If McNally changed the pressure of the footballs, the data suggests that one of two things is true:
a) he took out about 0.26 psi of air on average, thus getting the 1.39 psi average pressure drop predicted by one of the two gauges that might have been used, vs. a 1.13 psi drop as predicted by the ideal gas law. This ignores the effect of water on the footballs, which probably should not be ignored, and could account for all of the "extra" 0.26 psi drop
or
b) he put in about 0.12 psi of air on average, thus getting the 1.01 psi average pressure drop predicted by one of the two gauges that might have been used, vs. a 1.13 psi drop as predicted by the ideal gas law.
Yes, the data does not even make it clear whether he took air out, put air in, or neither.
The officials measured the drop in pressure four of the Colts footballs. Using one gauge they measured a drop of 0.37 psi. Using another gauge they measured 0.56 psi. Average of the two gauges: 0.46 psi drop. The drop in pressure of the Colts footballs is thus inconsistent with the valid scientific prediction of the ideal gas law that footballs will lose 1.13 psi in pressure, just due to the temperature drop. For some reason, the fact that the Colts footballs apparently did not obey the PHYSICAL LAWS OF OUR UNIVERSE has not, to this point, concerned anyone. It should. But...it is easy to explain!
The officials didn't even have time to test all of the Colts footballs because the 13-minute halftime break was ending. The Colts footballs had thus been in the heated room for at least 10 minutes before they were ever tested. They warmed up, maybe halfway to room temperature, which would explain a measurement of about half of the drop that was expected: 1.13/2 = 0.52 psi, saw 0.37-0.56 psi.
The Colts partly warmed-up footballs were then used as the "control" for the earlier-analyzed Patriots footballs. A ENORMOUS degree of importance, in the Wells report, was placed into the fact that the difference in the drop in pressure of the Colts footballs vs. the drop in pressure of the Patriots footballs was statistically significant. They even had a Princeton professor verify the statistical difference. The difference in the order in which the two groups of footballs were analyzed, as they were of course warming up toward room temperature, could fully account for this statistical significance, however!