That is not the case. The NFLPA had convincing testimony that the statistical methods used to deduce the amount of pressure that would be lost under game conditions were flawed.Originally Posted by sharpe1027
From Deadspin: http://deadspin.com/here-is-the-tran...ing-1722113452
Update (9:20 p.m.): After a lunch break, the NFLPA questions one of their expert witnesses Edward Snyder, Dean of the Yale School of Management. His role was to evaluate the findings of Exponent, the company used by Ted Wells’s team for scientific and statistical analysis of the deflation of the footballs. Snyder gets straight to the point, and identifies a number of errors he says Exponent made:
Q. Okay. So let’s go, let’s start with your slide deck. The first slide shows your three key findings. And if you could just sort of walk the Commissioner through each of the three key findings that you made and that we will elaborate on.
A. So first finding is that their analysis of the difference in differences, the analysis of the pressure drops and the difference in the average pressure drops is wrong because Exponent did not include timing and the effects of timing in that analysis.
Secondly, Exponent looked at the variation and the measurements between the Patriots’ balls and the Colts’ balls at halftime. They compared the variances. And despite conceding that there was no statistically significant difference between the two, they went ahead and drew conclusions, but those conclusions are improper.
And, last, and this goes to the issue of alternative assumptions, as well as error, if the logo gauge was used to measure the Patriots’ balls before the game, then given what the framework that Exponent provides us with scientifically, and if the analysis is done correctly, eight of the eleven Patriots’ balls are above the relevant scientific threshold.
OK. So 8 out of 11 meet the expected threshold of pressure loss due to conditions as calculated by Exponent. That still leaves 3 balls were tampering must have occurred, right? Not really, their analysis failed to be internally consistent.
Update (9:30 p.m.): Man, Snyder absolutely lays waste to the report Exponent prepared for the Wells Report. For instance, here he is explaining how quickly the PSI of a football changes when being brought into a warm room after spending a few hours out in the cold, and how Exponent didn’t even bother to account for timing in their report:
Q. So let’s go to our Slide 12. And what is this showing?
A. This takes the earlier Figure 22, and I will refer to that again. It takes the top schedule, what Exponent calls their transient analysis, that’s their scientific framework.
It says, okay, you bring in a Colts’ ball. It was pre-game at 13. It’s brought right into the locker room. It’s going to be 11.87. This is, like, so 2:40 is, like, in locker room terms, it’s minute zero. And then 12 minutes later, it’s warmed up and it’s roughly 1.1 psi greater in 12 minutes.
Q. The same ball?
A. The same ball.
Q. What did Exponent do in its difference in difference analysis to account for time?
A. Nothing.
Q. How do you know?
A. Absolutely nothing. If you look at their difference in difference equation in their appendix and you look at Table A3, where they report their results, they have explanatory variables for their difference in difference analysis and time is not an explanatory variable.
You can read the Exponent report forwards, backwards, upside down. You see time referred to again and again and again and again. However, you have to look at what they actually did, the statistical analysis that they actually did. They left time out of the analysis that they said was the most important.
Reply With Quote