The Washington Post just reviewed the widely reported study of a think tank (AEI), which dismantled the NFL's Wells report. AEI showed that the Pats balls were not abnormally deflated. That had already noted by many reputable scientists. What’s new about the Washington Post’s take, however, is that it comes out and says what many of us who've read the AEI report were too polite to say. The Post provides a scathing—almost directly accusatory-- rebuke of the Wells reports' analysts and of the NFL itself.
Why? When AEI reanalyzed the data, “The math in the Wells report didn’t add up….the results could not be replicated.... What’s worse, is the methods it used were not the ones it said it used. “The Wells report said it would use one equation, but then used a different (and weird) equation to arrive at its numbers (see my summary of the details below).*
It’s a standard principle in science: If you can’t replicate a set of results, then …a flaw or a fraud is at work. Either you made a mistake, or you made it up. Another plain English phrase possibly applies to all of this:
….”Lately the NFL has begun turning these special counsel investigations into manipulated campaigns calculated to enhance the commissioner’s profile and powers.
And they seem to be written to fit predetermined conclusions.”
(a not surprising fact given that the analysts it paid had previously written reports to help industries dispute links between cancer and 3 known causes of it: asbestos, toxic waste and cigarette smoke.)
According to the Washington Post, The AEI’s re-analysis of the Wells report supports the NFL Players Association’s charge that the Wells report “delivered exactly what the client wanted. It ….wasn’t an investigation; it was a frame job by the commissioner’s office desperate to reestablish its authority.”
“Twice now Goodell has ginned up false scandals that seriously and unfairly targeted individual players, and damaged franchises, on what turned out to be bogus or flawed evidence. Forget his bungled handling of Adrian Peterson and Ray Rice — at least those guys actually did something wrong. In the Deflategate and Bountygate affairs, Goodell hammered people who appear to have done nothing.” (even if they were apparently paid to do it in Bountygate)
Recall that the NFL also enabled Brady and the Pats to be convicted in the court of public opinion through daily leaks of false information about ball pressures, switched kicking balls, trips the john, and other incidents. Meanwhile, it withheld for over a month the true data, which could refute not only the data inaccuracies but also the faulty analyses)
The Post continues: “The AEI’s entry into Deflategate is important, because the institute was a major factor in righting the Goodell-driven injustice in Bountygate back in 2012. The Commissioner went all hanging judge on the New Orleans Saints, suspending several officials and players for a supposed bonuses system to injure opponents between 2009-2011. But then AEI analyzed injury data — something that surely the commissioner should have done. The AEI found that the Saints injured fewer opposing players than all but two teams in 2009 and all but one from 2009-11. After AEI’s report was presented at an NFL hearing, the suspensions were vacated……
Goodell is now in a truly interesting and awkward position. …Does Goodell stand by the conclusions of the Wells report, dig in and refuse to budge — thus establishing that he’s incapable of fairly considering evidence and is a serial abuser of his powers? Does he try to parse and sidestep the AEI analysis, by claiming that the scientific evidence is just a small part of the case against Brady? Trouble with that is, more than half of the Wells report’s 243 pages is taken up by pressure gauges and pounds-per-square-inch analysis – all of which must be thrown out according to AEI. If the balls weren’t deflated, then what’s left? One e-mail exchange, in which Brady complained that some game balls against the New York Jets were ludicrously overinflated. Is this evidence of ill intent? Hardly. Brady’s solution to the over-inflation was to suggest the refs check the rulebook. Not the act of a cheater.
Or does Goodell do the right thing and rescind Brady’s suspension on the basis of the new info in the AEI report — thus admitting that the league spent millions on a railroading farce? There is trouble for Goodell in this option too, because it suggests that the league office under Goodell’s leadership is either incapable of executing a proper investigation, or unwilling to….Brady may or may not win his appeal. But there is one sure loser here, trapped in a box of his own making: the commissioner.”
