Sports News Without Fear, Favor or Compromise
Sports News Without Fear, Favor or Compromise

# Wells Report Has Nine Page Statistical Analysis Of Ballghazi: LOL

Ballghazi is a perfect, perfectly hilarious scandal, and it just keeps getting better. To wit: The NFL, having ducked and dodged the many, varied complaints about the criminal behavior of its players, its own far-reaching municipal extortion schemes, and the long-term health effects of the game it promotes, has now released an independently gathered 243-page report on the matter of Tom Brady’s balls and their consistency during last year’s AFC championship game. Further, and somewhat surprisingly, the report comes stocked with nine pages of what seems at least meant to be taken as in-depth statistical modeling. Apparently, someone wants to leave us with the impression that the NFL’s investigative arm brought techniques generally reserved for Serious Scientific Issues like global carbon emissions or a batter’s expected BABIP as it correlates to True FB% to bear here.

As you might already know—regrettably, because no one should ever commit these things to memory, but NFL fandom often requires very specific, very stupid trivia—the NFL requires balls to be between 12.5 and 13.5 pounds per square inch gauge (psig) during use in a game. Before the AFC championship game, the report explains, the Patriots’ balls were right around 12.5 psig, and the Colts’ were 13 psig (or possibly, since it wasn’t written down, 13.1). By halftime the differences in the two teams’ ball inflation was statistically significant. Hard, or at least strongly circumstantial, evidence of real misbehavior has been sniffed out by the NFL, you are meant to conclude. Here is the model used, expressed in its additive terms:

Impressive! Here it is expressed mathematically:

Its terms are defined in the full document, embedded below, but probably this is plenty good enough to shock and awe the typical slack-jawed NFL dingus, an investigator would have to assume. Anyway, the model, which is a multiple linear regression and seems well done enough, says, basically, that some shit happened.

After accounting for the differences between two measurement instruments, and discovering that two officials apparently switched instruments at halftime—this takes up eight tables—here is a final table, with a summation of numerical results from the analysis: