The Carolina Panthers' Charles Godfrey (30) is congratulated by Haruki Nakamura after his interception return for a touchdown against the New Orleans Saints on Sunday. Many said the play changed the momentum of the game. They're wrong. Photo: Rainier Ehrhardt/AP
Charles Godfrey of the Charlotte Panthers anticipated the play perfectly.
As New Orleans Saints quarterback Drew Brees rolled to his right and threw the ball to tight end David Thomas, Godfrey sliced in front of Thomas, snatched the ball out of the air and strolled nine yards for a touchdown. The Panthers bench erupted. The crowd went nuts. And sportswriters breathlessly proclaimed the first-quarter play in Sunday's game a turning point where the momentum shifted decisively in favor of the Panthers, who won 35-27.
Except it didn't.
The widely held and devoutly believed idea that a big play can change the momentum of a game is, in a word, bunk. So say a trio of MIT stats geeks with a decade of data to back them up.
Aaron Johnson, Alex Stimpson, and Torin Clark analyzed 473,621 plays run during the 2,931 NFL games played between 2000 and 2010 simply to determine whether big plays change the momentum of a game, motivating teams to perform better in subsequent drives. The results, presented in Turning the Tide: Big Plays and Psychological Momentum in the NFL (.pdf), found no evidence that a big defensive play has any effect on offensive performance later in the game.
That flies in the face of all that sports writers and armchair quarterbacks will tell you. Like many football fans, Johnson and his friends believed the conventional wisdom, but being engineers, they're skeptical of everything until they've seen the data. So they decided to test it.
They gleaned stats from ArmchairAnalysis.com to examine 69,330 drives, which they divided into two categories: those that started after a big defensive play — an interception, fumble recovery, fourth-down stop, safety, or blocked kick — and those that did not. Then they analyzed three variables: the result of the first play after the change of possession, the success of the first set of downs, and the points scored on that drive. They discovered that a big defensive play doesn't improve the performance of the offense on the subsequent drive.
In other words, the momentum doesn't shift.
The key reason lies in something called the momentum chain, defined by scholars Jim Taylor and Andrew Demick in 1994. For a precipitating event to have an effect on performance, they found, a number of factors and internal attitudes must align perfectly. In Sunday's game, for example, the Panthers' offense would have had to gain positive momentum after Godfrey's interception and the Saints defense would have to see negative momentum. That's not what happened: Godfrey's pick-six tied the game at 7-7, but New Orleans bounced back with a pair of field goals to take a 13-7 lead before Carolina came back to take the lead for good.
So if the momentum doesn't actually change, why do we think it does?
"People do this with a lot of psychological phenomena that don't actually exist," Johnson said. "Fans watch the games, and we associate the positive outcomes with a particular perceived phenomena and we fail to associate the negative ones with our phenomena. So if a team scores a touchdown right after they intercept the ball, we say that the touchdown happened because of the interception. We take that causal leap. And these are the situations we all remember instead of all the 3-and-outs that happen after interceptions."
Dr. Robert Corb, a sports psychologist on the UCLA athletics medical staff, said it's the sports version of the placebo effect: If you believe something will work, it often does.
"We know from social learning theory that people learn from watching others," Corb said. "If the offense sitting on the sidelines watches their defense get pumped up after making a big play, those offensive players are likely to get pumped up as well, and perhaps go out and make a big play themselves. If that happens a few times, and coaches and announcers start talking about a change in momentum, perception becomes reality."
After their paper was published for the MIT Sloan Sports Analytics Conference earlier this year, the students expanded their research to examine whether coaches are significantly riskier after big defensive plays or turnovers. Do they "go for the kill" after the defense turns in a big play? Again, the answer was no. There was no evidence to suggest their plays grew more aggressive to take advantage of the big play.
For their next move, Johnson, Stimpson, and Clark hope to expand their research into college football to see whether younger players are more influenced by psychological factors. For now, though, they have plenty of opportunities each week to add to their NFL data.
"Now we pay more attention to when the commentators talk about momentum, which seems to be all the time," Stimpson said. "We also tend to point it out to our friends, which has probably become annoying by this point."
Sarah Mitroff 20 Sep, 2012
-
Source: http://feeds.wired.com/~r/wired/index/~3/3V2p67dQlr8/
--
Manage subscription | Powered by rssforward.com
Anda sedang membaca artikel tentang
Post-Interception 'Momentum' Is a Myth, Researchers Say
Dengan url
http://sizeofhappiness.blogspot.com/2012/09/post-interception-momentum-is-myth.html
Anda boleh menyebar luaskannya atau mengcopy paste-nya
Post-Interception 'Momentum' Is a Myth, Researchers Say
namun jangan lupa untuk meletakkan link
Post-Interception 'Momentum' Is a Myth, Researchers Say
sebagai sumbernya
0 komentar:
Posting Komentar