EMPOWERING SPORT WITH CLOUD
Author // Stuart Hodge
Amazon Web Services is partnering with sporting companies and leagues around the world with the aim of enhancing the experience for all stakeholders
Sports and data have enjoyed a symbiotic relationship for decades but, for a long time, data was used mainly by sporting institutions or the teams, and mainly internally, to gain competitive insights or advantages. Nowadays though, data and analytics are also at the heart of enhancing the fan (or customer) experience and, increasingly, holistically bettering the sports themselves either in terms of safety improvements or technological advancements which further competition. No company is making a more noticeable impact globally, or across a wider range of sports, than Amazon Web Services (AWS) to that end.
In the first half of 2021 alone, it has announced partnerships with the NHL (National Hockey League) and PGA TOUR golf, and has also revealed new advanced analytics with Formula 1 racing and Germany’s Bundesliga football league, amongst others. In fact, AWS has more existing sports partnerships than any other provider on the market, including with the Seattle Seahawks NFL team, the NFL itself, NASCAR racing, the Los Angeles Clippers NBA team, Pro Football Focus and Second Spectrum (American sports analysis platforms), Six Nations Rugby, Churchill Downs racecourse and the PAC-12 conference of collegiate athletics in the United States.
However, it’s not just the volume and variety of what AWS is doing that is so interesting, it’s what they’re doing and how they’re doing it that’s truly exciting.
Cutting-edge technology empowering enhanced fan and sporting experience
· Machine learning is producing advanced stats such as NFL’s Next Gen Stats, F1 Insights, and Bundesliga Match Facts · High performance computing is helping to redesign vehicles such as F1 race cars and America’s Cup boats
· Organisations such as the NHL are using artificial intelligence services to automatically tag and categorise archives · And the Internet of Things is connecting sensors from the field and track in various sports to deliver real-time data which partners can then use
Take Formula 1 for example: the fastest-moving, fastest-changing, most data-laden sport in the world. Rob Smedley, Director of Data Systems at Formula 1, joined Priya Ponnapalli, Principal Scientist and Senior Manager on the AWS Machine Learning Solutions team, recently to discuss how F1 is using Amazon's technology to enhance fan experience. Smedley said: “Formula 1 is such a data-rich sport and when we decided, as a league, to try to understand how to expedite, and use that data to look for broader fan engagement, and a more immersive experience with our fans, that's when our partnership between Formula 1 and AWS really kicked off. “We generate between 2.5-3 billion combinations of timing transponder events in every single race. Then we've got weather data, car telemetry data, the metadata, and lots more besides, all of which helps make up this massive live stream of data. Then it was (up to us to find) a way, esoterically, of trying to build something out from that in an engaging and usable format for the fans. “That's why, over the last few years, we've built this suite, this library of data analytics, and stats that sit on screen throughout any Formula 1 event.
“Formula 1 is such a unique sport in as much that the linear TV feed can only ever concentrate on one or two stories at the same time. But by using data, we can get behind the story in such a technical sport, in such a data-driven sport, using graphics to bring everything to life.” Smedley goes on to enthusiastically describe just one of a host of new graphical insights being unveiled this season for F1 fans. “Listening to the feedback on the new braking performance graphic we’ve introduced this year, for example, it's been absolutely great,” he says. “This is a visceral display through data and analytics of just how potent an F1 car is. Braking at more than 5Gs (gravitational force) – so, if you take Monza, turn one, as an example, that famous chicane, decelerating from 350-360 kilometres per hour down to 40, 50 or 60. “These Formula 1 cars, they're absolute monsters and we can bring that to life now. We can engage our fans by using that graphic (to show) exactly how late drivers brake. The breaking point, even one or two metres closer to the corner, (demonstrates) about a level of bravery that is just difficult to comprehend.” Formula 1 is also using its data partnership with AWS to enhance the standard of the cars themselves. It has been using Amazon’s cloud computing platform, Amazon EC2, for ‘Computational Fluid Dynamics’ (CFD) to simulate race car aerodynamics, essentially achieving the performance of a super computer at a much lower cost and reducing simulation time by an average of 70 percent, from 60 hours down to 18 hours. Within that project, Formula 1 used over 500 million data points to study downforce loss when two vehicles race in close proximity (a car’s downforce increases its tyre grip and cornering speed, thus reducing lap time) and the sporting world’s elite wind-tunnel aerodynamicists are now taking this data into consideration for the design of future F1 racing cars. That F1 partnership example is merely focusing on a couple of angles within a single sport. Some of the other ways that AWS’ technological infrastructure is being used by various sports are arguably even more intuitive.
