WHY THE NFL CHOOSES AWS TO STAT THAT

The vast majority of machine learning (ML) being done in the cloud today is being done on AWS, which is why AWS is the best choice for the NFL to leverage the power of its data through sophisticated analytics. The NFL uses the power of AWS ML to stat that: creating new stats and improving player health and safety, while creating a better experience for fans, players, and teams—all in real time.

Machine Learning

It’s how the NFL can Stat That, using machine learning and data analytics services to boost the accuracy, speed, and insights provided by its Next Gen Stats platform.

Quick Access

Using the business intelligence tool Amazon QuickSight, the NFL is able to gain greater insight while also opening a window for fans, broadcasters, and editorial to engage with data.

Speed

Using Amazon SageMaker to build, train, and run predictive models helped reduce the time to get results from as much as 12 hours to 30 minutes.

STAT THAT TO ENGAGE THE FANS

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. To see more, visit nextgenstats.nfl.com

Passing Score

A first-of-its-kind AI tool that combine seven ML models, including a new model to predict the value of a pass before the ball is thrown, to evaluate quarterback passing performance.

Completion Probability

This predictive model uses Amazon SageMaker to compute the probability that any given pass will be completed based on the distance of the pass, the receiver’s separation from the nearest defender, his spot on the field, the amount of pressure on a QB, and more.

4^th Down Decision Guide

Uses Amazon SageMaker to analyze win probability, which informs how the game will change based on hypothetical outcomes, and conversion probability, which predicts if the offense will convert a fourth down or 2-point conversion.

Expected Rushing Yards

This metric uses Amazon SageMaker to predict how many rushing yards a ball-carrier is expected to gain on a given carry based on the relative location, speed, and direction of blockers and defenders.

APPLYING MACHINE LEARNING TO THE DATA

By leveraging AWS’s broad range of cloud-based machine learning capabilities, the NFL is taking how it can Stat That on game-day to the next level—so that fans, broadcasters, coaches, and teams can benefit from deeper insights.

Training data from traditional box score statistics, as well as data collected from the stadium, will run through hundreds of processes within seconds with the output fed into Amazon SageMaker. From there, machine learning models built by the NGS team ingest the data, which continually train and refine the models. The machine learning models are then used or inferenced in real-time during games to generate outputs such as formations, routes, and events.

"We chose AWS because of its combination of advanced cloud offering, powerful machine learning capabilities, and experience operating at the scale we need. By powering Next Gen Stats with AWS, we’ll be able to kick off our [season] with even more impactful and meaningful content, uncovering deeper insights into the game of football than we’ve ever done before."

- Matt Swensson, Vice President, Emerging Products and Technology at the NFL

GOING LONG ON MACHINE LEARNING WITH THE NFL

Want to see the playbook? Read more about the NFL’s machine learning journey, with an introduction from NFL CIO Michelle R. McKenna—and hear from Matt Swensson firsthand how his Next Gen Stats team worked with the AWS Machine Learning Solutions Lab to build, train, and deploy their machine learning models on Amazon SageMaker.

Get the e-Book