Tweet
BEFORE “Deflategate,” the National Football League’s most recent controversy, there was “Bountygate,” in which New Orleans Saints officials were accused of offering bonuses to Saints players who injured members of opposing teams — a policy alleged to have been in place from 2009 to 2011. Initially, several Saints officials and players were penalized by the N.F.L.
In 2012, however, we published an analysis of N.F.L. injury data that found that the Saints injured fewer opposing players than all but two teams did in 2009, and fewer than all but one team did from 2009 to 2011. Even if Saints officials offered “bounties,” there was no good evidence that Saints players were influenced by them.
We presented our findings at an N.F.L. hearing in November 2012. The next month, the N.F.L. vacated all the players’ suspensions.
Considering that our impartiality was at least implicitly recognized by the N.F.L. in the past, we believe that our analysis of the evidence in Deflategate, in a study released Friday by the American Enterprise Institute, could help resolve this latest controversy.
Deflategate is a dispute about whether the New England Patriots used deliberately underinflated footballs in their playoff victory over the Indianapolis Colts in January. (Each N.F.L. team provides its own footballs when on offense, and an underinflated football may be easier to handle in cold or wet conditions.)
The N.F.L. commissioned a study, known as the Wells report, that concluded that it was “more probable than not” that Patriots personnel deliberately violated the rules and that Tom Brady, the Patriots quarterback, was aware of it. Following the release of the Wells report last month, the N.F.L. penalized the Patriots organization and suspended Mr. Brady for four games.
Our study, written with our colleague Joseph Sullivan, examines the evidence and methodology of the Wells report and concludes that it is deeply flawed. (We have no financial stake in the outcome of Deflategate.)
The Wells report’s main finding is that the Patriots balls declined in pressure more than the Colts balls did in the first half of their game, and that the decline is highly statistically significant. For the sake of argument, let’s grant this finding for now. Even still, it alone does not prove misconduct. There are, after all, two possibilities. The first is that the Patriots balls declined too much. The second — overlooked by the Wells report — is that the Colts balls declined too little.
The latter possibility appears to be more likely. The Wells report notes the expected pressure for the footballs at halftime in the Patriots-Colts game, factoring in the decline in pressure to be expected when a ball, inflated in a warm room, has been moved to a cold outdoor field. If the Patriots deflated their balls, their pressure levels at halftime should have fallen below the expected level, while the Colts balls at halftime should have hovered around that level.
But when we analyzed the data provided in the Wells report, we found that the Patriots balls declined by about the expected amount, while the Colts balls declined by less. In fact, the pressure of the Colts balls was statistically significantly higher than expected. Contrary to the report, the significant difference between the changes in pressure of the two teams’ balls was not because the pressure of the Patriots balls was too low, but because that of the Colts balls was too high.
How could this be? The report’s own findings suggest an explanation: At halftime, N.F.L. officials measured the pressure of “only a sample” of the Colts balls (four out of 12) before they ran out of time; the second half of the game was about to begin. This implies that the Colts balls sat in the warm room where they were to be measured — and thus increased in pressure — for almost the entirety of halftime before being measured.
All of the 11 available Patriots balls, by contrast, were measured at halftime, which suggests that they were measured earlier, when they were colder — and thus lower in pressure. Although this explanation contradicts the Wells report’s conclusions, it fits all the evidence.
There are other factors discussed in our study that undermine support for the Wells report’s conclusions. For example, there is considerable uncertainty concerning the actual pressure of the footballs. The N.F.L. official who checked the pressure before the game used some combination of two pressure gauges to measure the Patriots and Colts balls, but it is not known which particular combination.
One of the gauges, as the report notes, records pressures that are higher than the other. If the official used that gauge to measure the Patriots balls (but not the Colts balls) pregame, then those balls may well have started out with too little air, which could explain a later appearance of intentional deflation. The report, however, does not consider that possibility.
Our recommendation? When the N.F.L. hears Mr. Brady’s appeal of his suspension later this month, it should proceed with the knowledge that the Wells report is unreliable.