The Amazon technology chain underpinning it all
AWS’ respective partners build cloud-based video archives that will automatically tag specific frames, from hours of video, with metadata so that they can easily search historical footage and surface pivotal moments. Then, as new data is collected, it gets ingested using Amazon EC2. That cloud product is scalable depending on the amount of data each partner has, and data is then transferred via live APIs (application programming interfaces) into a data lake created using Amazon Simple Storage Service (Amazon S3) and processed using Amazon EMR (the company’s big data platform). Partners can then take the data and develop graphics in real-time to bring sports analysis to life on-screen. The AWS cloud is integral to making everything work: it means that data can be processed, streamed, stored and analysed in real-time.
Matt Hurst, Head of AWS Sports Marketing and Communications, explains: "Our strategy in sports is no different than in every other industry: we work backwards with our customers to help solve business problems through innovation. “For instance, in 2020, the NFL had invested millions of dollars in creating a major fan experience for the 2020 NFL draft and COVID-19 placed all of that investment in jeopardy. They came to AWS to develop a reliable, secure way of hosting its first virtual draft during the global quarantine and we partnered with them to create the most-watched NFL draft in history. “AWS technology really is changing the game of sports by enhancing and enriching the fan experience before, during, and after games through rich data and insights and second-screen experiences. “But AWS technology is also transforming more than the fan experience – leagues and teams are using AWS to innovate like never before across a range of sports.” A prime example of this is with the NFL, which uses data collected through RFID (radio-frequency identification) tags on player’s shoulder pads, in pylons, first-down markers, on the in-game officials, and even within the game ball. These all transmit location data to wide-band receivers, specific to the inch. It is a game of inches after all.
Data points such as (quick sporting jargon alert) air distance, target separation, sideline separation, pass-rush separation, passer speed, player velocity, acceleration, orientation, direction, and time to throw are calculated and stored in the AWS cloud – amounting to over 3TB of data per game week, and over 300 million data points during a full NFL season. The stats buffs are obviously the first to take advantage of this vast plethora of information, but there’s a bigger picture at play as well. By combining the NFL's vast trove of data, specialist knowledge and insights about the sport itself with AWS’ machine learning and cloud computing expertise, the partnership is working together to transform player health and safety by generating new insights into player injuries, game rules, equipment, rehabilitation, and recovery. The data generated will eventually be available to researchers, equipment manufacturers, trainers, coaches, and medical professionals to serve as a framework for future innovation. It’s all fascinating stuff. Of course, the most popular of the global sports is undoubtedly football (or soccer, depending on which side of the Atlantic you’re on) and in the Bundesliga, in Germany, they’ve been partnering with AWS to provide what are being called ‘Match Insights’ for fans, which enhance the viewing experience and provide additional material for analysis. Three new iterations: ‘Most Pressed Player’, ‘Attacking Zones’ and ‘Average Positions’ made their debut in February this year, joining ‘Speed Alert’, ‘Average Positions’, and ‘xGoals’, bringing the total number of insights available for Bundesliga fans to six. Again though, there is a bigger picture. Together, the partnership is building a next generation advanced statistics platform that covers historical data from over 10,000 Bundesliga games. The Bundesliga powered-by-AWS statistics service will be the only official source, deliver data in real-time and provide a new user experience and visualisation on mobile devices. The idea is to create a truly personalised fan experience which will enable real-time recommendations and adjustments based on the user interactions with Bundesliga platforms.
“Every Bundesliga match generates data that can improve play and help fans better understand team strategies, and we are making tremendous strides in leveraging the vast amount of data in our archives and from our league’s current games to develop and roll-out new Match Facts. The advanced statistics that we’re creating with AWS give fans an even deeper appreciation for how the game is played,” said Andreas Heyden, Executive Vice President of Digital Innovations for DFL Deutsche Fußball Liga GmbH. “Together with AWS, we’re delivering a new perspective on what happens on the field and offering a new and engaging way for fans to follow their favourite teams.” “Expanding our work with Bundesliga means more fans will gain an appreciation for the incredible talent on the field and the decisions made by teams, at the same time as the league differentiates itself through the use of advanced analytics to improve the quality of play,” said Klaus Buerg, General Manager for AWS Germany, Austria, and Switzerland, Amazon Web Services EMEA SARL.
As ever, the bigger picture is always part of the thinking on both sides of the partnership. The final word on all of this, we will leave to veteran NBA coach Stan Van Gundy. “Understanding the assumptions behind any statistic ‘makes you think’ about your own views,” he says. “That’s the essence of analytics: don’t trust, but verify – with data.” And the capacity to do that, thanks to AWS and the strong partnerships that it is developing, is undoubtedly greater than ever before.