had the best finish, at 183-84; ESPN's best predictor was Keyshawn Johnson at 176-91. So the Isaacson-Tarbell Postulate outperformed all but one football expert, which it tied. And here's the beauty of the Isaacson-Tarbell Postulate: You don't need incredible insider information; you don't need to spend hours in fevered contemplation; you don't even need to know who's playing. Simply always check-mark the team with the best record, or, if their records are equal, check-mark the home team. Whatever you do, don't think! I conducted a parallel experiment of forecasting all games using the Isaacson-Tarbell Postulate unless I felt really sure about a game going differently. A few times I was right where the Postulate was wrong, but on the season, this approach finished 179-88. That is, thinking about who might win was less successful than sticking to the formula. When making NFL picks, never think!And don't bet on sports.
Feb 12, 2008
the Isaacson-Tarbell Postulate and you
Looking for as close to a surefire NFL prediction scheme as you're going to find? Gregg Easterbrook explains: