Expected Goals

1. Football's data revolution: From skepticism to mainstream acceptance

"Football had flirted with data as far back as the 1950s, but it was only in the late 1990s – a few years before Bill James and his acolytes and apostles started to infiltrate and then influence baseball's thinking – that it started to gain any widespread traction."

Early resistance: Football, unlike baseball, was long considered too fluid and dynamic to be quantified. The sport's traditionalists believed that intangibles like passion and heart were more important than numbers. However, several factors contributed to the gradual acceptance of data in football:

• Increasing professionalism and globalization of the sport
• Growing financial stakes, incentivizing the search for competitive edges
• Influence of sports science and academic approaches
• Technological advancements enabling better data collection and analysis
• Changing fan demographics, with more analytically-minded supporters

Milestones in acceptance:

• ProZone and Opta introducing data collection services in the late 1990s
• Publication of "Moneyball" in 2003, highlighting data's potential in sports
• Expected Goals (xG) appearing on Match of the Day in 2017
• Top clubs like Liverpool FC integrating data into their core operations

2. The pioneers: Early adopters who paved the way for analytics in football

"Bolton were derided – not entirely fairly but not without reason – as a long-ball team at a time when English football was in thrall to a more exotic aesthetic, imported from continental Europe and inflected by South America."

Sam Allardyce at Bolton: Despite his reputation as a traditional, long-ball manager, Allardyce was one of the first to embrace data analytics in English football. His approach included:

• Hiring analysts and sports scientists
• Using ProZone data to inform tactical decisions
• Focusing on set-pieces and player recruitment based on data
• Developing the "fantastic four" principles based on statistical analysis

Other early adopters:

• Arsène Wenger at Arsenal, recognizing data's potential early on
• Damien Comolli at Tottenham and Liverpool, advocating for data-driven recruitment
• Decision Technology working with Tottenham to provide advanced analytics

These pioneers faced skepticism and ridicule but laid the groundwork for the wider acceptance of data in football.

3. The rise of Expected Goals (xG): A game-changing metric

"Expected Goals did not change the world straightaway. By 2017, when it first appeared on screen, it was not even particularly cutting edge."

Development of xG: Expected Goals evolved from early attempts to quantify shot quality to become a widely accepted measure of team and player performance. Key stages in its development included:

• Charles Reep's early work in the 1950s
• Academic research by Mark Dixon and Stuart Coles in the 1990s
• Opta's Sam Green developing a public xG model in 2012
• Adoption by professional clubs and media outlets

Impact of xG:

• Provides a more nuanced understanding of team and player performance
• Helps identify undervalued players in the transfer market
• Influences tactical decisions and player positioning
• Changing fan and media discussions about football

The widespread adoption of xG represents a significant shift in how football is analyzed and understood, bridging the gap between traditional scouting and advanced analytics.

4. Challenges of implementing data-driven approaches in traditional football culture

"Football holds its traditions tight; those who considered themselves the game's spiritual guardians did not welcome interlopers, with their bright ideas and their new ways of doing things."

Cultural resistance: The integration of data analytics into football faced several obstacles:

• Skepticism from traditional coaches and scouts
• Fear of job displacement among existing staff
• Difficulty in communicating complex ideas to non-technical stakeholders
• Resistance to changing long-held beliefs about how football should be played

Case studies in implementation challenges:

• Chris Anderson at Coventry City struggling to balance data insights with day-to-day operations
• Arsenal's StatDNA acquisition not fully utilized due to internal resistance
• Early attempts at Liverpool FC facing skepticism from managers and fans

Successful implementation often required:

• Buy-in from top management and ownership
• Patience and long-term vision
• Effective communication between analysts and football staff
• Gradual integration of data insights into existing processes

5. The role of gambling firms in advancing football analytics

"Both make their money by using the probabilities determined by those algorithms to spot areas where the open market is either over or undervaluing a team or a player, and passing that advice on to – and placing bets for – high-rolling clients, who can now gamble safe in the knowledge that the odds have been tilted, however slightly, in their favour."

Gambling's influence: Professional gambling firms like Starlizard and SmartOdds played a significant role in advancing football analytics:

• Developed sophisticated models to predict match outcomes and player performance
• Invested heavily in data collection and analysis
• Created metrics similar to Expected Goals before they were widely used in football

Transfer of knowledge to clubs:

• Tony Bloom (Starlizard) applying analytics principles at Brighton & Hove Albion
• Matthew Benham (SmartOdds) implementing data-driven approaches at Brentford and FC Midtjylland

These gambling-rooted approaches helped smaller clubs compete with larger, wealthier teams by identifying undervalued players and tactical advantages.

6. Liverpool FC: A case study in successful data integration

"Over the last decade, Liverpool have been able to do what no other club of their size has tried or dared or managed, and laced the use of data and analytics into their very fabric."

Liverpool's approach: Under Fenway Sports Group ownership, Liverpool FC has become a model for integrating data analytics into all aspects of club operations:

• Recruitment: Using data to identify and evaluate transfer targets
• Tactics: Informing on-field strategies and player positioning
• Player development: Tailoring training and recovery programs
• Business operations: Optimizing ticket pricing and fan engagement

Key figures in Liverpool's data revolution:

• Michael Edwards: Technical Director who championed data-driven approaches
• Ian Graham: Head of Research, developing sophisticated analytics models
• Jürgen Klopp: Manager open to incorporating data insights into decision-making

Results of data integration:

• Successful player acquisitions (e.g., Mohamed Salah, Virgil van Dijk)
• Premier League and Champions League victories
• Sustainable financial model balancing performance and profitability

Liverpool's success demonstrates how data analytics can be effectively integrated into a top football club when combined with strong leadership and traditional football expertise.

7. The future of football: Balancing data with human expertise

"There is a value in the intangible, in the parts of the game that are not reflected in, but may be affected by, the stark reality of the data."

Evolving landscape: As data analytics becomes more widespread in football, the future of the sport will likely involve:

• More sophisticated data collection and analysis techniques
• Increased integration of analytics into all levels of the game
• Growing emphasis on finding new edges and inefficiencies

Balancing act: The most successful clubs and managers will be those who can effectively combine:

• Data-driven insights
• Traditional football knowledge and experience
• Human factors like psychology and team dynamics
• Intuition and "feel" for the game

Potential areas for future development:

• Advanced metrics beyond Expected Goals
• Real-time analytics informing in-game decisions
• Personalized training and recovery programs based on individual player data
• AI and machine learning applications in tactics and recruitment

The future of football will not be solely determined by data, but by those who can best interpret and apply analytical insights within the context of the sport's human elements.

Last updated: August 15, 2024