As with just about any industry, data analytics has bled into every aspect of professional soccer. From injury prevention to team selection, there isn't a part of soccer management that hasn't been affected by the explosion in data collection and the insight drawn from it, and never is the impact more greatly felt than during the transfer window. Not only are players more closely judged on their statistics when there is speculation around their futures, but add transfer fees into the equation and footballers are commoditized and analyzed more than ever when the window is open. As fees become astronomical and seemingly free of any grounding, though, there is an argument to be made that analysis of a player's actual output, rather than imagined worth, is more important now than ever before.
This transfer window has already seen what would have previously been considered ludicrous sums spent. In an attempt to shore up an often shaky defence, Liverpool spent £75 million on highly rated Southampton center-back Virgil van Dijk, a move that raised eyebrows given the club's comparatively reasonable transfer policy. Then again, when you consider the money raised from the sale of Philippe Coutinho to Barcelona - a reported £142 million - the transfer is little more than a reinvestment of funds. On top of this, clubs have ways of mitigating the impact of expensive transfers on their finances, amortizing payments over a number of years in some cases. Similarly, the percentage of a club's revenue traditionally spent on a 'large' transfer in the modern era is less than 20%, even in the case of Paul Pogba's £89 million move to Manchester United. Such is the revenue that global clubs are pulling in that seemingly ludicrous transfer fees aren't as damaging as they appear.
Tim Bridge, a senior manager at Deloitte’s sports business group, explains: 'Clubs do not account for a transfer all upfront, instead spreading the fee across the life of the contract, so a £30m to £40m revenue uplift in one year translates to £200m in transfer spend across a five-year period. The scale of the numbers attracts attention and gets people asking the question about whether this is sustainable. But on a proportionate level, we are seeing it remain at a similar level.'
So, if transfer fees have become all but irrelevant and circumstantial at the top level, data becomes a more important tool in judging the objective value of a player. Soccer's fluidity would make a Moneyball-style analytics strategy almost impossible, but a player's statistics can and should inform decision making.
Coaches that allow team selection and player acquisition to be influenced by data will gain a competitive advantage, even if it means relinquishing some creative control. 'Football is hard,' Campbell says. 'But saying that it is not suited is missing the point. In my mind, hard is good. Because there is the most opportunity to gain a competitive advantage if you use it right. This is why poker players make more money than chess players. Because in chess the best player almost always wins, and in poker there’s variance and luck, and that’s a good thing.'
This is why data in football thrives as it does - where there is a seemingly chaotic spectacle, the numbers underpinning it are inherently interesting. The data, in the case of soccer, can tell coaches who is over performing or underperforming, who is likely to get injured, which tactical decisions have the best chances of success against any given team. All of this information can be drawn from either the chaotic fluidity of the match itself or from the players' biometric data, and this is why clubs are piling resources into it.
The way soccer is broadcasted this month will also be heavily influenced by the explosion of data. Pundits will cite goal conversion rates, pass completion rates, number of successful dribbles, and seemingly innumerable other data points to form their analysis of any given game or player. Where fans would once be presented with only the possession stats and shots on goal from either team, for example, they can now delve into how many 'quality' chances a playmaker created.
The development is perhaps best exemplified by the 'Expected goals' statistic. Put simply, each goalscoring chance is assessed and assigned a 'quality' value (xG), with 1 being the maximum. A very good chance will represent something like 0.91xG, with a more missable opportunity coming in at something like 0.34xG. The running total of these figures is now included among the more traditional match facts on most networks, much to the disdain of the less data-conscious pundits, who seem to view it as unnecessarily analytical and arbitrary.
Scouting teams can even now assess whether a player's goalscoring record comes from easy (high quality) chances or from more difficult positions, from which they can build a picture of how reliant a striker is on their playmakers, for example. Ultimately, expected goals gives us a much better picture - based on a number of factors - of how many chances a player should be scoring, rather than simply the percentage of ill-defined chances they convert.
Harry Kane, for example, scored 29 goals in last season's Premier League campaign, despite only tallying up 18.59 expected goals. This return is part of the reason the player is so highly rated; his ability to score chances against the odds is invaluable. Kane's transfer fee, if he were to leave Tottenham, would likely be sensational. Assessing his worth to a club like Real Madrid, on the other hand, would take deep analysis of his impact and of his ability to add value to the team in terms of the numbers. Players will no longer see their transfer fee as intrinsically linked to their worth to a club, rather they will work to improve metrics as specific as pass accuracy in the final third to improve themselves as prospects.