Comment: How AI can be used to improve sport content

“The biggest thing about data is that it allows you to be objective rather than subjective. In sport, people have historically relied on what they think they see or feel. Data removes the emotion, and helps you make decisions based on facts, not gut instinct.”

Those are the words of Billy Beane, a baseball player turned analyst who is one of the key innovators of the use of stats in sport. His vision of objectively looking at data to determine issues in sports is one that has taken hold not just in baseball but in almost every popular sport on the planet.

Post-match discussions for Premier League football fans seem to now focus as much on xG (expected goals), the probability of scoring from corners, and players’ tackles won and lost as they do on tactics and controversial incidents.

Meanwhile clubs in many sports use a mixture of data and video content to assess player contributions, plan future transfers and enhance their coaching programmes.

The challenge of data overload

Underpinning the growth of the use of data in sport is technology. High-quality video means it is possible to capture the most granular details of a game. For example, the Red Sox baseball team utilises up to 50 cameras across seven locations, shooting hundreds of frames a second, for two to three hours of a game. 

At the same time, the growth and availability of cloud object storage provides sports institutions with a highly secure, central repository in which to store its data.

Yet capturing and storing data also presents a new challenge for sports teams, companies, and broadcasters. They might be regularly archiving terabytes of data, but how are they going to access it? 

If you are a broadcaster looking for a specific clip from years ago, or maybe a social media editor who needs to stitch together a few moments from a game that has just finished, manually wading through enormous video libraries is simply too time consuming. 

Cloud storage plus AI – a gamechanger

The key is to harness AI to tag, sort and analyse the video content that has been archived. This is not only essential for newly created sports content, as a lot of archived video content has not been assigned the appropriate metadata – this becomes possible with AI tools.

The benefits of being able to swiftly find and retrieve sports video content can be transformative for sports, media, and entertainment industries in several ways.

For example, from a commercial perspective AI can detect when corporate sponsors’ names appear in advertising in the stadium, so companies can ensure they are getting the airtime they have paid for. 

New opportunities for broadcasters

Broadcasters can enhance their coverage by focusing on a particular player and instantly compiling mini clips of each involvement they have in a specific game. Another example could be that they use AI to find and stitch together footage that celebrates the career of a particular player or rounds up highlights of the best moment of historical interaction between two teams.

It can create closer collaboration between sports organisations and broadcasters in that the former can more easily find and share appropriate content with the latter. For example, content produced by Tennis Australia is sold to broadcasters, sponsors, and partners to repurpose for their respective outlets around the world. 

It can also create media opportunities that sports companies can use to engage and deliver closer relationships with their fans. Liverpool FC, for example, boasts its own 24/7 TV channel, as well as an OTT platform. It creates all the media assets for live games, the club’s websites, fan-facing applications and 15 social channels.

AI is also the catalyst that has made instantaneous personalised content on demand possible. It can even generate the commentary to accompany the footage that has been created.

One key thing that sports organisations and broadcasters need to watch for though is egress fees (charges from cloud providers for moving or transferring data from the cloud storage where it was uploaded). Paying a fee for each time a company accesses a cloud system can significantly increase the cost of producing the content.

With the combination of the right AI tools and appropriate cloud-based storage, sports companies and broadcasters can keep fans happy with dynamic, personalised content. This combination will also open many more content-based and commercial opportunities in the years to come.

Aaron Edell is SVP of AI at Wasabi