BOSTON -- Here's a fun game to play in your free time. Imagine you're an NFL general manager scouting cornerbacks in advance of free agency.
Your coach plays press coverage, putting defensive backs tight against receivers on the line of scrimmage. So his scheme sometimes requires cornerbacks to chase receivers at a sprint to recover from missed or insufficient contact.
What available cornerbacks would be the best fit? The only tools you have are 40 times from a long-ago scouting combine and your subjective evaluation of his speed on tape. Soon, however, the NFL's more shrewd general managers will have a new and potentially landscape-changing trove of data to help.
Sometime in May, when the draft is complete and free agency is largely concluded, the NFL is scheduled to provide its teams access to the next-gen statistics it has been compiling during games for the past two seasons. What teams do with it is up to them, but discussion at last week's MIT Sloan Sports Analytics Conference teemed with possibilities as analytics moves closer to the football mainstream.
"I think we can answer a lot of questions," said Adam Beard, the Cleveland Browns' newly hired director of high performance. "Without it, we've got a lot of subjective opinions. Everybody's an expert on Monday. They all knew what was going to happen. But the data can help us be more objective, look at trends and what can help us win the next one."
Since 2014, the NFL has worked with Zebra Technologies to outfit its stadiums with RFID (radio frequency identification signals) technology that tracks and records the real-time position and movement of all players using a chip embedded under their shoulder pads. The league shared a small portion of the resulting data with television broadcasters -- you might have seen graphics that revealed how many miles per hour a receiver ran -- but it's valued most by the teams themselves.
The league has withheld most of the information to better understand the potential impact and competitive disadvantages it might spring. The NFL will provide its own digital platform to synthesize the numbers to any team that wants it, but there were plenty of third-party vendors hawking custom platforms to team representatives last week at Sloan.
So let's return to our original example of a general manager scouting veteran cornerbacks. Instead of relying only on his judgment off tape, the general manager could use RFID data to know the actual speed a cornerback ran to recover when initially beat at the line of scrimmage on every play it happened the previous season. The general manager could compare that speed to the rest of the available cornerbacks, to all cornerbacks throughout the league and even to the top receivers he'll face within the division.
There are plenty of applications to "game speed" alone. Exactly how much speed has your veteran running back lost? And how do you reconcile game film of a player who tested poorly at the combine three years ago but looks plenty fast on tape?
"There is a bit of a reality check here," said Dean Oliver, vice president of data science at TruMedia and a former ESPN analyst. "We talk about combine speed, and then there is game speed. There is potential here for capturing those things. It doesn't eliminate the discussion about what a player's game speed is, but you can see how fast a player is on certain plays in the game. If you're seeing [in the data] that a player is much slower or much faster, that can give you something. You can say, 'OK, he may look faster but he really is in fact faster.' That reality check can be helpful of keeping arguments from going too long in the wrong direction."
Player tracking is common among professional sports in Europe and Australia, among other places, and most NFL teams are using some form of it to monitor player exertion in practice. Cleveland Browns receiver Andrew Hawkins returned to the Sloan conference last week to repeat a message he delivered last year: Players like the idea of increased health awareness, but wonder whether the data could be used against them in contract negotiations.
"Instead of them telling you that you've slowed down," Hawkins said, "they can hand you a piece of paper showing you that you're slower. That's tough."
But the truth is that most teams aren't sure how the data will be best used, part of the reason the NFL delayed its release and finally scheduled it for after the "player acquisition" portion of this offseason. Will it only be a tool for evaluating players? Or can coaches incorporate it into game plans and in-game adjustments? Might they rest players based on their exertion data instead of simply a raised hand or body language?
At last year's Sloan conference, New Orleans Saints coach Sean Payton suggested in-game RFID data could be helpful in determining matchups between not only receivers and cornerbacks, but other positions such as pass-rushers and offensive linemen. If available during the game, it could allow coaches to change matchups based on players who might be slowing down and thus getting tired.
Perkins Miller, the NFL's chief digital officer and head of media operations, has been working to develop data-related applications for television broadcasters. He recently conducted a "Hackathon" in which college students were challenged to find football stories in data.
"The winner," Miller said, "was a group that found a way to tell you the probability of a successful pass play at the snap of the ball based on the configuration of eligible receivers and the defensive line. It can be a fascinating piece of evolution and maybe even understanding the decision-making a quarterback goes through."
If television viewers can be given a real-time heads-up of what is in essence a "good look" or a "bad look" for a quarterback, could coaches capitalize as well? What data scientists call "pattern recognition" is sometimes not diagnosed by coaches or scouts until the next-day review of a game. These are the questions that must be answered. They're good questions, and fun, and if used well, should do nothing but enhance the game.