The “Moneyball Revolution”
How numbers are turning football into an exact science

With the “Moneyball Revolution” well and truly underway in terms of revolutionising sports such as baseball and American football, football is now next in line. Football clubs have recently been actively looking for means to increase their advantage over their competitors through the extensive use of data which will result in on-field improvements. Going beyond finding data and analysing it, the “Moneyball Revolution” in football presents football clubs with ways to implement this raw data into their player-purchasing and on-field strategic approaches.


The idea that “moneyball” is the answer to any football club’s deep-rooted issues is a massive misconception that must be discounted before we start delving into this subject. The way in which the moneyball strategy can help a football club or a manager achieve better results is far more complicated than simply receiving data and implementing it. It is also important to note that immediate results are not to be expected — this approach mainly offers solutions for the long term; for reasons which we will discuss below.

We have to remember that when moneyball was famously implemented in baseball, it was done so in a game which is, historically, extensively data-driven — thus explaining why it had such a visible impact (Beane’s Oakland A’s reached the MLB playoffs from 2000 to 2003, after Beane’s appointment in 1997). Beane’s on-field success ultimately came from his — and his team’s — ability to identify and field a competitive team on a tight budget. This is where his philosophy applies to football.

In an age where the football industry is proverbially drowning in money due to multi-million TV and sponsorship/advertising deals, ‘smaller’ top-tier clubs are finding it hard to compete with clubs who have higher ressources. Working with a budget of $30mil versus your competitor’s $200mil is an obvious hurdle — one which more and more clubs are trying to tackle by adopting the moneyball approach.

So, how can football clubs operate within their financial means by using this method? The first obvious response is to identify ‘quality’ football players at a lower price — relatively unknown football players who meet a manager’s quality requirements, and who can be purchased at a low price.

When scouting players, there are always basic factors to take into consideration such as:

  • Goals scored
  • Assists provided
  • Clean sheets (defenders and goalkeepers)
  • Appearances
  • Distance covered per match

This gives you an overall view of the player and his performance during a game. Then you look at more in-depth data, including:

  • Passes completed (%)
  • Interceptions
  • Shots on target
  • Clearances (defenders and goalkeepers)
  • Saves (Goalkeeper)
  • Sprint speed
  • Accelerations / Decelerations
  • Tackles completed
  • Duels won
  • Aerial duels won
  • Recoveries
  • Shot accuracy (%)
  • Average pass length
  • Average defensive actions
  • Defensive errors
  • Total bookings (Yellow/Red)
  • Total chances created
  • Offside(s)

But of course, it doesn’t stop there. With the help of specific software, you are able to determine a player’s fitness, movement, and physical performance during a game. Does your centre forward feel more comfortable taking an extra touch before shooting or is he more inclined to take his chance on the first touch? Does your goalkeeper position himself before jumping for a save or does he read the game well enough to be in a favourable spot the majority of the time? How many times does your center back get bettered by an opposition player dribbling, and how often does he recover into position in those situations?

As one can imagine, there are dozens upon dozens of pieces of data which contribute towards a player’s overall efficiency on the pitch.

As mentioned above, football mainly benefits from moneyball in the long term because ultimately, a manager will be able to improve certain aspects of the player once he joins the club (strength-building to win duels, drill training for recoveries, etc) but it’s a player’s raw ability (overall shot accuracy, ability to read the game) that gives the manager a fair view of the player. Focus, drive, work rate and sheer heart are also important factors which might push a transfer over the line, but those are aspects which cannot be identified in statistical data.

Coupled with the fact that the use of raw data can be used to analyse a player’s training and in-game performances as well as fitness levels, the moneyball model can only successfully be applied over a longer period in order to identify patterns and inconsistencies — resulting in a lack of immediate results.

Consider the collection of data and its analysis into a coherent narrative as an ongoing process which not only cannot be rushed, but furthermore one which will only yield results over a certain timeframe due to the very nature of its mechanisms.

Because football is a much more free-flowing game than baseball, where the game mainly consists of 1v1 actions, there can be up to 10 thousand on-ball actions during a game of football. The multitude of data collected then has to be interpreted in a comprehensive narrative which can ultimately be used for the purchasing of a player. The manager ultimately decides the most important factors for a certain player in a specific position, and the filtering process goes something like this :

“I’m looking for a centre-back who wins aerial duels over 75% of the time, plays the ball out to the forwards on the counter-attack at least 90% of the time, has a clean sheet record of at least 4 out of 10 games, and commits less than 2 defensive errors per game”. This criteria is then applied to a list of players, from which a shortlist is created.

Then, you can go more in-depth to identify a player’s strong or weak points, such as finding data of a player’s right-footed passes of more than 20 yards, towards the left wing in the final-third of the pitch. You are able to create a list of 10 or 20 of these instances and determine whether the player can be used in a team’s system where he would be required to complete such actions. This process is then repeated a multitude of times for each potential recruit, and interpreted by the manager and his team of data researchers.

A success story of the moneyball concept applied to football is Matthew Benham’s involvement in  FC Midtjylland and Brentford FC. Although some would argue that the moneyball model hardly carries any weight in top-flight football, Benham’s acquisition of Midtjylland in 2014 was shortly followed by their first league championship a year later, since their creation in 1999. Moreover, Benham’s acquisition of Brentford in 2012 was followed with promotion to England’s second-tier in 2014, after a 21 year wait.

Although both clubs are not enjoying consistent success, Benham’s involvement in Midtjylland and Brentford has pushed both clubs to adopt different approaches, which resulted in a league title and promotion, respectively. Both clubs then have to adapt once more when they enter the top-flight and have to compete with bigger teams, though this simply means readjusting their approach and going from strength to strength. Midtjylland’s Europa League funds and Brentford’s promotion funds will allow both clubs to purchase more expensive players and improve their back-room activities.

Sport science has certainly come a long way in the last 20 years, but certain factors can simply not be analysed in a statistical narrative — or at least not as efficiently as a seasoned scout’s eye. As much as the moneyball philosophy has evidently benefited many football clubs, it remains to be seen whether or not its model is sustainable and above all, competitive.