Football is a booming business. You only have to look at the latest TV deal, worth £5.14bn to see how hot it really is. Football clubs are seeing an unprecedented influx of money and are trying to ensure they stay at the top, to make sure they turn this windfall into a regular income. This has paved the way for Big Data & Analytics to play its part in aiding success on and off the field, much like its impact in other business sectors and industries.
Stadiums across the UK are festooned with cameras to track players’ movements at all times, even when not in possession of the ball. 1.4million data sets are collected each game, that’s 10 data points each second for every player. It’s this that gives managers, performance analysts and other backroom staff the chance to see the different layers on the on-field action that would be impossible to accurately track with the human eye. Player’s are also heavily monitored off pitch, with GPS trackers, acceleration sensors, heart rate monitors, diet and sleep tracking to deliver metrics to help optimise training and game tactics. Not only this, but a statistical approach to training can help pre-empt injury, forecast injury recovery times and avoid overtraining. Their every move is creating large data sets and we’re now seeing that they can be used for much more beyond the pitch.
We’ve seen performance monitoring at work in football for a while, but what else can these statistics do in shaping the football industry? Teams are now utilizing this data to scout and sign players to best prepare their tactical advantage for the season ahead.
Brought into the limelight by Michael Lewis’ Moneyball and the following 2011 movie, Data Science in sports games has been on the rise. For sports, such as Baseball, predictive analytics work well as its easier to apply metrics when players are in fixed positions with limited moves etc – but how do we go about incorporating such tactics in a fluid game like football? It’s much harder to apply statistical analysis to shape the outcome of a game with more free and sporadic player movements.
By hiring Data Scientists to work alongside the traditional scouts, teams are able to seek out the next big thing by implementing more analytical approaches. As competition for players becomes more and more fierce, there is a huge opportunity to spot quality in the transfer market that others might not see beyond the big name on the back of the shirt.
Clubs like Liverpool have employed analysts like Michael Edwards who, leading a team of analysts, has brought in new methods to find new players. Working with traditional scouts, these data analysts are leading the way in the future of football. A scout may see a footballer on a good run and recommend they be signed, but an analyst can see the entire traceable performance history to make a better judgement on a players’ long term abilities and what they can bring to the club. Edwards has supported in the signing of players who have gone on to success, whilst working for Liverpool and Tottenham Hotspur.
By replacing some part of the scouting process with more data based approaches, clubs are also able to somewhat justify their massive spends. Real Madrid’s purchase of Gareth Bale for £85million was backed by not only his performance data, but his forecast profit of £41million over 6 years through football shirt sales alone.
Companies such as Prozone are leading the way of performance analysis in sports and are used by most Premier League clubs. It’s been a couple of years since their integration with Football Manager from Sports Interactive’s worldwide database holding player information into their Recruiter tool used by most Premier League teams.
This made biographical and contractual data for over 80,000 players available to clubs. Information on players’ dribbling, goal attempts, aggression, ball possession, saves and distance travelled on pitch and much more is available with this tool, which allows clubs to make the most informed decisions during transfer season. It’s also giving players from smaller clubs the chance to get spotted and signed by bigger clubs.
But, data analysis can only go so far in helping decision makers make the right moves during transfer season. Finding not only new talent, but the right talent still requires a human element. Is this player a good fit for the team? Do they have the right mentality? How will the player develop and grow? There are still some gaps and a long way to go, but the early signs are that the marriage of Big Data & Analytics and football is a match made in heaven.
Co-Founder, Big Cloud