“The NFL is a tremendous content creator,” imparts Michelle McKenna, the National Football League’s Senior Vice President and CIO.
“With over 188 million fans worldwide, NFL games accounted for 38 of last year’s top 50 telecasts. Every week the league generates 3 terabytes, equivalent to 1,500 hours, of data,“ the CIO adds. That includes a generous payload of stats from the Seattle Seahawks and the 12’s.
Machine learning has transformed the league. Coaches can use the technology to officiate playbook formations and automatically draw out plays, saving them time on the sidelines. On and off the field, the technology is being applied to better understand fan engagement, how a game is presented, the potential impact of adjusting the rules of play, how the game is called, and player performance and safety.
Next Gen Stats allows the NFL to use real-time data to engage, inform, and empower fans in new and unique ways.
Want to quarterback the game plan for yourself?
Download the Ebook:
“ Going Long on Machine Learning: How AWS and the NFL teamed up to transform a 100-year-old league”
It’s interactive and it’s Free.
Sensors in the Ball and on the Players
“We’ve got chips inside of the footballs now,” conveyed AWS CEO Adam Selipsky in a recent online interview.
“The quarterbacks thought they could tell which footballs had the chips in the game but that’s been tested — and they can’t.”
What kind of data is being collected? “We are able to capture data on the separation of the receiver from the defender. We’ve got probability of catch. And really importantly, we’re partnering tightly with the NFL on player safety. We’ve got over 300 sensors on the football field and on players’ bodies. And we’re looking at things like weather and the field itself and the equipment,” Selipsky said.
The vast majority of machine learning (ML) being done in the cloud today is being done on AWS, which is why AWS proved the best choice for the NFL to leverage the power of its data through advanced analytics.
If there’s a play, any play, the NFL can Stat That. The league has built several Machine Learning stats on AWS, each of which relies on different data points. Check out a few examples by clicking on the videos below. To see more, visit nextgenstats.nfl.com
Innovation in Player Health and Safety
Cutting-edge football helmets, data-gathering sensors, and a turf testing machine called the BEAST are just a few elements of the behind-the-scenes revolution making the game safer for NFL players. While many different technologies are driving this shift, they all have a common thread: data. Indeed, the NFL and its partners are collecting more data than ever to better understand how injuries occur and ultimately keep players safer while making the game more exciting.
In a new special episode of NFL Explained, NFL Commissioner Roger Goodell, Amazon Web Services (AWS) CEO Adam Selipsky, NFL EVP overseeing player health and safety Jeff Miller, NFL Chief Medical Officer Dr. Allen Sills, and several NFL-partnered engineers and innovators take fans behind the scenes into the labs designing the technologies, and into the locker rooms and onto the fields where they are used.
This 30-minute episode offers an inside look at the innovations driving the future of player safety, View it here:
- Helmet Testing: A team of biomechanical engineers at Biocore – an equipment-testing laboratory that also assesses automobile safety designs – has developed a system of testing, scoring, and ranking the effectiveness of helmets. This framework helps facilitate innovation and improvement in the helmet industry, and the results of these NFL-NFLPA laboratory tests are displayed on a poster that’s shared with NFL players and displayed in every NFL locker room.
- HeadHealthTECH Challenges & the NFL Helmet Challenge: A program of grants incentivizes entrepreneurs, small businesses, and universities to develop better protective equipment for use on NFL fields. The program has granted over $4.5 million since 2017 to advance 17 new technologies, including through the most recent competition, the NFL Helmet Challenge.
- Rule Changes: Since 2002, the NFL has used data to inform more than 50 rules changes to eliminate potentially dangerous tackles and reduce the risk of injury.
- Data Collection: Through myriad sources, including tags of players’ equipment and video analysis, the league collects ample data about every move that each player makes on the field. This provides context for understanding how, where, and when injuries typically occur – invaluable data for helping improve the safety of the game.
- The BEAST: The NFL has developed a biomechanical testing device – dubbed the BEAST – that analyzes the interaction between players’ cleats and the surface of football fields. This measurement can then be used to verify whether excessive forces are transmitted to players’ bodies.
- The Digital Athlete: A virtual representation of an NFL player is being developed with AWS to leverage artificial intelligence and machine learning to better predict and eventually prevent player injury.
These projects have improved player health, as the League has seen a sustained 25% drop in concussions over the past three seasons. With exciting new technologies in development, the bar keeps rising season after season for player health and safety.
Stat That
If there’s a play, the NFL can Stat That. The league has built several Machine Learning stats on AWS, each of which relies on different data points. Here are just a few examples:
Passing Score
4th Down Decision Guide
Completion Probability
Expected Rushing Yards
