The majestic appeal of the Tour de France is its winding route for 2,200 miles through 21 stages, five mountain and four countries on the 2017 course, but that sprawling route creates challenges not just for the riders but also for the broadcasters.
The standard contextual clues in other sports that fans can rely on to make sense of the unfolding action — football’s 50-yard line, baseball’s second-base bag, hockey’s red line or soccer’s goal box, to name a few — are missing. To help viewers understand the cyclists’ speed and distance traveled, the Tour organizing group Amaury Sport Organisation has teamed up with Dimension Data, a global company with headquarters in Johannesburg, South Africa, to provide tracking data through under-seat sensors on each bike.
For the first time in 2017, that effort has expanded to include machine learning technology for more predictive real-time analytics. While last year’s Tour de France generated some 128 million data points, that total is expected to reach 3 billion this year, incorporating course gradient and weather conditions, the information gleaned from each cyclist’s GPS transponders and historical data from prior tours and the cyclists’ personal riding profiles. Among the promised new uses: projecting whether the peloton will be able to overtake any out-front riders separated from the pack.
For instance, this graphic showed the final stretch run in which Slovak rider Peter Sagan held off late charges to win Stage 3.
Always in front, Peter Sagan (BOH) proved (again) he had the power to hold off any rival despite unclipping 200m from the finish.#TDFdata pic.twitter.com/FcB4lCFj5X
— letourdata (@letourdata) July 3, 2017
And here are the speeds of the top three in Stage 2, with Marcel Kittel out-sprinting Demare Arnaud and Andre Greipel over the final 500 meters.
Kittel went full steam from +300m out & kept his lead to the end.
This is the 3rd time he wins the 1st bunch sprint of @letour.#TDFdata pic.twitter.com/RPr2EIAG78— letourdata (@letourdata) July 2, 2017
Get The Latest Sports Tech News In Your Inbox!
As the Tour progresses, Dimension Data plans to use the hashtag #TDFpredict for social interactions related to race predictions. The intricacy of the projections extends beyond simple time and distance parameters — Rider 1 is a quarter-mile off the pack with two miles to go, for instance — based on the upcoming terrain, temperature and rider history, with seven different categories for past performance such as sprint, flat ground, mid- and high-mountain.
“What’s especially exciting for us is how we’re helping ASO to attract a new generation of digitally savvy fans,” Scott Gibson, Dimension Data’s chief digital practice executive said in a release, “and how advanced technologies like machine learning are opening up new possibilities for providing the insights that today’s fans demand.”
Christian Prudhomme, the director of the Tour de France, added, “Today, our followers want to be immersed in the event. They’re more digitally engaged on social media than ever before, and want a live and compelling second-screen experience during the Tour. Technology enables us to completely transform their experience of the race.”