Embracing AI Beyond the Transfer Window

The Association of Sporting Directors are proud to partner with Zone7 who continue to help clubs across the industry achieve their goals. Their latest piece looks at AI and how it has become ingrained within the industry.

Artificial Intelligence (AI) is no longer just a buzzword in football – it’s becoming part of the fabric of the sport.

To date, the primary purpose of investments made into data analysis and AI within clubs has focused on talent identification (ID) and recruitment.

Player transfers are becoming bigger business than ever, with fees regularly exceeding the $50m mark, clubs are always looking to unearth talent that will improve their performance on the pitch and, in turn, generate increased revenue via prize money, marketing, or commercial means. 

In fact, in 2022, clubs globally spent a staggering $6.5 billion on transfer fees, 33.5% higher compared to 2021 and a substantial increase from the $3.9 billion in 2013, as cited in FIFA’s Global Transfer Report 2022.

The direction of travel on player transfers is clear; to support these sizable investments, clubs have embraced data-driven solutions for a more informed, moneyball style recruitment process. 

In November, reports emerged that Manchester United had begun the process of recruiting data scientists, machine learning (ML) scientists, analysts, and engineers as they look to become “dominant” in data science.

Sports data companies have also emerged to quantify previously qualitative metrics such as playing styles and athleticism, providing valuable insights on potential recruits.

When examining these companies, the evolution of performance data within talent ID and recruitment is evident.

Football data began as almost entirely ‘event data’. This had limited impact as it only detailed what happened to the ball (how many shots, passes etc.), but didn’t provide a full picture of what all 22 players were doing at any given moment. 

Tracking data has revolutionized the field. Cameras around the stadium now record the movements and actions of every player on the pitch, providing a much more holistic data picture of each match.

This abundance of high-quality data has since enabled solutions to graduate from data analysis conducted manually, to systems controlled by AI.

Today, it would be difficult to find a single top tier club that isn’t investing in these tools and building out their data analytics and AI capabilities. Those that aren’t investing are rapidly being left behind. 

Millions of dollars are spent on identifying the right players to fit into the right systems within a club.

Organizations clearly understand the ROI that data analytics tools and AI provide in assisting the human operators within a club to make data informed decisions on who to buy, or not to buy – a decision that is ultimately worth tens if not hundreds of millions of dollars.

But what happens once a multi-million-dollar player is signed and integrated into a new club remains less data informed, says Tal Brown (CEO, Zone7).

“The main tool in your arsenal when looking to bring in the right players and to justify the transfer fee is through using data and AI. That is already happening at almost every top club. They’re buying the data and deploying AI,” Brown explained.

However, there is an imbalance in the investment and innovation directed towards new player transfers, versus the maintenance and development of the squad already within a club.

Brown added: “The same playbook is also available for looking after the players that are within their existing squad. It’s just a matter of taking that leap.” 

Currently, clubs collect a significant amount of performance and medical data on players such as GPS, heart rate, and force-plates, but there are few clubs which have deployed the resources to get these data points to speak to one another for positive impact.

“Finding meaningful insights across multiple datasets from multiple athletes on a daily basis is almost humanly impossible to achieve with manual analysis processes, even with well-organized data,” said Rich Buchanan (Performance Director, Zone7). 

“More sophisticated analytics tools are required.” 

Without these more sophisticated analytics tools making sense of all the relevant and available performance datasets, clubs rely on limited data and basic analysis processes to aid them in their quest for greater, more sustainable player availability.

Given the high stakes directly related to keeping players fit, this is an area where AI can have an overwhelmingly positive impact. 

The National Football League (NFL) has pursued innovative, data-driven programs to improve player safety and assist players in achieving their full potential. Since 2019, the NFL has partnered with Amazon Web Services (AWS) to implement a data-driven strategy using cloud-based analytics, ML, and AI, to gain a deeper understanding of every aspect of the game.

Alternatively, clubs such as SC Napoli, Liverpool FC, Rangers FC, Leeds United FC, Los Angeles FC and others have adopted Zone7, to rise to the challenge of making sense of performance and medical data for greater player availability.

Zone7 is an AI-powered platform which unlocks disparate performance data for clubs by recognising and analyzing patterns in the performance and medical data that they are collecting and comparing it against hundreds of millions of hours of anonymised training, match and injury data. The result is an ability to forecast injury risk for players and offer actionable interventions that have ultimately increased player availability by upwards of 40%.

This makes sense for a number of reasons.

• Improved player availability not only translates to better on-field performance but also offers significant financial efficiency rewards that are often overlooked. This is especially crucial for clubs seeking long-term sustainability through business efficiencies. 
• Additionally, players who are available and not injured are likely to command higher transfer values, providing additional financial benefits for clubs.
• And lastly, as they are less likely to partake in emergency transfers due to injuries, clubs can save on costly temporary signings in the January transfer window.

Clubs are investing huge resources in data analysis and AI for players they don’t yet own. However, there is clear evidence for also expanding this investment into systems that enable more proficient data and AI processes for the players they already have within their squad.  

Brown poses the question, “Why would you spend money on data and applying the latest methods around the transfer of a player, but then manage that player once they are at the club using a more rudimentary method?”

Why indeed? A rebalance, undoubtedly, is needed.