What does the explosion in data means for the future of sport?
Will talk of players being on a “hot streak” or “in form” still be the norm by 2020? Or will post-match analysis be based on machine learning and predictive analytics?
And instead of a team’s success being attributed to players or management, will it come down to whose data centre can deliver real-time strategical and tactical information?
The idea of data usage in sport was popularised by Moneyball, a book and film about the Oakland Athletics baseball team. The general manager (Billy Beane) needed to build a team to compete against richer rivals. He opted for a data-driven, sabermetric, approach to finding undervalued players. This involved establishing different parameters for measuring performance. For example, instead of looking at how many runs a player scored, sabermetrics would evaluate the context of the runs scored. When a game had already been won? Or when the pressure was on?
This knowledge led to the team setting a league record for consecutive wins in 2002. In the playoffs Oakland narrowly failed to beat the New York Yankees, a team with four times their spending power.
Of course, data is now a key part of professional sport. And with advances in wearable technology, innovation is showing no signs of slowing down. At sport’s elite level, a split-second can mean the difference between winning and “what if”. Here’s how some sports are using data to deliver better results.
Combined revenue for the top 20 football clubs reached EUR6.2 billion for season 2013/14. Yet it’s only relatively recently that the game has begun to embrace the power of data. Managers are still seen in the dugout of a game, giving orders based on what they see. But behind the scenes, new partnerships are happening that could change that.
SAP has an arrangement to provide real-time data analytics to German club TSG Hoffenheim. Players wear sensors which enable club officials to analyse their activity during a game. This Internet of Things approach is giving the club unprecedented volumes of data. This is used to measure a player’s performance, identify areas of concern, and even predict if a player is going to get injured. “The real-time analysis of player and team performance based on SAP HANA will help not only to elevate levels of play, but provide foresight to customize trainings accordingly and in a way not previously possible,” explains SAP’s Gerd Oswald.
F1 team Williams generates around 120GB of data during a race. Hundreds of sensors send back information on everything from tyre pressure to fuel efficiency. Technology company Avanade delivers this so that engineers can make strategic decisions mid-race. If a driver has a bad start, or gets held up in the pitlane, predictive analytics can inform the team whether to make tactical changes.
“Data is so good for drivers that the pit crew could literally coach a driver to the most optimal racing conditions based on a live data feed of everything imaginable,” said Tony Jardine, Sky Sports Formula one analyst.
This year’s Tour de France saw big data and analytics used for the first time.
Sensors were fitted under riders’ saddles to give fans real-time performance insight (from 42,000 geospatial points and 75 million GPS readings). All 22 teams participated, with fans able to analyse cyclists’ race strategies as they hit speeds in excess of 100 kilometres an hour.
A live tracking website was built to handle up to 17 million viewers, 2,000 page requests and 350,000,000 CPU cycles per second.
At the start of this season the NBA installed SportVU cameras in its 29 arenas. These send algorithmic data to coaches, helping them analyse how a player will perform in a certain scenario, or when faced with a particular player.
“In the past you would need a human to log in this information themselves,” explains Ryan Warkins of Stats, the company which owns the SportVU cameras.
“You can find out efficiencies of teams and players and it’s all based on storing that raw information at 25 frames per second.”
The advantages of data-driven sport are clear. As more teams, coaches and individuals harness its potential, it’s clear that data centres will play a pivotal part.
And who knows, alongside those end-of-season Player of the Year and Manager of the Year awards, there may soon be a trophy for Data Scientist of the Year.