How big data in sports is changing the game
In today's world of elite sports, Big Data can help predict matches - but that's not all it can do.
Remember the famous crumpled sheet of paper lodged in the sock of German goalkeeper Jens Lehmann that gave away the likely habits of the penalty takers – “Riquelme left high, Heinze left low, Ayala long wait, long run right” – which helped him save penalties against Argentina in the 2006 World Cup quarter-final shootout?
In the world of elite sports, there is a huge competitive advantage to knowing your opponent’s next step, which is precisely why it is hardly a surprise that the sporting universe is experiencing an explosion in the use of analytics.
Imagine if teams had access to a real-time heads-up database that could accurately predict how AB De Villiers would react to a particular ball bowled by a particular bowler in a particular situation? Given the venue, pitch, weather conditions, crowd, time of play and the composition of the opposition attack, what should be the batting order and target score?
Coaches have always sought to spot patterns and exploit weaknesses in their opponents’ game during planning. And ever since the story of “Moneyball: The Art of Winning an Unfair Game” alerted the sports world to the power of data science, sports culture has undergone a paradigm shift.
The world of sport generates far, far more data today than could have been imagined just a few years ago. Today, teams are going beyond the field of play and searching for innovative ways to use the unprecedented mountains of data they have been stockpiling. While the trend of using analytics in sports to improve on-field performance is continuing to grow, it's certainly not stopping there.
What started with the manual counting of assists, possession percentages, shots on goal or the distances clocked by players has now evolved into an industry where analytics tools can be leveraged for a glut of decision making instances - from improving fan experiences to even foreseeing injuries.
Nowadays, EPL and La Liga clubs leverage data from Opta, Prozone, Whoscored and others to formulate strategies. Sam Allardyce, during his time at West Ham, would keep track of his players’ on field and off field activities and tailor trainings and proper conditioning for individual players.
Much of the value in sports analytics will come from being able to marry information from variables like stress, fatigue, sleep, intensity and nutrition with other bio-metric data such as heart rate, speed and acceleration from matches and even gym sessions. This kind of real-time data will not only help coaches and physiotherapists fine-tune training regimes and optimize recovery/rest times but also uncover hidden trends that cause injury.
If a player practises too hard, he runs the risk of incurring an injury, and can be expected to slacken up on match day. It can even feed into match day decisions, as to when to drop or substitute a player, and help coaches to adapt formations and tactics based on data gathered about opponents’ teams. Also, by monitoring data on hydration levels, pulse rate or hits to the head, sports can be made a lot safer.
Taking a cue from e-commerce and other consumer centric firms, teams can adopt analytics to play an integral role in off the field wins and advancing fans’ experiences. Analyzing social media data and profiling fans helps teams hear the voice of their “customers”.
By using variables such as game attendance, purchase of team merchandise and combining them with knowledge of fan preferences and behaviour, teams can predict the needs and wants of fans and deliver their targeted content with laser precision. They can even scour other profiles who share similar characteristics with existing fans and target marketing campaigns to propagate them.
While there is no doubting the fact that data can give teams a truly competitive edge, teams should still leave room for intuition in the decision-making process. After all, getting all the information right isn’t a piece of cake. Unlike in some other industries, there are factors in the sports business that can still be difficult to quantify in an actionable way. The biggest challenge is to build machine-learning models that aren’t biased by the historical data they are triangulated by.
All said and done, data science in sports is certainly here to stay, and with companies like Adidas investing heavily in wearable technology, more data will be available in the hands of everyone, which only means that the viewing experience is only going to get better. Eventually it is the fans who will be the biggest winners.