The data arrived weekly last season at each NFL team's headquarters. Packed inside was every imaginable measure of a player's in-game movements: Speed on each play. Yards covered, both horizontally and vertically. Precise location of a receiver or a defensive back on the field.
Generated by RFID (Radio-Frequency IDentification) chips embedded in players' shoulder pads, the information was powered by the league's Next-Gen stats program that sports analytics experts consider nothing short of game-altering in terms of how we understand football.
And what did the teams do with the numbers last season -- the first year these advanced figures were made available to them?
"Nothing," Atlanta Falcons coach Dan Quinn said, with a laugh and a shrug.
"I'll be honest," said Tampa Bay Buccaneers coach Dirk Koetter. "I didn't look at that data during the season."
"All that stuff is good to have," Indianapolis Colts coach Chuck Pagano said. "But it's on film, too, and the film don't lie."
The NFL's relationship with analytics remains a matter of competitive semantics. Every team uses data on some level, and while a few individuals admit it, no one embraces it. Game-day player tracking has long been expected to revolutionize evaluation (as well as planning and strategy for games), but most coaches ignored the debut dump of 2016. It will be available again this season, perhaps soon supplemented by data gathered from chips embedded in the football itself. No one, however, has convinced coaches that poring over these statistics can help them win games.
"With that stuff, it's kind of spotty," Seattle Seahawks coach Pete Carroll said. "We don't have enough background yet to kind of make sense of it, how it's helping us at all."
Interviews with nearly half the NFL's head coaches during the past year yielded similar answers. Carroll was one of the few who could recall a specific piece of information he gleaned from the data: Receiver Tyler Lockett outran everyone else on the field at a speed that was not close to matching his previously recorded high. For the most part, though, coaches believed crunching the numbers was redundant to their weekly film study.
"I can see if he was fast enough," New Orleans Saints coach Sean Payton said. "I don't need to see [in numbers] how fast he was."
It is early, however, to draw final conclusions on how chip-generated player data can affect the game. Some players have ideas, as it turns out. So does Dean Oliver, a pioneer of sports analytics and the current vice president of the data/analytics giant TruMedia. And sometimes it takes a few years for football people to embrace a new tool, as you can see below.
Thumbs up: Practice tracking
Despite some early reservations, nearly every NFL team now uses some level of player tracking to monitor exertion in practice and help prevent injuries. Coaches who once planned practices and individual workloads by feel now have a more precise set of information that helps them determine how long and how hard to go.
GPS chips inserted in shoulder pads record how many miles a player has run, how close he has come to hitting his maximum speed and other data points. Combined, they tell a coach how hard the player has worked, and indicate whether he's approaching a point where he is more susceptible to injury. Coaches use the numbers to guide practice length and intensity, as well as decide when to give individual players time off. All in all, the process that takes some of the guesswork out of health management.
"The awareness of workloads and the wear and tear on players has been helpful," Carroll said. "It's allowed us to communicate with players on a better playing field, and they know we can tell better how they're feeling."
The trend surfaced in the NFL less than five years ago, and is now embraced without much debate across the league. Payton considers it the true and best football-related advantage of chip technology.
"Our benefit is player-injury related," he said. "The league's benefit is production on game day. So, everyone has a different goal with it in mind. With us, it's training and reducing injury."
Game day production
Payton was referring to the NFL's decision to supply Next-Gen game day data to television broadcasts. In theory, viewer experience could be enhanced by data-culled explanations for what takes place on the field. How did DeSean Jackson catch up to that Jameis Winston pass? (By increasing his speed from 16 miles per hour to 18, for example.) How many yards toward the sideline did LeSean McCoy run before turning up field? (Perhaps 12 yards sideways to gain nine yards downfield.)
Payton and other coaches suspect the NFL means to implement its game-day data for commercial and entertainment use, rather than as a strategic supplement.
Oliver, who has worked in the NBA as well as at ESPN, said: "Player tracking is a complete game-changer in how football can be understood." On a broadcast, Oliver said, the information can be used to support what otherwise seem to be subjective judgments.
