Sports, data analytics and machine learning. Three words you would never expect to be in the same sentence, right? Well, what if we told you that they already are in the same sentence in sports teams the world over. That’s right, we’re already seeing the inclusion of data analytics and machine learning in sports – some even as early as 15 years ago. You’d be surprised how advanced things have gotten when it comes to data analytics and sports; we’re even seeing companies use Amazon Web Services (AWS) to help deal with and store the data.

In sports such as the F1, American football and even rugby we’re seeing more and more decisions being made when taking into consideration probabilities and numbers generated by machine learning. In fact, one of the sports most adept at using data is the Formula 1. Teams generate up to 600GB of data per lap from the 200 to 300 sensors in the cars. When it comes to the American NFL (National Football League) each player is analysed based on over 100 data points. These data points drive the plays we, as fans, cheer and look for when we watch the athletes play.

Dilemma: Where to store the data? How to capitalise on it?

When it comes to dealing with the data generated from these sports, the first dilemma is where to store the data. Of course, Amazon Web Services has a slew of container and data lake services such as Amazon S3 storage and more these teams are already using to store their data. However, just keeping the data in the cloud isn’t enough. They will need to run through and analyse the data for it to truly be useful to the teams. That’s where machine learning comes in.

While it might seem like a brand-new paradigm, we can assure you, that it’s been happening behind the scenes for quite a while. Teams in the F1, NFL and even rugby have been collecting data and analysing them to help players perform better, drivers drive better and engineers optimise their technology further. In fact, there are companies out there such as Pro Football Focus that actually process and analyse the data in real time. In fact, at AWS Re:Invent, Cris Collinsworth, CEO and Co-Founder of Pro Football Focus, said that what used to take coaches around two to three days to analyse is done in less time. He said that with this improvement, coaches are given more time to strategize and tweak their plays to help their teams win.

The data collected during the races of the F1 doesn’t just go to the cloud for storage. Analysts on the ground are constantly looking at it to help tweak and make critical decisions for that edge. In fact, the data plays a big role in the teams pitting and undercutting strategies in a race. The engineers are also using this data to help with their car design and tweaking between races. However, the F1 has a pretty good head start compared to other sports out there. They’ve been using data analytics in their sport for over a decade now and have been able to use it to help with performance. However, that isn’t the only way they use their data, they also use it to create new regulations that affect the whole game and the welfare of the drivers.

Machine Learning in Capitalising on Collected Data

With the advent of machine learning in the past few years, the work of analysing the data has been made even easier. Using services like Amazon SageMaker, companies and teams are able to take advantage of the numerous data points in real time. Machine learning algorithms can churn out predictions and probabilities based on the collected data near instantaneously.

That said, the data generated by the machine learning algorithms is only half the picture. It informs the coaches and players of not only the probabilities and possibilities but also what could be done to help give the teams an edge over the competition. The decision making process on the pitch or track is no longer only a question of gut instinct, it’s about tempering and guiding the gut instinct with mathematics.

We are at the crossroads of a change in sports paradigms. Coaches are beginning to accept that the data being processed by machine learning algorithms as guides for their game time decisions. The game is changing based on how teams are able to use and optimise machine learning to get the edge they need during game time.

Creating New Fan Experiences

That said, machine learning isn’t just giving the edge during game time. It’s also being used to create new fan experiences. Watching sports can become a pretty mundane experience for some. However, using machine learning and data analytics, broadcasters can create new experiences for fans to keep them more engaged.

In the United States, broadcasters have been experimenting using data lakes and machine learning to enhance the sports viewing experience. This isn’t just restricted to F1, NFL, NBA or the MLB. It’s across the board. These broadcasters are using machine learning to create overlays and explanations of complexities that help fans better understand the sport. In fact, with the amount of data they have at their fingertips, shout casters and commentators are able to see plays before they happen or even suggest some that would have led to a better outcome. These hints of information are also opening up the sports world to new audiences. It is also creating a more engaging experience for long time sports viewers and fans.

Given the amount of data being collected, it also comes as no surprise that broadcasters and even teams are looking into giving fans a better experience via a second screen. They are looking at what information would make sense and enhance the experience for viewers. Of course, raw data isn’t the answer but the data processed by machine learning algorithms are able to give a better understanding and appreciation to fans. In fact, they expect that it would engage a whole new type of viewer.

With all the emphasis on machine learning and data analytics, it would seem that sports will be reduced to 1s and 0s. However, as Rob Smedly highlighted, artificial intelligence and machine learning can never replace the driver or player. In fact, the thing that makes sports engaging is the human element in the game. It’s about how athletes are able to push boundaries of human performance and how we use the data to improve, not only the game, but also other aspects of human life.