Kevin A. Hassett is the director of economic studies, and Stan A. Veuger is a resident scholar, at the American Enterprise Institute.
http://www.nytimes.com/2015/06/14/op...gate.html?_r=2
In 2012, however, we published an analysis of N.F.L. injury data that found that the Saints injured fewer opposing players than all but two teams did in 2009, and fewer than all but one team did from 2009 to 2011. Even if Saints officials offered “bounties,” there was no good evidence that Saints players were influenced by them.
We presented our findings at an N.F.L. hearing in November 2012. The next month, the N.F.L. vacated all the players’ suspensions.
Considering that our impartiality was at least implicitly recognized by the N.F.L. in the past, we believe that our analysis of the evidence in Deflategate, in a study released Friday by the American Enterprise Institute, could help resolve this latest controversy.
Deflategate is a dispute about whether the New England Patriots used deliberately underinflated footballs in their playoff victory over the Indianapolis Colts in January. (Each N.F.L. team provides its own footballs when on offense, and an underinflated football may be easier to handle in cold or wet conditions.)
The N.F.L. commissioned a study, known as the Wells report, that concluded that it was “more probable than not” that Patriots personnel deliberately violated the rules and that Tom Brady, the Patriots quarterback, was aware of it. Following the release of the Wells report last month, the N.F.L. penalized the Patriots organization and suspended Mr. Brady for four games.
Our study, written with our colleague Joseph Sullivan, examines the evidence and methodology of the Wells report and concludes that it is deeply flawed. (We have no financial stake in the outcome of Deflategate.)
The Wells report’s main finding is that the Patriots balls declined in pressure more than the Colts balls did in the first half of their game, and that the decline is highly statistically significant. For the sake of argument, let’s grant this finding for now. Even still, it alone does not prove misconduct. There are, after all, two possibilities. The first is that the Patriots balls declined too much. The second — overlooked by the Wells report — is that the Colts balls declined too little.
The latter possibility appears to be more likely. The Wells report notes the expected pressure for the footballs at halftime in the Patriots-Colts game, factoring in the decline in pressure to be expected when a ball, inflated in a warm room, has been moved to a cold outdoor field. If the Patriots deflated their balls, their pressure levels at halftime should have fallen below the expected level, while the Colts balls at halftime should have hovered around that level.
But when we analyzed the data provided in the Wells report, we found that the Patriots balls declined by about the expected amount, while the Colts balls declined by less. In fact, the pressure of the Colts balls was statistically significantly higher than expected. Contrary to the report, the significant difference between the changes in pressure of the two teams’ balls was not because the pressure of the Patriots balls was too low, but because that of the Colts balls was too high.
How could this be? The report’s own findings suggest an explanation: At halftime, N.F.L. officials measured the pressure of “only a sample” of the Colts balls (four out of 12) before they ran out of time; the second half of the game was about to begin. This implies that the Colts balls sat in the warm room where they were to be measured — and thus increased in pressure — for almost the entirety of halftime before being measured.
All of the 11 available Patriots balls, by contrast, were measured at halftime, which suggests that they were measured earlier, when they were colder — and thus lower in pressure. Although this explanation contradicts the Wells report’s conclusions, it fits all the evidence.
There are other factors discussed in our study that undermine support for the Wells report’s conclusions. For example, there is considerable uncertainty concerning the actual pressure of the footballs. The N.F.L. official who checked the pressure before the game used some combination of two pressure gauges to measure the Patriots and Colts balls, but it is not known which particular combination.
One of the gauges, as the report notes, records pressures that are higher than the other. If the official used that gauge to measure the Patriots balls (but not the Colts balls) pregame, then those balls may well have started out with too little air, which could explain a later appearance of intentional deflation. The report, however, does not consider that possibility.
Our recommendation? When the N.F.L. hears Mr. Brady’s appeal of his suspension later this month, it should proceed with the knowledge that the Wells report is unreliable.
Kevin A. Hassett is the director of economic studies, and Stan A. Veuger is a resident scholar, at the American Enterprise Institute.
http://www.nytimes.com/2015/06/14/op...gate.html?_r=2