Can We Really Predict Injuries in Football? Insights from FSI Talks Special Edition
In this special edition of FSI Talk, recorded live at the Benito Villamarín Stadium during the VII FSI Conference, leading sports scientists tackled one of modern football’s most controversial questions: Can we really predict injuries?
The session was moderated by Fabio Nakamura, Professor at Universidade da Maia and an international authority in football performance science. With extensive research experience and applied work with clubs and national teams, Nakamura led a discussion centered on one of the major challenges in high-performance football: injury prediction.
Joining him were two top-level experts: Chris Carling, Head of Elite Performance at the French Football Federation, and Chris Barnes, Sports Scientist at UEFA and consultant for Brøndby IF. Both offered evidence-based insights on the use of data, load management, and the real limits of injury prediction models in professional football.
Stop Talking About “Predicting” Injuries
One of the most revealing points came early in the discussion: the focus shouldn’t be on predicting injuries.
“It doesn’t work like that. It’s multifactorial.”
While models such as the Acute:Chronic Workload Ratio have gained popularity as tools to “predict” injury risk, Carling and Barnes emphasize that injuries result from a complex combination of factors: physical, psychological, contextual, and even genetic. No single metric can predict exactly when a player will get injured. The focus should instead be on risk management and preparing for variability.
Internal vs External Load: Both Matter
Another key point was the imbalance between monitoring external and internal load. Many clubs rely heavily on GPS data, distances, speeds, and sprints—external load—without giving equal attention to internal responses such as heart rate, perceived effort, or neuromuscular fatigue.
“Internal load tells us how the player is actually responding.”
Combining both dimensions provides a more complete picture of a player’s condition. External load is what we impose; internal load is how the body reacts. Ignoring either can lead to poor decisions and increased injury risk.
Load Management is Highly Individual
A critical takeaway: load doesn’t affect all players equally.
“Same load. Different risk.”
Two players with the same age, position, and physical profile may have vastly different injury risk levels, even when exposed to identical workloads. Individualization is therefore essential in load management.
It’s not just about collecting data but interpreting it in context. A player’s training history, age, sleep, stress, match demands, and even travel can all influence their capacity to handle load.
Coaches Know More Than We Think
The discussion also highlighted a common bias in sports science: underestimating the intuition of coaches.
“Some coaches just know. They see a player and sense something is off, even before the data shows it.”
Experienced coaches who work with the same players over time develop insights that can be validated and integrated with data. Intuition and science should coexist, not compete.
Technology Isn’t the Problem—Systems Are
Despite advances in tracking technology, one major bottleneck remains: data access and integration.
“It’s not just technology, it’s how we manage the data.”
Common issues include limited access to raw data, difficulty identifying outliers, poor integration between GPS and Athlete Management Systems (AMS), and unstable APIs. Improving workflows, data channels, and communication between platforms is as important as investing in the latest hardware.
Is AI the Future of Injury Prevention?
Artificial Intelligence emerged as a promising topic, though often misunderstood. While AI won’t predict injuries in the short term, it can add value by:
- Automating pattern detection
- Identifying risk clusters
- Improving load planning
- Reducing time spent on repetitive analysis
“AI won’t replace coaches or sports scientists, but it will save them time.”
AI should support human decision-making, not replace it.
What Modern Sports Scientists Need
Towards the end of the episode, the conversation shifted to the evolving role of performance staff. Modern sports scientists are highly skilled in data, programming, and physiology, but often lack essential interpersonal skills:
- Effective communication with coaches
- Teamwork across departments
- Leadership under pressure
- Flexibility and critical thinking
“They know how to use R and Python, but not always how to talk to a coach.”
This highlights the need for well-rounded professionals who combine technical expertise with emotional intelligence and collaboration.
Conclusion
If you work in performance, rehabilitation, training, or sports science, this episode is a must-watch. It challenges the notion of injury prediction and emphasizes a holistic, evidence-based approach: combining data, intuition, individualization, and smart load management.
For those seeking to go deeper, FSI offers advanced training programs like the FSI Master in Strength & Conditioning Coach and FSI Master in Injury Rehabilitation, both providing hands-on experience with professional football teams—a unique opportunity to apply knowledge directly on the pitch.
Stay informed about the latest in football performance science by subscribing to the FSI newsletter and following FSI on social media.
