Player acquisition in sports is a complicated thing. In theory, a good player in one league or at one team should be a good player anywhere. Talent, work ethic and adaptability are requisites to making it in top level sport, but there are innumerable other factors at play that can see athletes wildly over or under perform following a move.
There are, of course, some transfers that justify themselves. Real Madrid’s Galactico model generally looks to avoid these external factors, with varying degrees of success, bringing in hugely successful players and accommodating them at all costs. Even for a club as wealthy as Madrid, though, the integration of new players is far from perfect. Imbalance and clashes of ego are relatively common, yet the Galactico system has continued to yield trophies. Other top level teams can operate a similar system, but for smaller clubs it's more important that the player be a good fit.
To find one, scouting teams need the full picture to pick out players. Every sports fan will be able to rattle off a list of missteps made by their clubs on the transfer market, each with their own notable flops and calamitous debuts. It’s often hard to pin down exactly why a move didn’t go according to plan. It could be the lack of familiarity with a new formation, personal issues around relocating to a new country, inability to handle the intensity of a new league, a lack of discipline in training or, simply, a language barrier.
And, onto this complex and nebulous process, enter data. Billy Beane’s Moneyball is an inspirational story of looking at a game differently, picking out the numbers others ignore and abandoning the truisms of a sport in favour of the cold metrics of success. The Oakland As performed way above expectations not because they signed the world’s best undiscovered baseball talent, but because they understood that gut feeling and accepted best practices can be qualified or disqualified by the data.
Traditional scouting doesn’t give teams enough to go on. Attend four different fixtures and you’ll get four different performances. Watch six hours of matchday video footage and you’ll miss the player’s commitment on the training ground. Even if you watched an entire season’s worth of footage you’d be unable to see how a player reacts when the ball is lost down field, or their positioning when they’re not in the frame. It’s because of this that any scouting department worth their salt in today’s game will back their hunches up with numbers.
Experienced scout Rob Mackenzie lauded analytics in an interview with Sky Sports: ‘Data is neutral, reliable and it allows you to assess a significant number of players in a time-efficient manner. It provides a platform for you to identify players and benchmark expected performance levels. It also allows you to compare similar profiles, help establish what else is available in the market and therefore work out who the most valuable players are to your club.’
Mackenzie then went on to explain the fundamental point to remember when inputting an analytics program geared for talent acquisition. ‘Having access to data and career biographies of players in competitions across the world is certainly empowering,’ he says. ‘But as with any resource available to you, it is important to acknowledge the level of insight that it can and cannot provide and place that into much-needed perspective when making decisions. Anything in isolation tends to be insufficient.’ The former Tottenham Hotspur scout hits the nail on the head here. Data is just the same as any new tool - a useful addition to bolster already established scouting practices.
Because of the myriad potential factors at play, no one is suggesting that teams go out and buy players based solely on their passing accuracy statistics for the past five seasons, for example. Rather it should be used to streamline the process. Data can give scouting teams a list of potential candidates based on a minimum expectation in certain metrics. It can help make the final decision once a list of players with the suitable mental and physical attributes have been isolated. But it can’t, crucially, make the decisions alone.
It’s also important to note that some sports lend themselves to data more than others. Sports like baseball, for example, are more about isolated events and the data produced is far less interdependent. The Oakland As could use on base percentage as a key metric because you can say with a fair degree of certainty that the batter will be able to keep it up in a new team. Strip away a soccer goalkeeper’s band of defenders and place them in a new environment, though, and their save percentage could be anyone’s guess. So dependency on data should vary from player to player, from sport to sport, and from team to team.
And yet, nothing beats the human eye. Data use’s astronomic growth in sports will have significant impacts on the industry, but it won’t make scouts redundant. On the contrary, it’ll only empower scouts more to make better informed decisions when it really matters for major teams. The only fear is that data will rid global sports of comical and often legendary flops, though it’s difficult to see manager’s losing too much sleep over that.