*you can read the AEI report link below. Among the Wells reports’ questionable practices:
1. it claimed to include a mean term in the statistical model (ANOVA) as well as a second error term and other interaction effects, But it actually did not use this model to obtain the reported results.
2. The authors also give the impression of running a regression using all the data but instead used a series of individual regressions.
3. They falsely claim that substituting different pressure gauges’ results in the analysis yielded the same results.
4. The halftime measures suggest the two referees switched gauges between testing the Pats and Colts balls, and this is not noted in the report. (the two gauges different by as much a 0.7 PSI, about half the degree of deflation claimed).
5. In their statistical analysis, the Wells study only compared the pressure changes in the Pats balls with the Colts balls, not with the expected pressure changes based on atmospherics. This not only invalidated the statistical assumptions (since similar changes from pregame to halftime measures will occur due to atmospheric conditions, leading to correlated error terms in an incompletely specified model). Beyond merely changing confidence in the statistical significance of results, however, it also made the results completely misleading. To show this, AEI does a separate and study of atmospherics (the relation of the ball’s pressure to temperature and other factors). When the latter are analyzed, the Patriots balls do not significantly deviate from the prediction of the Ideal Gas Law in the direction that one would expect based on the Wells report’s conclusions. By contrast, the Colts halftime pressures were higher than predicted, implying less deflation occurred than actually should have been the case. That was because the balls were given more time to warm up and were not measured until just before the halftime ended.
The AEI’s analysis of this point is incredibly detailed--even to the point of analyzing sequential pressure changes in the measures of the Pats, then the Colts’ balls. It thereby shows how the pressure changes could be explained by the order of measurement of the Pats and Colts balls. It is not consistent with the NFL’s allegations that the Pats deflated the balls. .
ADDENDUM: Summary of the appeal letter sent to the NFL (with expected arguments)
1. Brady was not proven guilty.
The accusation disregards contrary evidence. It’s based on speculation piled upon speculation about Brady’s involvement with two Pats’ employees’ purported conduct. It grasps at dubious, contradictory and mischaracterized circumstantial evidence merely to conclude that it is “more probable than not” that Mr. Brady was “generally aware of” “inappropriate activities.”
(Also, the balls were not abnormally deflated in the first place).
2. Brady's punishment is unfair and inconsistent.
The NFL stipulates a $25,000 fine for a team, not for a player. Yet, Brady suspension would cost him nearly $2,000,000 in unpaid salary for an alleged “general awareness” of actions
(not to mention endorsements lost already from the NFL’s defamatory and inaccurate media leaks, a million dollar Pats’ fine, the loss of 2 draft picks worth more millions, and the consequent threat to the future success of Brady’s teams).
Even if there were one iota of definitive proof of deflation and guilt, no player in the NFL’s history has ever gotten anything approaching this level of investigation or discipline for similar behavior.
(Brady’s initial suspension far exceeds that of Adrian Peterson for child abuse or Ray Rice for knocking his wife unconscious and dragging her out of an elevator by her hair. Also, 2 teams escaped any punishment at all after definite proof that they overinflated kicking footballs).
Never before in history had the NFL even tested football pressure at half time, let alone conducted a sting operation on other players for similar behavior.
(A former Bears QB admitted deflating balls. Colts’ sideline employees were never investigated despite the suspicious actions they accuse the Pats of. Reportedly, they carried under their sleeves the pins that could be used to deflate balls illegally. Other Pats critics like Jerry Rice were never investigated or penalized despite admitting that he applied stickum to his gloves to make it easier to achieve a completion. That’s a clearer advantage for Joe Montana than Brady).
3. Goodel should not arbitrate this case and the Exec VP Vincent should not have determined discipline in the first place.
An independent arbiter is needed due to NFL bias Only Goodel—not Vincent—is supposed to determine discipline. (Such delegation is a ruse to let him control the investigation and avoid embarassment). In fact, a previous independent arbitrator said Troy Vincent was unfamiliar with proper disciplinary procedure and should have no role in it. Also, Vincent cannot be unbiased as he was directly involved in game day events. As such, he must testify about his own involvement in such events. Goodel must too.
