If there’s one thing that’s transformed football over the past decade, it’s data. From the trackers players wear under their shirts, to the dozens of cameras mounted around stadiums, to the teams of analysts in every elite club’s backroom staff, data is football’s most valuable currency.
But it’s not only the elite clubs with their billionaire backers who are benefiting from this influx of football data. Thanks to affordable cloud computing platforms, even junior teams playing on local park pitches can benefit from video analytics that would have made professional clubs jealous a mere decade ago.
Cloud computing has helped bring data analytics to every level of football, from Barcelona down to the grassroots. And it’s changed the game for good.
The data elite
That data has infused the elite game will come as no surprise to anyone who’s spent any time watching Premier League matches. Coaches hunched over tablets in the dugouts, real-time statistics being displayed by broadcasters and players running around with visible tracking tags in the back of their shirts are just some of the signs that the elite game no longer relies on the human eye to interpret what’s unfolding on the field of play.
Football data has become an industry in its own right. Clubs make buying decisions based on the player’s data metrics, while terabyte upon terabyte of video is collected from every elite game. Analysts spend days pouring over opposition data and footage to find that marginal gain that will give their team the edge in a forthcoming fixture.
Data is collected to feed unprecedented levels of information to fans, too. At the Premier League level this involves collaboration with some of the world’s biggest cloud computing companies. In 2021, for example, the Premier League announced a partnership with Oracle Cloud to deliver a new set of AI-powered analytical models that give supporters on the sofa a clearer insight into what’s happening on the pitch.
These included average positioning tracking, which show how teams organise themselves, both when they have possession of the ball and when they’re defending. It included AI models, working on data collected from thousands of historical games, to predict how likely the team in possession is to score in the next ten seconds. And it used AI modelling to provide a win probability for each team, by simulating the remainder of the match 100,000 times.
None of that would be possible without a huge amount of cloud computing power on tap. The analytics engine in the Oracle Cloud infrastructure, for example, can count the number of sprints made by a team’s midfielders over the past five minutes compared to the previous 40 minutes. And compare that trend with historical scoring outcomes, according to Lee Bonfield, a member of Oracle Cloud team. “All of those different criteria can be taken in to model the game a thousand times and say, 700 times out of a 1,000 this kind of movement and speed in the midfield, combined with this number of successful passes, make for a score in the next five minutes,” he said.
The video bank
Although football analytics isn’t solely reliant on video footage, it’s a huge part of it. And with multiple 4K cameras capturing the action from different angles within the stadiums – and even on the training pitches – of elite clubs, that’s an awful lot of data that needs storing.
Of course, it’s not only the analysts who rely on huge banks of video footage at the elite clubs. Video is a key part of the mix for the marketing teams, too, often with 24/7 TV channels, club apps and social media to feed. The huge storage requirements for video, coupled with the need to give access to staff working in locations right around the world, has seen many clubs move their infrastructure to the cloud.
Liverpool, for example, struck a partnership with Wasabi to move all of its digital assets to the cloud, helping the club reduce the cost of maintaining its on-premise infrastructure and making it easier for the different stakeholders to access content. “The second anything gets ingested into our building, it is replicated onto the Wasabi active archive,” Drew Crisp, senior vice president of digital told Wasabi’s website. “This makes it instantly accessible by anyone on our team, whether they are on-premises or off.”
Crisp added: “We often get external requests for assets that may be rights restricted. By connecting Iconik [the club’s media management system] to Wasabi, all of this can be end-to-end. Content can be found and accessed wherever it is. Team members will be able to self-service the clips they need for their projects, which can automatically be fed through our approval/rejection workflow.”
One of the great things about the availability of cloud computing is that it’s not only the Premier League giants who are benefitting from this technology, but tens of thousands of grassroots teams too.
Veo is a Danish company formed in 2015 after one of its co-founders, Keld Reinicke, was running late for his son’s football match and missed him scoring his first ever goal. The frustration of missing his son’s big moment led him to wonder why all football matches weren’t being recorded, and within a short time the company had created Veo Cam, a one-camera system that automatically tracks the ball across the pitch. If you’ve been to any Non-League or grassroots game recently, you may well have seen one of Veo’s distinctive green cameras mounted on a tall tripod at the side of the pitch.
Veo is more than just a camera. It relies on Amazon’s AWS cloud computing infrastructure to house the video captured by the cameras and run detailed analysis on the footage, allowing even amateur coaches and players to benefit from high-level analytics such as heatmaps showing player positions, 2D maps of player positions during the game and automatically generated statistics showing possession percentages, shots on target etc.
“It’s all about proving our ethos, which is that we want to democratise technology in sport,” said Rob Scotland, head of brand marketing at Veo, who uses the technology with his own son’s under-14s side. “Typically, when you go to one of the top teams, they have a staff of people, they have the technology that you would actually wear, all those kinds of things that have super-accurate data. They’re working around the clock. A Non-League team, or even a school team, just wouldn’t have that.
“We don’t use trackers that make it more complex to set up – everything goes through the camera. And therefore there’s a pretty smart artificial intelligence going on to be able to detect everything, down to heat maps, possession and shots, and those types of things.”
And although the cloud computing infrastructure was and remains key to Veo’s offering, the company’s increasingly relying on in-camera processing to accelerate the time it takes to get analysis to impatient coaches. “The majority of the processing is done in the camera itself, as it’s being recorded,” said Scotland. “Then the upload goes up into the cloud where additional bits are then adjusted. From there, you’re then able to access your game a lot quicker than you would have done before.
“Before it could have been anything up to 12 hours [before you got analysis back] because it would have gone up into the cloud, then it would have been processed. This way is a little bit more efficient.”
In fact, because of the shift to in-camera processing, impatient coaches can even watch the match back as soon it’s finished in streaming mode, even if they have to wait a little longer for the full stats to be verified when the match is uploaded to the cloud. “It’s a mixture of massive innovation and massive cost efficiencies in cloud computing,” said Scotland. “Put the two together and it starts to become quite an interesting competitive advantage for coaches and players.”
Nathalie Parent, Chief People Officer at Shift Technology: “HR is the conscience of an organisation”
For more than 30 years, Nathalie Parent has led global HR teams, working primarily with software companies. Today she’s Chief People Officer at Shift Technology
Amazon introduces new storage class that makes it cheaper to store rarely used files
Robot carers are real, but caregiving has bigger problems, writes Richard Trenholm in this FlashForward edition