Follow Datanami:
February 1, 2019

Three Ways Analytics Impacts the NFL

(Africa Studio/Shutterstock)

Fans will hopefully enjoy a ton of action on the field Sunday when the Los Angeles Rams line up against the New England Patriots in Super Bowl LIII. But what fans likely won’t see is how analytics is impacting football behind the scenes.

Up to this point, the adoption of analytics in the National Football League has been tepid at best. Compared to Major League Baseball and the National Basketball Association, where openly available player data and abundant analytics has energized fans and analysts alike, the NFL has lagged behind.

But by all accounts, that’s drought is on the wane as teams look for novel ways to apply big data analytics to improve their odds of winning. The NFL is also getting into the act, but can it do more?

Here are three ways analytics are being used in the NFL today.

Player evaluation

Football is a team sport that’s unlike any other. Winning at football requires having the right roster of 53 men who not only are appropriately sized and physically skilled, but whose personalities fit well into the coach’s scheme and are also team players.

Teams spend millions to gather all available personnel data and develop insightful information — not only about their own players, but other teams’ players, as well as the most promising college players for the draft.

During the draft in April, there isn’t much difference in the intelligence collected about the top prospects, who are likely to go in the first few rounds. But the better teams are increasingly turning to analytics to make more informed picks in later rounds, as well as during free agency.

“We really let our analytics group lead us in sixth and seventh round – and especially undrafted free agency, they run it all,” Los Angeles Rams Chief Operating Officer and Executive Vice President of Football Operations Kevin Demoff said during a panel discussion last February at the 2018 MIT Sloan Sports Analytics Conference.

It’s all about spotting hidden traits that could indicate a good player down the line. “Your scouts may have seen these guys one time, but increasingly teams are looking to analytics to fill in the gaps,” Demoff said. “It still always comes down to human beings and developing them, but [analytics] gives you an even better chance.”

Player Tracking

Starting in 2016, the NFL started placing sensors on players for the purpose of tracking player movement during games. The League kept the data to itself for the first two seasons, but last year it released the data to all 32 teams. The result has been somewhat of a data analytics arms race among the teams to see which ones can best utilize the data.

(Mike Orlov/Shutterstock)

“It’s really helpful in self-scouting,” Dave Anderson, a former tight end with the Texans, Broncos, and Redskins, and a co-founder of Gains Group, said during the panel discussion at the 2018 MIT Sloan Sports Analytics Conference. “It used to take weeks and weeks to figure out your own tendencies,” Anderson said. “Now you can pull up tendencies in a matter of seconds. I think that’s probably the most helpful thing for a team.”

The player movement data will provide objective analysis of players for coaches and scouting teams, said the Rams’ Demoff. “One of the hardest things when you’re talking about football analytics is it’s not completely objective. It’s subjective,” he said.

“I think what the [player tracking] data will do is strip it all out,” he continues. “You get rid of the inherent bias that we all bring in in scouting and give a chance for players who are over performing their name or reputation. And conversely, maybe some player who has a great reputation, you look at them as a dot on the screen, and say maybe they’re not as good as we thought they were.”

Game Planning

The player movement data will also help with game planning, said Tedy Bruschi, the hard-hitting former linebacker with the New England Patriots.

(inspired-fiona/Shutterstock)

“[L]et’s say I play defensive end and I’m about to go up against this offensive lineman… it’s week 15 and I have 14 week’s of data,” Buschi said during the Sloan Sports Analytics Conference panel discussion says. “Give me this data where every time he was single-blocked that he got beat inside. I want to see every play.

“Boom!” he continued. “Now I can study these 29 plays, and out of those 29 plays, he got beat by the same move 16 times. I’m going to use that move.”

The use of statistics in football is not new. Teams have collected the data for years and built databases of every play they’ve ever run. They’ve built playbooks from these databases that tell them which offensive plays will be effective in all sorts of different situations, and defenses have their own playbooks that tell them which packages are most effective against the most likely play the offense is going to run.

There are many variables in the game of football that cannot be adequately modeled. From the chance of injuries and the willingness to play hurt to the adherence to team-play and even the bounce of the football, the randomness of football is an integral part of the indelible fun.

But there’s one thing that seems almost absolutely certain: If the Rams rely purely on a statistical model to tell them how to play offense and defense this Sunday, they will probably lose. That’s not because the Patriots have their own superior analytics group. But more likely because their head coach, Bill Belichick (who is also the GM), seems to have already memorized every possible play.

“[Belichick] does it with intuition,” one AFC executive told author Kevin Clark in a story on The Ringer titled “The NFL’s Analytics Revolution Has Arrived.”

“You know because you’ve been coaching for so long, how you match these 11 guys against those 11 guys,” he said. “It all makes sense to you. At some point, maybe we can all come to those conclusions without having Bill Belichick’s brain. We’re still a long way from that.”

Related Items:

Are You Ready for Some Football Analytics?

Deep Learning Is About to Revolutionize Sports Analytics. Here’s How

How Madden’s Ratings Czar Rules NFL Analytics

 

Datanami