The newest NFL star could be a data scientist: Using ML and AI to keep players safe

The NFL on Monday opened its crowdsourced challenge to help the league analyze and identify player contact during games using AI and machine learning. The challenge is a part of the NFL’s innovative partnership with AWS to create the Digital Athlete, the virtual simulation of an NFL player that will help the league predict and prevent player injury.

The challenge is hosted on Kaggle, an online community that invites data scientists and machine-learning experts to solve real-world challenges using data sets. The NFL’s Contact Detection Challenge will be open until March 1, 2023, and the first-place prize will be $50,000.

The competition’s goal is to detect physical contact the players experience in a game. Participants will watch game films to analyze moments of player contact and help improve player safety. Then, machine learning and computer vision will help the NFL identify at which moments in a football play players experience contact.

In past NFL-sponsored Kaggle competitions, participants aided the league in creating helmet impact detection and identification algorithms, but the league is looking to take it a step further. The results of The Contact Detection Challenge aim to improve the league’s ability to predict moments when players will be in contact with each other and with the ground.

The NFL says these results will “improve the NFL’s ability to quantify when contact occurs on the field, allowing the league to better predict – and ultimately prevent – player injuries,” according to a press release. 

The NFL and AWS’s collaboration to create the Digital Athlete is a years-long technological feat that should supply the NFL with quantifiable data to help predict and prevent severe player injuries. 

The Digital Athlete will serve as an automated experimental NFL player who will experience infinite simulations of football plays based on in-game data analyzed by participants on Kaggle. 

The Digital Athlete will supply the league with information that will determine player safety based on the type of cleats a player wears, the surface they play on, the weather conditions, and many other factors. 

The data extracted from the Digital Athlete’s simulations will allow the NFL to construct training and injury recovery programs better, conduct real-time injury risk analysis during games and help the league craft improved rules to ensure player safety. 

The partnership between the NFL and AWS could help prevent injuries like the ones that happened to Tua Tagovailoa, the Miami Dolphins quarterback. In weeks three and four of the current NFL season, he suffered back-to-back head and neck injuries, conjuring scrutiny from the NFL Players Association, fans, and former and current NFL stars.

The simulations could also prove the risks of playing on slit-film turf fields. Recently, players have complained about lower extremity joint pain and sprains caused by the turf. Though they are not burdensome injuries, the Digital Athlete could help the league understand how the turf causes them, how to treat them, and how to mitigate them.