Deflategate favored foul play over science

Hopefully after next week we will never again hear about Deflategate, the controversy surrounding the role of underinflated footballs in January’s conference championship game between the New England Patriots and the Indianapolis Colts. On June 23, the NFL commissioner will hear the appeal of Patriots quarterback Tom Brady’s four-game suspension, one of the punishments that resulted from the controversy. Patriots’ team equipment managers may have intentionally underinflated game balls and Brady may have known about it — I won’t weigh in on that here. But the scandal, which propelled the ideal gas law to the front pages of sports sections, inspired an odd mix of experts to choose science over sports, and that’s almost always a win.

In case you haven’t followed the story: During the first half of the January 18 AFC championship game in Foxborough, Mass., the Colts intercepted a pass thrown by Brady. Suspecting that the ball was underinflated — rules allow a pressure range of 12.5 to 13.5 pounds per square inch (psi) — the Colts requested an inspection. Brady is known to prefer his footballs on the low end, around 12.5 psi, and pretty much everyone agrees that that’s what the Patriot balls were inflated to before the game started. But at halftime, officials tested the Patriots’ game balls: All measured below the minimum required level of 12.5 psi.

Patriots coach Bill Belichick was one of the first to take a stab at the science, citing “climatic conditions,” “equilibrium states” and “atmospheric conditions,” to explain the deflation. This wasn’t a surprising stance; he’s the coach. But it didn’t take long for real scientists — and non-Patriots fans — to weigh in. While Bill Nye (mechanical engineering degree, Seattle Seahawks fan) declared that rubbing the footballs to break them in couldn’t account for pressure changes, others took a measured approach. If the initial pressure of a football measured in a warm locker room during pre-game inspection was 12.5 psi, could the roughly 25-degree-Fahrenheit drop in temperature between the locker room and the rainy field that day account for the lower air pressure of a ball measured at halftime?

Scientist Michael Naughton (expert in condensed matters physics, Buffalo Bills fan) lent his expertise to the matter when the controversy initially blew up. Naughton’s lab at Boston College inflated a football to 13.5 psi at 72° F. Then they stuck it in a fridge and measured the pressure at 42° F (slightly cooler than the low on game night of 47.7° F, the average of measurements from two weather stations near Gillette Stadium). The pressure dropped to 10.5 psi.

HeadSmart labs, a Pittsburgh-based engineering firm that ordinarily conducts research related to helmets and concussions, also turned its attention to the matter. Experiments done by CEO Tom Healy (mechanical engineering Ph.D. student, Patriots fan) and others in the lab (not Patriots fans) simulated field conditions by placing 12 balls inflated to 12.5 psi in a cold room for 2.5 hours. Measurements revealed an average drop of 1.07 psi, well within the range of the halftime measurements. Saturating the balls with water to mimic field conditions bumped the measurements down another 0.75 psi, they conclude in a technical paper. (HeadSmart has launched a crowd-funding campaign to raise research funds to further investigate the matter.)

The kerfuffle provided a teachable moment for physics teachers everywhere, and despite Deflategate fatigue, homework problems featuring the ideal gas law — which relates temperature, pressure and volume to an amount of a gas (in moles) — will likely be assigned for years to come. This science matters well beyond the football field: Understanding the gas law means knowing whether a scuba diver will experience potentially fatal bends when returning to surface waters, why life-saving contraptions like fire extinguishers and airbags work, and how hot air balloons and combustion engines do their stuff.

But instead of acknowledging that game day conditions could have accounted for the psi changes, an acknowledgement that wouldn’t preclude other evidence of foul play, the NFL’s Wells Report concludes that there’s an “absence of a credible scientific explanation for the Patriots halftime measurements.”

It would be one thing if the Wells Report (which consulted Daniel Marlow, experimental high energy physics expert at Princeton) just said that additional evidence (bathroom breaks and text messages, among other things) was more compelling than the pressure data. Or if it noted that the pressure data are ambiguous, collected so haphazardly that they wouldn’t be allowed in a high school science fair: Two different gauges that differed by approximately 0.4 psi were used in taking measurements, and it isn’t clear which one was used in the pre-game measurements because those data were not recorded. At halftime, 11 Patriots’ balls and four Colts’ balls were measured, and while all of the Patriots’ balls measured below 12.5 psi, three of the four Colts’ balls also did, according to one of the gauges.

Post-game psi measurements of four Patriots balls ranged from 12.95 to 13.65. These data, the Wells Report acknowledges (in a footnote), “did not provide a scientifically reasonable basis on which to conduct a comparative analysis.” If the report can acknowledge poor methodology for the post-game data, why not acknowledge that for the pre-game and halftime data as well?

Roderick MacKinnon of Rockefeller University specifically addressed the scientific methodology in a letter posted to The Wells Report in Context, the Patriots’ rebuttal to the report’s conclusions (MacKinnon, professor of molecular neurobiology and biophysics, and chemistry Nobel laureate, was conducting experiments in a basement microscope facility and couldn’t immediately respond to my requests for his team allegiance):

“The scientific analysis in the Wells Report was a good attempt to seek the truth, however, it was based on data that are simply insufficient. In experimental science to reach a meaningful conclusion we make measurements multiple times under well-defined physical conditions. This is how we deal with the error or ‘spread’ of measured values,” MacKinnon notes.

Football fans are a loyal bunch. (Let it be said that I live in Boston and while I appreciate a quarterback who can make fun of himself, I do not have a favorite football team.) But it’s refreshing to see some put aside team loyalty in favor of Team Science.