From live televised games to lunchtime kick-offs, from fixtures on Sundays to goal-line technology, and from directors of football to false nines; English football has been a reluctant and nervous embracer of innovation over the years - and yet it has always got there in the end.
So will the arrival of innovative Danish club FC Midtylland at Southampton in a Europa League tie this evening (kick-off 8pm) be the initial spark that inspires a new way of doing things?
Midtylland, one of the youngest clubs in European football, having only been formed in 1999, are punching way above their weight thanks to a data-driven revolution that has seen them crowned Danish champions for the first time and begin slowly to push the conservative boundaries that restrict the way the majority of people think about football.
Inspired by the vision of majority shareholder Matthew Benham, a professional sports gambler who also owns Championship side Brentford (between 50 and 100 Bees fans will join Danish supporters in the away end tonight), they have vowed to disrupt the market with a totally different approach to recruitment and training.
Benham’s use of mathematical modelling, algorithms and data mining has made him millions in the business world and now the test is whether the same principles can be applied in football, using stats in a new way to unearth under-valued players, aid team selection and inspire tactics which can help a small club compete against bigger rivals.
“We can’t outspend our competitors,” said FC Midtylland chairman Rasmus Ankersen, a former player and coach at the club who also holds the role of Co-director of Football at Brentford. “So we have to out-think them. And we think careful analysis of data on leagues, teams, and players can give us an edge.”
Inevitably, the process has been labelled a ‘Moneyball’ revolution, a reference to the Michael Lewis book and film starring Brad Pitt which told the story of American baseball team the Oakland Athletics who used a similar model of statistical analysis or ‘sabermetrics’ to great effect. But Midtylland’s system is still forward-looking.
The club, based in Herning and Ikast in the western part of Jutland, is seriously challenging long-held and accepted ways of doing things in football by gradually replacing traditional scouting with a mix of statistics and video analysis that seeks to unearth new talent and help current players improve.
Scouts, for instance, are no longer encouraged to focus on watching matches live to ‘discover’ new players. The statistics identify who has the talent - and the scout’s job is not to decide if they are good enough (the figures already show they are) but simply to provide feedback on whether the player has the psychological and technical stability to make that talent count.
An assessment, the theory goes, more likely to be accurate after watching 38 videos of a player’s performances over a season than by sitting in the stands for a single match in the rain. Of course, the use of data in football is not new in itself. Cameras film every player in every Premier League game, for instance, and reams of data is produced on their performance. But mining, understanding and utilising that data is a skill – one that requires serious investment.
This is where Benham’s backroom team of data analysts comes in. His model effectively ranks the performance of every team and every player in Europe as if they played in one giant league, allowing Brentford and Midtylland to spot players who are under-valued. Midtylland midfielder Tim Sparv, a Finnish international who ironically began his career in the youth team at Southampton alongside Theo Walcott, was found this way.
Benham’s data showed his team, German Second Division side Greuther Furth, were performing well enough to play in the Premier League despite their status – and Sparv was integral.
“The data is not perfect but I think it’s less imperfect than the human judgement,” said Ankersen. “You’ve got to know where human judgement has a role to play and you’ve got to know what part of the process the data has a role to play in. It’s a lot about understanding what stats can be used for and where the weaknesses with stats are.
“There are things where the eye can spot something the stats can’t and there are other matters where stats can reveal something the eye and the ear can’t.”
Several of Midtylland’s title-winning squad were discovered using the club’s data model and, allied to a heavy focus on practising set-pieces - statistics showed the importance of success from dead-ball situations had been undervalued by traditional coaches – it has brought success.
Both Midtylland and Brentford employ full-time set-piece specialists, and last season the Danes scored 63 goals in 34 games, including 1.05 goals a game from dead-ball situations. So Southampton know they will have to concentrate at St Mary’s tonight.
Interestingly, match preparation also includes input from a ‘kicking coach’ who encourages players to practice movements over and over again, much like honing a golf swing. During the game, statistics play a part, too. All matches are analysed in real time by backroom statisticians who pass them on to the coach to utilise if he wishes.
Other innovations - there was talk of a ‘sleep coach’ being employed at Brentford - are likely to emerge as both clubs earn greater buy-in to from fans and board members for doing things differently. So this may well be just the tip of the iceberg.
Ankersen said: “It’s still early days but we have tested a lot of things that seem to have potential - some have worked, some haven’t. But you have to believe in the model.” At Brentford, manager Mark Warburton wasn’t quite able to do that; baulking at losing so much control over signings and tactics to outside influences. But the club’s ruthless decision to announce he would leave at the end of the season as a result – when the Bees were still in play-off contention – shows how determined they are to make it work.
If FC Midtylland can win at St Mary’s tonight, there will be plenty of other teams taking notice, too.