NHL on AWS
AWS teams with the National Hockey League to be the Official Cloud Infrastructure Provider of the NHL.
New Metric, Ice Tilt, Measures Momentum During NHL Games
AWS and the NHL quantify this nuanced aspect in hockey that had previously been deemed immeasurable. Ice Tilt provides accurate, real-time visibility into which team has the edge – and by what margin – at any given point during game play. It does this by continuously tracking the player position and the puck, gathering and transmitting over 1,300 thirteen data points per second.
WHY THE NHL CHOOSES AWS
Hockey has gone from a game of seconds to milliseconds. But deriving insight from the large-scale volumes of data remains challenging. Hear from the NHL on how they’re using AWS to develop new advanced insights that differentiate players and their performances in a way that wasn’t previously possible.
Unlocking Data's Potential
Using more than 10 years of the NHL’s historical Hockey Information and Tracking System (HITS) data, the league will harness the power of this data to produce advanced statistics through the use of innovative AWS technologies such as AWS Lambda, Amazon SageMaker, and more.
Engaging and Delighting Fans
Fans will gain deeper understanding into players, teams, and the league because AWS’s leading technology will help deliver the next generation of advanced analytics. New real-time hockey stats will transform the fan experience before, during, and after games to help fans better interpret and appreciate the action on the ice.
Enhancing the Broadcast Product
Video feeds processed by AWS and the NHL’s new 4K cameras will capture every angle on the ice, providing better knowledge of what’s happening throughout the game for better storytelling. The NHL is building a central repository on AWS which will help retrieval of archived video footage giving instant access to NHL content to deliver new in-game analyses, predictions, and video highlights to enhance mobile, online, and broadcast experiences.
Learn how NHL EDGE IQ, powered by AWS, is scoring big with game-changing technology. From real-time analytics to player performance insights discover how AWS is helping the league and its players enhance the game with data-driven insights.
See How AWS Can Predict Likely Face-off Winners
Go under the hood to see how the ML Solutions Lab team built Face-off Probability with the NHL
Key Obstacles
Modeling the face-off event as an ML problem, collecting parameters needed for prediction, calling the model, and submitting results to broadcasters
Building an ML Model
By continuously listening to NHL’s expertise and testing hypotheses, AWS’s scientists engineered over 100 features that correlate to the face-off event.
Determining A Face-Off
Go under the hood to see how the AWS AI and Machine Learning teams built Face-off Probability with the NHL.
Top Factors Impacting Opportunity Analysis
While there are multiple factors that will ultimately contribute to each scoring opportunity – the Opportunity Analysis model has shown the following three factors most consistently contribute to the quality of the opportunity.
Goalie Positioning
Opportunity Analysis looks at 7 different factors across goalie depth, distance, and height to contribute to a rating.
Puck Movement
Opportunity Analysis looks at 5 different factors across puck direction changes, puck speed range, average puck speed, and angle discrepancies to contribute to a rating.
Traffic
Opportunity Analysis looks at player positioning between the shot origin and goalposts as well as possible vision obstructions and defensive puck positioning to contribute to a rating.
ENGAGING THE FANS
Hockey is predicated on fast-moving action on the ice, and the NHL is producing its set of advanced statistics by leveraging AWS’s unmatched set of cloud computing technologies, the NHL is producing its set of advanced statistics. Using over a decade of historical data, the NHL can now provide in-depth analysis from its rich history with real-time game action.
-
SHOT ANALYTICS
-
SAVE ANALYTICS
-
FACE-OFF PROBABILITY
-
SHOT ANALYTICS
-
SHOT ANALYTICS
Each shot taken in every game will include a number of variable factors such as from where on the ice the shot was taken; if a team is playing even-strength, short-handed, or in a Power Play; the score the game; the time of game; and more.
-
SAVE ANALYTICS
-
SAVE ANALYTICS
Every save a goalie makes will now include a goalie’s save percentage based on location the shot came from, different team strength conditions, time of game, game conditions, and more.
-
FACE-OFF PROBABILITY
-
FACE-OFF PROBABILITY
Our first AI & ML driven stat that generates predictions for who will win the upcoming face-off based on the players on the ice, location of the face-off, and current game situation.
"AWS is the perfect partner for the NHL, helping us to aggregate our rich historical content and deliver real-time advanced analytics that will continue to transform the fan experience. The delivery of these stats will unlock opportunities for fans, along with players, coaches and broadcasters, to see the game in new ways and to gain a deeper understanding of the strategic decisions made during a game."
- Dave Lehanski, Executive Vice President, NHL Business Development and Innovation
"AWS is the perfect partner for the NHL, helping us to aggregate our rich historical content and deliver real-time advanced analytics that will continue to transform the fan experience. The delivery of these stats will unlock opportunities for fans, along with players, coaches and broadcasters, to see the game in new ways and to gain a deeper understanding of the strategic decisions made during a game."
- Dave Lehanski, Executive Vice President, NHL Business Development and Innovation
