WASHINGTON (AP) - It's advice that sounds like heresy on the gridiron: Go for it on fourth down. Try more onside kicks. Running backs don't matter much.
But to stats geeks, this is the gospel from the spreadsheets.
The number-savvy economists, statisticians and refugees from Wall Street who revolutionized much of baseball in a way celebrated in the book and movie "Moneyball" are aiming their calculators at football.
But so far they aren't making much headway. The complex team-oriented game is difficult to break down into interchangeable numbers, and the men who run the sport are even tougher to change. If baseball is in the 10th year of adopting "Moneyball" concepts, football is barely in year one, several sports number gurus say.
There is one NFL team that seems to best epitomize an increased amount of number-based decision-making, top sports statisticians agree. And that team, the New England Patriots, will be playing on Super Bowl Sunday for the fifth time in 11 years.
"The Patriots are the `Moneyball' team of the NFL, no doubt about it," said Brian Burke, founder of Advanced NFL Stats Inc. "The Patriots are in the Super Bowl because they have this system and stick to it. The Giants are in the Super Bowl because they are lucky."
The statisticians are trying to bring mathematics and economics to decisions made on field strategy and player drafting. It's called analytics and in many, but not all, ways it runs counter to decades of football philosophy.
The numbers show that:
- Icing the kicker - calling a last-second timeout to put more pressure on him - doesn't work.
- The cliche that defense wins championships is wrong.
- Top first-round draft picks are overvalued compared to picks a bit lower. So says Tobias Moskowitz, a University of Chicago economics professor and co-author of the book "Scorecasting."
Stats guys can spout the numbers proving them right. But when football teams hit the field and general managers compile their rosters, all that doesn't seem to matter. They punt on fourth-and-2 even though they have a 75 percent chance of making a first down. They ice the kicker. And they talk about how it all starts and ends on the defensive side of the line, even though when Moskowitz ran the numbers, the best offense won as many championships as the best defense.
The presence of the Patriots in Sunday's game, and the win by the New Orleans Saints two years ago, show that it's nonsense to say defense wins championships, said Aaron Schatz, founder of Football Outsiders and a pioneer in the young field. "This year is the most offensively focused of all time," he said. "These are the two worst defenses to ever play in the Super Bowl in terms of yards allowed."
Sunday's game also supports another mantra of the numbers geeks: "a running back is a running back is a running back." Shane Reese, a statistics professor at Brigham Young University, said an average-to-good running back with an excellent line works just as well as a superstar running back.
"The running back is an overrated position," he says.
"Yes, it's a tool, but it's one tool," Mula said. "On the carpenter's belt. It's the one that he should regularly be checking his work against."
Mula said there's far more to decision-making than numbers in the Patriots' way of thinking. While he figures nearly every club uses analytics to some degree, it's the way they use it "that probably separates New England" from most other teams. And if a club isn't crunching the numbers, "to me that explains why they'll forever be in the basement," Mula said.
Boston is a mecca for sports number junkies. The Red Sox, Celtics and Patriots are big into using numbers and every year MIT hosts a conference on it.
And while the Pats win respect for following the numbers guys' advice in trading players for the future, a failed fourth-and-2 attempt at a first down in their own territory a couple of years ago still gets talked about.
Yet that strategy has been studied so frequently that Schatz calls the topic too boring and accepted to talk about.
Moskowitz looked at a decade of about 7,000 fourth-down situations in the NFL. He calculated how often fourth-down attempts are successfully converted. He calculated the results of improved field positions from punts. He looked at the outcome of the decisions. He charted out every position on the field and different yardage situations in most of them - even fourth-and-5 yards to go - and it made more statistical sense to go for a first down, he said. And the other stats guys agree wholeheartedly.
Even more than the Patriots, a high school football team in Little Rock, Ark., has embraced football analytics with a passion. Coach Kevin Kelley of Pulaski Academy has gone for it on fourth down more than 500 times since he took over in 2003; getting first downs 49 percent of the time. In the past five years he's only punted five times; but he's also won the state championship twice in those five years and three times in his nine-year tenure.
He figures punting gains a 30-yard advantage in field position, but a 1-in-2 chance of keeping possession of the ball is far more worth it. He also only does onside kicks, but he has a dozen intricately choreographed versions, and gets the ball back about one-quarter of the time. In one game against a powerhouse rival last year, his team scored four touchdowns and led 29-0 after consecutive recovered onside kicks before the other team got the chance to run an offensive play.
When you ask why Kelley doesn't play by "the book," he responds: "Which book are you talking about? The one everyone uses or the one that's right?"
Kelley said many people don't adopt football analytics "because self-preservation kicks in. It's `I'm worried about my job and I don't want to fight the battle."'
And Kelley wouldn't mind if other football coaches don't follow his lead: "I have an advantage right now over everybody else."
Online:
Football Outsiders http://footballoutsiders.com/
Advanced NFL Stats: http://www.advancednflstats.com/
Scorecasting: http://scorecasting.com/
Pulaski Academy football: http://www.pulaskiacademy.org/football