German football league picks AWS to score with real-time data
The Bundesliga to build an NFL-style statistical platform with AWS machine learning and analytics services
Germany's top football league has partnered with AWS to bring more in-depth insight into live broadcasts and enable more personalised fan experiences.
The Bundesliga will use a range of machine learning, analytical and storage services to build a statistical platform to provide viewers with real-time information on player statistics, game outcomes and predictions on future play.
The league has the highest average number of goals per game and the highest stadium attendance globally, according to AWS, and the partnership is about improving the fan experience. It also makes Bundesliga the first soccer organisation in Europe to work with Amazon's cloud computing arm.
"Innovation means challenging the status quo," said Christian Seifert, CEO of the Bundesliga. "Working closely with AWS significantly enhances the investment we've made in innovation over the past two decades, all of which contributes to us being able to deliver a world-class football experience for our fans."
A number of sporting organisations have taken on such digital transformation projects, using some of the biggest technology companies around. AWS has had plenty of success in this regard, with deep partnerships in Basketball, Formula 1 and the National Football League (NFL).
For fans of Germany's top division, the work AWS has done with the NFL is arguably the closest example of what they can expect.
With the NFL, during live games, broadcasters can pull up statistics as the game develops. The available pass options a quarterback has will appear on the screen as a graphic; each receivable player will have a percentage for the likelihood they will catch the pass in their given situation.
The Bundesliga system will be based on a similar concept, with the goal of providing insightful information on how likely a goal is to be scored or highlight how teams are positioning and controlling the field.
Such smart real-time functions will be powered by Amazon SageMaker - a managed service that can build, train and deploy machine learning models - which will come up with its predictions and insights through analysing historical data from more than 10,000 Bundesliga games, as well as live match data.
Diversity in the digital workplace
The future of work is a collaborative effort between humans and robotsDownload now
Match day-ready machine learning
Other AWS machine learning systems will also be used by Bundesliga. Amazon Personalise, which is used to create individually personalised recommendations in real-time, will be used to offer fans personalised footage, and search results based on their favourite teams, players or matches, as well as serving up tailored marketing promotions. However, processing all the information could be an arduous task for Bundesliga’s IT team, so to optimise such operations, Amazon’s Rekognition facial recognition technology will be put to the task.
Bundesliga will create a cloud-based media archive with Rekognition being used to automatically tag specific frames with metadata on the players, team, game, and venue. This will avoid the need for the manual tagging of some 150,000 hours worth of video, and will make it easier to search for historical match footage and “pivotal plays” to be added into in-game broadcasts.
As such, Amazon’s cloud-based machine learning services will not only help Bundesliga provide its football fans with a new way to receive match and league information, but also streamline the way the football league accesses and handles its match and team data.
With cloud platforms and services being ever-more capable and accessible, Bundesliga is yet another example of how the cloud can be used to drive digital transformation in an organisation, whether that involves overhauling archaic IT systems or finding new ways to deliver services to customers.
Become a digital service provider
How to transform your business from network core to edgeDownload now
Optimal business results with the cloud
Evaluating the best approaches to hybrid cloud adoptionDownload now
Virtualisation that enables choices, not compromises
Harness the virtualisation technology that's right for your hybrid infrastructureDownload now
Email security threat report 2020
Four key trends from spear fishing to credentials theftDownload now