Kelley is a rogue, just a high school coach, but his ideas are taken seriously. He’ll be featured next week, for the second time, on HBO Real Sports. Several NFL coaches have called to discuss his strategy, he said. At the MIT Sloan Sports Analytics Conference this spring, Atlanta Falcons executive Scott Pioli asked to meet Kelley and picked his brain for an hour.
Kelley’s philosophy hatched from his devotion to statistical analysis. The underlying math and probability made him believe that possession in football had been astonishingly undervalued, and that it was irrational to give the ball away when you had a chance to keep it. Only convention dictated normal punting and kickoff patterns. Imagine, he likes to say, if punting had never been part of football. What would fans think if a coach suddenly sent out a specialized player to kick away the ball after three plays?
At that Sloan conference this spring, another statistical insight led Kelley to another radical notion, which he plans to unleash this fall. It could be crazy. It could be genius. It’s probably a little bit of both. With football season around the corner, it’s definitely fun to contemplate.
“It might be horrible,” Kelley said. “But there’s a lot of thought that went into it.”
Kelley used an ESPN database to study college football history. He found that historically, there was no bigger indicator of victory than winning the turnover margin – teams that forced more turnovers than they committed won 80 percent of the time. But last season, Kelley said, a new trend emerged for the first time: Teams that recorded more plays of at least 20 yards won about 81 percent of the time.
It made sense to Kelley – bigger chunks of yardage meant scoring quicker and less opportunity to commit turnovers and drive-killing penalties. He became obsessed with finding a system designed for big plays. He found that on plays when two players touched the ball – a typical handoff or pass – teams gained 20 yards about 10 percent of the time. But when at least three players touched the ball – a trick play with a lateral involved – the percentage for gaining 20 yards rose to around 20 percent.
“That got me thinking,” Kelley said. “How could we develop a system for more than two people to touch the ball?”
One day, watching television, Kelley stumbled across a rugby game. That was it. Rugby teams built designed plays despite constant movement, an intricate series of laterals. Teammates didn’t block for the ball carrier; they rushed to the right spot to receive a pitch.
And so Kelley instituted a new system. When he calls out “Rugby!” before an offensive series, his wide receivers change their assignment. Rather than blocking downfield, they rush toward the receiver who catches the ball. If they’re open, they yell the receiver’s name and which side they’re on. He tells his players only to pitch the ball when they’re sure it’s safe.
Essentially, Kelley’s offense will run the option – after a completed pass down the field.
But one potential glitch stands out. Trick plays usually gain more yardage because of deception rather than design. Would those plays gain 20 yards less often because the defense could prepare? Kelley understands the potential problem, but believes the defense’s expectation for a lateral would actually help his offense.
One of the hardest tasks on a football field, Kelley reasons, is open-field tackling. It’s why defensive coordinators tell their players to swarm to the ball. By changing his extra receivers from blockers to potential pitch men, Kelley hopes he’ll force extra tacklers to stay with them and give the pass receiver more space to operate.
The field would be stretched in a different way. Think about a basketball offense with strong three-point shooters stationed in the corner. Even if they never get the ball, their presence means more space for other players.
“Let’s say we could successfully complete a pitch three times a game,” Kelley said. “The guy with the ball is going to be in more one-on-one situations down the field. Even if it’s not working as well, I do think opposing coaches are going to have to change the way they defend the field.”
The other problem is that laterals will ratchet up the risk of turnovers. Kelley is even less concerned with that. If his players practice the plays, there’s no reason downfield laterals shouldn’t be precise and relatively safe.
“The downfield pass sounded risky at one time, too,” Kelley said. “Oh my gosh, you throw it in the air, whoever catches it, it’s their ball!”
Even if Kelley’s offense works this fall, it’s not going to change much outside of the Arkansas 5A-Central Conference. Despite his success derived from not punting, no copycats have sprung up at higher levels. Football coaches are too wedded to convention, scared by the knowledge that losing traditionally is safer than trying to win radically. Kelley is just fine with that.
“I don’t want anybody else doing this,” Kelley said. “With not punting and the onside kicks, I know I have a stat advantage. If this works, I want everybody thinking this is stupid, too.”