"You watch a game and they're telling you a player did well or did poorly," Oliver said. "Well, why? Take an offensive lineman. Is he holding his block longer? Is he getting downfield for his second block more frequently? You hear broadcasts and read articles about a player being good, but they don't explain what makes them good. The information in this data can do that. It can get beyond where we are in terms of knowing the game."
The question, of course, is whether the information can get beyond what coaches know about the game and how they plan for success. Companies such as TruMedia do not have access to the data but are planning for that day. To that end, Oliver said, much of the value could actually be found by personnel departments who need to shortcut or modernize their analysis of a large group of players. Where coaches are concerned, Oliver and others are intrigued by the technology's ability to pinpoint the position of the receiver, defensive back and ball at any given time. How close did the defensive back stay to the man he was covering? How much ground did he make up, or lose, when the ball was in the air? How much separation did a receiver gain on an average nine route? How long did it take him to get off the line of scrimmage?
"We can measure all of that with this data," Oliver said. "Stats see everything on every play. They might not see it quite as well as a trained NFL coach. But if you could use it to come up with a stat that a coach trusts, it could be of help to them. Maybe it's different for different coaches. I'm not going to say that there is one specific stat that can come out of this that will apply to everybody, but it could help some of them focus what they see before they start watching film."
At the moment, however, few feel compelled to try.
"If you're watching tape," Koetter said, "[you can see] he's about 6 yards off, he's about 5 yards off. He's pressing. He's in cloud corner. You can see it."
Said Houston Texans coach Bill O'Brien: "That's something you see on tape when you watch it. But I guess the data could quantify it and be a nice confirmation of what you're already seeing."
To be fair, the NFL severely limits distribution. Each team gets only its own data, leaving coaches unable to compare the numbers to a larger set and eliminating the possibility of using the data to scout opponents. It also arrives after most teams have completed the process of reviewing that game. Some coaches would love to receive it during a game, which would provide a boost in making real-time matchup adjustments, but the NFL's competition committee is concerned this information would spark a competitive imbalance.
Players are a bit more amenable to the idea of technological aid, especially those who were exposed to some of the tools at the high school and collegiate level. This spring, the NFL Players Association contracted with the wearable technology company WHOOP to outfit each player with a device that would generate its own workload numbers while providing an avenue for future commercialization.
In speaking to them over the course of the past year, at least one potential tool drew interest. While fans and media focus on the velocity of the football, quarterbacks and their coaches are more interested in trajectory. Redskins quarterback Kirk Cousins, for example, practices three different types of throws and would be interested to see them plotted for analysis and improvement.
A "Level One" throw, Cousins explained, is a line drive. "Level Two," he said, "is when you get the ball up and over somebody and then back down quickly on an intermediate throw." Finally, a "Level Three" involves the quarterback throwing "a deep ball that hangs up in the air and has a huge arc to it."
Every route in the Redskins' playbook requires one of those three trajectories, depending on the overall play call and the defense employed. Using data chips in the ball on game day could help quarterbacks not only evaluate their throws, but also as a more efficient way to project their impact on specific defensive backs.
"That would be pretty cool," Redskins coach Jay Gruden said. "It would be kind of like what you see with a golf ball and the trajectory it takes down the fairway. That could be something especially good for deep balls because you see so many quarterbacks throw it too flat with not enough air."
Redskins backup Colt McCoy, meanwhile, thinks that trajectory reports could help a quarterback adjust to his receivers.
"Some receivers, you might want to put more air under it on a deep ball than others," McCoy said. "Some guys are better at tracking a 50-50 ball in the air. That would be a good thing to know."
These are the types of conversations that lead to innovation, which is a fancy way of talking about competitive advantage. It's unlikely that anything gleaned from game-day tracking will revolutionize how the game is played. As Oliver noted, there is more to be gained here for fans and other observers. But creative minds are digging, and the story is only just beginning.