Big Data in Sports: From Professional Leagues to your Smart Phone

Sports is about a lot of things: physical prowess, game intelligence, and sportsmanship, among many other things. But sports are also more and more represented and perceived through numbers, data, and statistics. Statistics in sports is nothing new: pretty much every American kid has at least once in their lives exchanged baseball trading cards covered with player’s statistics. But today, we’re going further with data in sports than ever before: even a casual runner today has, at the very least, a smartphone tracking his or her performance. Tennis Grand Slam tournaments such as Roland-Garros and Wimbledon have partnered with IBM Watson to track performance on and off the tennis court. Sensors and computers at the venue will be collecting about 3.2 million pieces of data from 19 tennis courts every day. The question is: what can we do with so much data? It turns out that we can do a lot of things. After all, it was an algorithm that helped Leicester City’s recruiters build a team that, in just two seasons, went from the bottom to the very top of the Premier League – deemed “the greatest upset in sporting history” by The Economist. That being said, there are many broader use cases. With the data collected on sports fields today and Natural Language Generation (NLG) technology, match commentary can be generated automatically and in real time. There are already several functional “robo-journalists” covering different sports events. However, the power of AI in Sport is much more than just robo-journalism. Perhaps a more valuable and interesting case is using the data your smartphone or other connected objects collect on your performance to provide you with guidance on how to improve: a personal, accessible, 24/7 coach, right there in your pocket. Or another use case could be an expert system that explains data to professional players in written language, because most professional coaches are not data scientists, but they are dealing with large amounts of data that would overwhelm even the most experienced data scientists. Today, applications render such data in the form of graphs and a dashboard, and eventually a few pre-defined sentences to encourage you to carry on. And while it is always pleasing to see your performance gradually increasing in a graph, it really does not tell you much on how to care for your body and improve performance at the same time. Feed that data to an AI system that knows how to formulate sound recommendations based on your own habits and objectives, however, and you’ll get a lot more value out of all those sensors tied to your body while you’re exercising. Yseop will be present at the Viva Tech event in Paris on Thursday and Friday this week on the theme of data and sports. Feel free to stop by (stand PMU_035) and we’ll be able to talk more about what we can do in the field of personalized coaching.