Pythagorean Win-Loss: RS^1.83/(RS^1.83 + RA^1.83) Team, Actual Wins, Pythagorean Wins Yankees, 98, 96 Indians, 82, 89 Tigers, 83, 85 It does a good job predicting the wins for the Yankess and the Tigers, but it misses on the Indians. So I went looking for why... Clearly they scored a lot of runs, they either score runs in large chunks, or there is something runs allowed isn't showing us that cost the Indians a few games. I think it's the chunk theory... http://www.baseball-reference.com/teams/CLE/1923_sched.shtml Their WPCT (.617, 29-18) in blowouts is 8% higher than their WPCT overall (.536, 82-71) which is 8% higher than their WPCT in one run games (.457, 21-25). They also, by my count, were 11-4, .733 in games decided by 9 or more runs, including wins of 27-3 and 22-2. If you could shuffle around those runs in those huge wins, it'd make up the 7 games the pytagorean WL overshot by.