(The implicit accusation is that the NFL and Indy set up a “sting operation” to implicate Brady and the Pats. Who was the driving force behind the investigation? Mike Kensil, whose father was the Jets president and who himself worked 20 years for this team—one that has had longstanding legal disputes with the Pats. Kensil reportedly walked up to the Pats equipment manager at halftime and said, “We weighed the balls. You are in big f—ing trouble.” The NFLPA believes this statement not only showed prejudgment but also that Kensil took joy in trying to catch the Pats in the act. To make matters worse, Kensil destroyed the alleged “evidence.” Kensil inspected the footballs at halftime and instead of preserving them as evidence had them reinflated. As such, it was not possible to judge the pressure of all Pats and Colts balls together under the same atmospheric conditions. Remember that the AEI report found that such conditions fully explained the pressure differences).
Also, both Goodel and Vincent must both testify about when they became aware of the Colts’ complaints about ball deflation and what decisions and steps were thereafter taken. Specifically, the NFL had claimed it did not suspect deflation until a ball was intercepted in the game’s 2nd quarter. But there is now written evidence that Indy informed the NFL of their concerns a day earlier. If the Colts had notified the league that the Patriots were breaking the rules, the league is supposed to notify the Patriots about the complaint.
Also, since we now know that league officials were alerted before the game, they must explain why the exact PSI of each ball wasn’t recorded by NFL officials before the game.
(Apparently, the refs were not told of the concern of league officials prior to the game so that an improper sting operation could proceed. The NFL officials’ sting operation proceeded even though refs could have prevented this crucial game from being played with presumably underinflated balls).
The NFL is biased and lacks credibility in this case. Goodel, Vincent, and other NFL officials are themselves suspected of improper behavior. So, Goodel must explain why a neutral party with no ties to the League should not be appointed for Brady to maintain the integrity of the investigation. Goodel previously concluded that one was needed to hear Ray Rice’s case for that
4. In a footnote, the NFLPA letter also says that Brady did not knowingly violate rules or fail to cooperate with the investigation.
(But would any celebrity hand over their cell phones and emails to a biased organization that previously defamed his character through unauthorized and inaccurate news leaks?).
5. If the NFL does not appoint an independent party, the Brady and the NFLPA will sue the NFL.
Pittsburgh Steelers move Devin Gardner back to QB http://t.co/Vtu97FgLdj— Patrick Claybon (@PatrickClaybon) June 4, 2015
I don't know if this is a good or bad sign, but best of luck to him. I think he's got a shot at battling with Gradkowski for backup to Big Ben. I was picturing he could be pretty versatile at receiver for them and maybe make some magic like Randle El did ten years ago. Hopefully he can beat out Landry Jones and make the squad.
Tom Brady will still appeal suspension through NFLPA.
Could one of the resident MGoFootballExperts please explain why it is that tackles are so much more highly valued than guards? I understand why they'd be a bit more valued, but why is a great OT worth so much more than a great OG, and why are tackles so much harder to replace?
I ask because my NFL team, the Washington NFL franchise, drafted a guard this year and the DC press has surprisingly mixed feelings. This despite the fact that he is, by all accounts I've seen, a skilled beast and the best offensive lineman in the draft. And Washington needs help in the middle! Scherff could start right away! But people are saying that it's just not worth taking a guard with the fifth overall pick. Why not?
A man we once knew as one of the worst safeties in Michigan's history (although he finished his senior season as a pretty good SLB) continues a pretty solid NFL career as a safety, signing with the Texans.
As a Texans fan, I like it. But rumors are we're looking to trade DJ Swearinger, which seems like a waste on such a young player.
Probably overdue, but interesting nonetheless. Shouldn't impact much, as most taxable revenue runs through the member clubs. Mostly a PR move and leverage play against Congress re: concussion, domestic violence, etc.
This has been my first post. Neat.