Telecom Italia (TIM) is the main sponsor and partner of Serie A, the major italian football league. Their dedicated mobile app broadcasts football content to millions of fans worldwide. TIM also manages accurate, live, raw data and statistics coming from the football fields. We have been asked to bring this data to life, creating storytelling out of numbers. For the past 3 seasons, Algo created dozens of live video stats for every match of the competition, entertaining football fans and explaining match dynamics.
To cover the 90 minutes of game, we conceived 14 different templates on different topics: Attack, Defense, Line-ups, First Half, Hat-Trick and so on. During every game, a bespoke algorithm analyses hundreds of data points to evaluate, for example, if a video about the Lack of Fairplay would be more interesting than one about Aerial Passes, in order to tell the story of today’s match. To increase variety, once a video template has been rendered, it gets pulled out, so that it won’t render twice.
Among this set of templates, we’ve developed some dedicated to promote the match and raise expectation before the event starts. In these videos we display information like timing & venue, line-ups and even more complex data like when exactly a given team is most likely to score, based on historical data from the past seasons.
Technically speaking, on every match day, our software picks the right templates and customises them by changing team logos, adding data about every kick, every tackle, every scorer, producing +120 videos per weekend, that are pushed to viewers on the app.
To enhance the live nature of sport events, the videos are created and delivered in the shortest timeframe possible, and by shortest we mean in tens of seconds. They also need to be super light-weight: when simultaneously streaming to a few hundred-thousands football fans, every bit is a cost.
Since sports data can vary a lot, customisation is a complex aspect to tackle. Although in most cases scores are going to be 2 to 4, what happens to a video in a no score scenario? What happens instead if they are 12?
For this reason, we usually end up creating a rig, a series of code expressions that connect all the elements of a scene to a single input data, so that by changing that single data, the whole scene transforms to deliver a different message visually.
In order to maximise how different the videos appear to each other we make a smart use of colours. Each video is customised by cycling through the swatches of a 5-colours palette, so that consecutive videos always feel visually different.
Premium content distribution
Videos are published as premium content within the Serie A TIM app - where they rank as the third most watched content (after match live streams and highlights, but above match synthesis and interviews). It’s worth mentioning we achieved this result without using live action imagery -covered by expensive rights. Instead, we engage users by delivering meaningful insights through colorful animations and provide a 2nd screen experience to millions of fans.
All sports love data.
Beyond helping football teams and leagues to communicate with data-driven videos, Algo can work with any kind of sport organisation to enhance the way video is used to entertain, attract, convert your audience into fans. We did so for the Olympics, producing one video in real-time for every medal won, and experimented with data-visualisations for many other sport competitions: from basketball and tennis to formula 1, cycling, golf, running and skiing.