Key variables from EPTS devices

/in /by

Sports medicine professionals need to consider key values when analysing data from EPTS devices

The advent of wearable and tracking technologies, commonly grouped under EPTS (Electronic Performance & Tracking Systems), has transformed how professional football clubs monitor athlete workload, performance, and injury risk. Devices such as WIMU Pro, Catapult, Fitogether, among others, provide a wealth of data. But for that data to be meaningful, sports-medicine professionals must know which variables to prioritise, and how to interpret them.

EPTS data generally fall into two broad categories. First, external load variables (eTL), which quantify the mechanical and locomotor stress placed on the athlete. These include metrics such as total distance covered, distance at high velocity or sprinting, number of accelerations/decelerations, player load, and neuromuscular load.

Second, internal load variables (iTL), reflecting the biological stress experienced by the athlete, for instance, physiological parameters (heart rate, oxygen consumption, metabolic markers) or perceived exertion, when the device or complementary systems allow.

Among external load metrics, certain parameters have shown particular relevance in elite football. Studies tracking high-speed running, sprint distance, accelerations/decelerations and overall “player load” correlate these metrics with risk of overload and injury.

For example, excessive high-intensity load without adequate recovery may predispose to muscle strains or overuse injuries.

However, not all measurement systems are equally reliable across all variables. A review of local positioning systems (LPS, including UWB-based ones) demonstrated acceptable validity for measuring distances (< 3.5 % difference compared to reference systems), but greater variability and lower reliability when assessing instantaneous speeds, accelerations, or decelerations, particularly in sport-specific movements with changes of direction.

Given these limitations, sports medicine professionals must carefully interpret EPTS data and avoid overreliance on any single metric. In practice, analysing a combination of variables offers more robust insight. For instance, coupling external load metrics (e.g. high-speed distance, number of sprints) with internal load indicators or subjective markers of fatigue better reflects the athlete’s true stress and recovery status.

Moreover, temporal context matters. Monitoring raw data across sessions is less informative than establishing individual baselines (e.g. typical weekly load distribution, pre-season load, match load) and detecting deviations. Also, trend analysis over time (accumulated load, spikes in intensity, insufficient recovery periods) is more valuable than isolated snapshots.

Finally, data quality and consistency are crucial. It is advisable to maintain the same tracking technology (e.g. same device type, positioning system) across sessions to avoid biases introduced by differing measurement systems. Some studies have shown that different devices or technologies may register divergent values even for identical sessions.

In a nutshell, for EPTS-based data to inform medical or performance decisions, practitioners should: select validated systems, prioritise a set of key external and internal load metrics, contextualise data longitudinally rather than cross-sectionally, and combine objective measures with subjective or physiological markers.

Through adopting such rigorous data-driven monitoring, VALITICA can support clubs in making informed, evidence-based decisions about training load, recovery protocols, and injury prediction strategies, ultimately enhancing athlete welfare and performance longevity.

Shall we talk?

Can injuries in elite Athletes be predicted?

/in , /by

Big Data, AI and ML are the best solution for elite athletes to avoid injuries.

A few years ago, we enjoyed the Football World Cup 2022, one of the most important and long-awaited sporting events for a long time. Therefore, it is important to understand different technologies that for some time now have been helping athletes to raise their performance and to know if they are going to have an injury during their physical preparation. Artificial Intelligence (AI), Big Data and Machine Learning have been the answer for those who once thought this was not possible; these, besides being great technologies, make sports clubs save money.

We know the worst thing can happen to an athlete is to get injured. We have seen great soccer stars, like Golden Ball winner, Karim Benzema, who missed the World Cup due to an injury in his left leg and had to stop his sporting career for months due to injury, thus weakening his team and opening a big emotional gap in himself. Although many think this is something usual in the sporting process, it is also an aspect of the athletes’ lives that is about to change.

Dr. Irene Aguirre, our Chief Medical Officer, stated that “these technologies are revolutionizing medicine in all areas and now, in the field of sports, they are very important from the point of view of prediction and sports performance, both individual and collective. Biomechanics, for example, is another area in which artificial intelligence is being developed through key studies in the development of multilevel neural networks.”

Data: The best (non-pharmaceutical) remedy for injuries

According to a study conducted on the Premier League, squads with 25 players suffer approximately 50 injuries during a season. This is 2 injuries per player per year, which costs the team about € 300 MM per year, approximately.

Due to this situation, physical trainers and sports technicians set themselves the task of investigating the correct way to keep track of each of the athletes, realizing that in the data there is a way to help both their club and their athletes. As Dr. Aguirre explained, “Athletes and physical trainers are aware of the importance of injury prediction methods and have been able to become familiar with and understand the importance of developing them, being, for example, the use of GPS during training hours the technology that helps the most in collecting data during the game and capturing most of the variables to be taken into account in injury prediction. The use of this geolocation technology in devices that are easy for athletes to wear such as wearables (smart electronic devices incorporated into clothing) allows the collection of medical data (heart rate, respiration, temperature, etc.) and physical data (position, speed, acceleration, etc.) of the player during training and matches. An example of this is FIFA’s EPTS technology which, thanks to devices inserted in the players’ inner tops, can collect all game data.”

While data is valuable, data alone does not prevent injuries: a method is needed to organize it. There are 5 variables for classifying data:

  • Volume: The size of the data is calculated. These can be in petabytes, exabytes and even zettabytes.
  • Variety: The consideration of structured and unstructured data in numerous forms and combinations.
  • Speed: Data transfer and analysis that can be done in fractions of seconds.
  • Veracity: Accuracy and reliability.
  • Value: Placing value on the information collected.

Likewise, just by organizing them we do not prevent an athlete from getting injured. It is necessary to cross large volumes of data at high speed (almost automatically) to project, for example, very precise training sessions that take into account the capabilities and biometric conditions of each athlete, thus avoiding possible overloads that lead to injuries in the future. In this sense, AI is key to perform this information processing and analysis.

It is also essential that all this information is used to predict, based on performance and workloads during each training session or competition, when an athlete is on the verge of injury. This can be achieved by technologies based on learning logics such as machine learning, a branch of AI.

An example of this is the work VALITICA has been doing to convert data into useful tools for athletes to deploy their full potential and to maximize injury prevention. This is achieved thanks to the systematization of variables that previously could not be measured due to their high level of subjectivity (such as the perception of physical effort, as well as the effect of rest periods and recovery times), but which are fundamental when it comes to reducing the risk of injury. In addition, this information is also key for the coaching staff to make the best decisions (based on scientific information and not on assumptions) when designing training sessions and even injury recovery.

Would you like to learn how VALITICA could help your team? Schedule a demo now →.