It takes Jesper Gudde two clicks.
For the purposes of our chat, we’re a Premier League club with a middling budget and a hole in our forward line.
The challenge is to find a Harry Kane clone playing in the lower leagues who fits our modest budget – and a quick search brings up 27 names. Another toggle – narrowing the search even further by limiting it to finishing ability – hones it down to three contenders (Teemu Pukki, Dom Solanke and Ivan Toney).
When Toney comes out top, you can then interrogate his profile to determine his recent development, playing style, comparisons to other Championship top performers and even how proficient he is from right wing crosses delivered from a certain area of the field.
But that’s just the start: you can then match him to other clubs in European football, telling you whether he would suit their preferred playing style and even drill into exactly what his status (first team regular, back up or potential first teamer) would be if he was recruited by a rival club.
Welcome to SciSports – the “real life Football Manager” influencing multi-million pound transfers across European football.
The Dutch company was launched in 2013 by self-confessed Football Manager geek Giels Brouwer and has made waves as one of the fastest-growing of the companies being harnessed by the increasing number of clubs joining the data revolution in professional sport.
Through a whip smart artificial intelligence model that was honed in conjunction with advice from coaches, sporting directors, agents and players, they’ve built a system that squeezes the pips of the plethora of data out there to create a clear picture for agents, managers and Sporting Directors.
“In Holland we’d say it was like a toy and it’s fun to play with – but there’s serious money at play here,” Gudde – Manager Football – for Sci Sports explains over an hour chat from Rotterdam.
Football’s data arms race has become even more fiercely fought in the COVID era, when getting eyeballs on players has become more difficult. Platforms like Wyscout and Hudl Instat offer video analysis, allowing scouts to watch matches on the other side of the globe while a growing number of companies can provide insight into that footage.
But shrinking recruitment budgets have created a challenge for clubs: how to buy smart in an era where so much is unknown?
From their base in Holland, SciSports see their challenge as “translating” the data into the simple, easy-to-understand chunks.
“Right now, everyone is hyping data scouting: “We need to do data scouting!” No, what you need if you’re a Director of Football is good results and to make a profit in the transfer market and good decisions,” Gudde says.
“Data helps but it’s the same as live scouting or video scouting – it helps you in making decisions not only based on gut. It’s on facts that you can’t manipulate.
“We know teams will never buy a player based just on his SciSports profile – we wouldn’t expect you to and we wouldn’t want you to. You have to do the due diligence on someone – what is his mentality, what is his personal position? There are all sorts of factors.
“But we know the data works and helps because we have evidence from our work with these clubs.”
In a season when Barnsley, Norwich and Brentford were promoted using data-driven recruitment, football’s best kept secret is no longer under wraps. Expected goals are commonly used now, so new ways of measuring ability that can be applied across leagues are being developed.
SciSports’ SciSkill index crunches all of the numbers to create what they see as a definitive ranking of the best players and teams in the world. What makes it unique is that clubs can import potential signings into their profile and see exactly what difference they would make to their results.
“There’s a big difference between stats and data insights,” he explains.
“What data providers measure are things like passing accuracy, shot accuracy, length of passes. So a player may have a passing accuracy of 80% but what we are looking at is how much a player contributes with his passing because if I have an 80% passing accuracy but play everything backwards, it’s not what a team might want.
“We get the raw data provided elsewhere and try to create extra intelligence on top of the data to give more context to the numbers.
“We get requests from clubs – maybe they’re looking for a right back to play more as a wing back. We create a model which provides a player and then we validate it with the market by asking coaches, assistant managers, sporting directors whether it is providing the right sort of answers.”
Artificial intelligence has allowed seriously smart software to cross over into sport.
“We use a machine learning model,” Gudde says. “So, for example, maybe 1,000,000 times a certain type of pass was played. It looks at what the situation was that occurred afterwards – did a team create a chance, advance up the pitch, was it blocked? Then based on this it creates the model of what a certain move – a pass in a part of the field, for example – contributes.
“We are the translator of this information with the platform. That’s how our name came about – you have science and sports. We are the bridge between the two.
“Data is excellent but you need to be able to present it in a way that the coach can understand it and use it. That is a very undervalued thing in football these days – clubs are appointing data scientists and all sorts of wizards and gurus but they’re creating islands it is extremely important that no islands are being created within the club and that everyone speaks the same football language.
“The presentation of the data is 50% or more of the importance. If you have a fantastic story that you’ve discovered with data but if you can’t communicate it, it’s useless.”
At the moment the system is being used by club like Leeds United, Norwich, Sheffield United and Burnley – but talks are ongoing with other Premier League clubs, including Newcastle.
While the elite are interested, the price of the software means lower league clubs are looking too – with upwardly mobile League Two side Forest Green Rovers another client.
It’s not just clubs – individual players and agents are starting to act as their own Directors of Football. When Memphis Depay left Manchester United, he used SciSports to pick his next club.
“We developed a ‘career advice’ mode for agents as well where players and their representatives can find the right next move,” Gudde explains.
“On the platform there are detailed data-driven player profiles and there is also a detailed section on team playing style – so our scientists have found a way to match playing style of an individual player with team style of a club.
“We did this for Memphis Depay when he left Manchester United, the next move was a big move for him personally. It gave him a list of six clubs and Lyon was one of them. We feel like we had an added value in that transfer – of course, there’s got to be a personal side to it as well which we don’t factor in. But we feel that we provided him with important information which influenced his choice.
“We have worked with Matthijs De Ligt and two players in the Premier League in the same way.”
Ahead of a difficult transfer window – Gudde thinks clubs “will spend in the end”, especially if fans are back in grounds – the company believes a sea change has occurred in the way deals are researched.
“Teams are rethinking their processes and using their data more and more. That’s one benefit but on the other hand they don’t have a lot of incoming so budgets are tight. That’s a difficult one. I think in terms of the money being spent, it’s also an opportunistic world,” he says.
“If you stop spending, you lost the option to make money as well. The world has changed and people have to adapt.”