If you’re lucky enough to get a ticket to this year’s Wimbledon tennis championships, be prepared to be scanned by a supercomputer: IBM Watson.
Cameras linked to IBM’s Watson “machine-learning” platform may be monitoring your facial expressions and trying to work out what emotions you are displaying.
If Watson learns quickly enough over the fortnight, it will apparently be able to work out which player you are supporting just by reading your face.
The All England Lawn Tennis Club (AELTC) and its tech partner IBM are remaining tight-lipped on the details of the new technology – not least because it needs legal approval and raises privacy concerns.
But it is another example of how sport is becoming increasingly digital, for fans, players and venues alike.
Even if Watson isn’t tracking your every cheer and grimace at the championships – which begin on Monday 27 June – it will be digesting millions of conversations on social media platforms, such as Twitter, Facebook and Instagram, and using natural language processing to identify common topics – not necessarily just about tennis.
“During last year’s final we were analysing about 400 tweets a second,” says IBM’s Sam Seddon. “Expand that out into Facebook, Instagram and more long-form content, and that’s a lot of data.
“We can come up with insights much faster than humans can and inform the media team so they can decide what kind of content they should be offering.”
Wimbledon’s digital team has a global audience to serve – the website received 71 million visitors last year – and a window of just a few seconds to persuade people to read its social media content rather than that of other publishers.
So, armed with IBM’s social media analysis, the team will be able to entice people chatting about their own country’s performance in the current Euro football championships, say, towards Wimbledon content about a tennis player of the same nationality.
“Social media is growing exponentially and is increasingly becoming the primary voice with which we communicate with our fans,” says Alexandra Willis, Wimbledon’s head of communications, content and digital.
Data-driven sport
On top of this social analysis by IBM’s “cognitive command centre”, sensors and computers at the venue will be collecting about 3.2 million pieces of data from 19 tennis courts across the fortnight. The tech company claims a sub-second response time and 100% accuracy.
This performance-monitoring data – everything from live scores to fastest serves to the number of backhand winners – is made available to fans via smartphone apps, the website, and now Apple TV.
You can personalise the app and receive every piece of relevant content on your favourite players, using data going back eight years.
But, to the surprise of many tech commentators, Wimbledon still has no plans to introduce wi-fi in the grounds, so visitors will have to rely on an imperfect mobile network to access all this data and content.
It will be interesting to see if this limited connectivity – and potentially higher mobile data costs – mars the user experience, particularly for international visitors.
While the tennis players can also use sensors inside tennis racquets and wristbands to monitor their own performance, under current International Tennis Federation rules the data is not allowed to be used for coaching purposes during matches.
But is there a danger players will become too reliant on detailed data analysis of their opponents and end up cancelling each other out?
“The great players know how to understand and react to what’s happening on the court – no amount of data analytics can prepare for that. It’s only one element of a sportsman’s preparation,” says Mr Seddon.
Super Bowl economics
The increased use of smartphones and apps is giving sports venues reams of valuable data about the way fans move around, the things they buy, and the content they want to watch.
For example, during February’s Super Bowl final at the Levi’s Stadium in Santa Clara, California, data analytics helped improve the fan experience and drive up sales of drinks and merchandise.
Tech firm VenueNext developed an app on behalf of the NFL (National Football League) and nearly half the stadium’s 71,000 capacity used it on the day to make purchases and access game stats.
“By offering an in-seat beverages delivery utility orders increased 67% during the Super Bowl,” says John Paul, VenueNext’s chief executive. “Delivery times averaged less than 10 minutes.”
A crack team of 200 ushers delivered the drinks and fed back data to the app on how long the queues for the toilets were, for example, and where the quickest place to buy a hotdog was at any time during the event.
“We also implemented express pick-up of merchandise after ordering online via your mobile,” says Mr Paul. “We ran out of inventory because it was so popular – we could’ve sold five times more than we did.”
The average spend was $212 (£145) and the most popular item was a woman’s Denver Broncos jacket costing $225, he says.
Horses for Courses
Mr Paul admits that the many breaks in play during an American football match make in-seat ordering practicable, but it wouldn’t be suitable for other sports and venues.
Indeed, Wimbledon’s Alexandra Willis says that’s the last thing they want during an intense tennis match. Nevertheless, location-based app data does help them improve signage and navigation for visitors across the complex site, she says.
During a summer of sport that includes the current Euro football championships, Wimbledon, and the Olympics, global audiences are set to grow and digital will be the main way most people access the action.
Andrew Chant, head of networks at cloud services firm Exponential-e, notes: “Since the Uefa European Championship began on 10 June, an average of 30% extra traffic has been added to networks.
“Excited by real-time intensity, we predict that the reach of this year’s biggest sporting events – from the Euros to Wimbledon and the Olympics – will extend far beyond the stadium and into the workplace, as connected sports fans live every second of the game, wherever they are.”
– Matthew Wall
