The role of invisible monitoring in elite sports

What is invisible monitoring in athlete health and performance, and how does it differ from traditional monitoring approaches?

Invisible monitoring is a process to track and analyse an athlete’s training and performance while minimising disruption to their routine, developing approaches to collect and assess a player’s training in a way that feels natural, without interrupting training or daily life. We provided a first set of constitutive and operational definitions. This is an initial step and needs to reach consensus, which hopefully should ease its implementation by creating a common language and avoiding misconceptions.

Invisible monitoring differs from traditional methods through the aims to reduce the burden and increase the versatility of measurement tools so they can capture multiple constructs. Part of our work was integrating those two dimensions (burden and number of constructs measurable) within this conceptual framework.

What are the primary challenges associated with implementing invisible monitoring in elite sports settings?

Implementing invisible monitoring is a smooth idea until you hit reality, as numerous challenges exist ^[1]. First, without good processes in place, managing the different data sources can quickly become overwhelming. Additionally, some of these new monitoring techniques rely on estimates (e.g., predictive modelling), which are not always valid or reliable.

Athlete awareness / buy-in is another big factor. If players don’t understand how their data is being used, they might feel uncomfortable or even resist using these new approaches ^[2]. To succeed, sport organisations must integrate valid and reliable data seamlessly, ensure ethical transparency, and educate both staff and athletes on the benefits and limits of invisible monitoring. Addressing these concerns proactively can enhance adoption, trust, and long term success. 

How does invisible monitoring address the issue of athlete burden compared to traditional assessment methods?

Traditional monitoring (e.g., CMJ, wellness questionnaires) requires athletes to participate in structured assessments that are time consuming, maximal, and repetitive, leading to decreased player buy-in. The data collected is also used for a limited number of purposes (e.g., fatigue monitoring).

Invisible monitoring approaches try to eliminate this additional burden by adapting the primary purpose of a measurement tool to a secondary purpose to provide additional insights indirectly into players’ status. Multiple pitch-based examples are now available in the scientific literature: Martin Buchheit with a standardised running test, Amber Rowell with small sided games, or Jace Delaney with training load efficiency index ^[3–5]. These take advantage of the ubiquity of the accelerometers built into to GPS units, which are usually used to evaluate training load, to develop insights into player status. Mathieu Lacome and Mauro Mandorino have extended this work using predictive modelling approaches.

Just because a monitoring strategy is deemed invisible does not mean it will solve all the problems. We remain tentative about the term “invisible” itself, because it likely creates a false dichotomy: invisible vs. visible.

Monitoring strategies need to be considered within a continuum to avoid misconceptions. This is why we made a first attempt to classify different measurement techniques by measuring different levels of burden and constructs.

What technological advancements, such as wearables and AI-driven analytics, are enabling invisible monitoring in sports science?

Wearable sensors are becoming smaller, smarter, autonomous, and more seamless than ever, and might provide access to more and greater insights into player responses, and at multiple levels of the organism.

Imperial College London’s The Future of Wearable Technologies provides a useful overview of developments that are likely to infiltrate sport sciences practices in the future ^[6]. We have been missing practical measures of physiology in the field (particularly in team sport contexts) for a long time, and solutions are looking likely. Also, computer vision applications are consolidated in the technical-tactical realm, but are likely to be equally as useful for the physical monitoring of an athlete.

Cloud-based solutions and advanced AI modelling techniques can allow practitioners to be more precise by leveraging impactful trends from their data sources. This double enhancement will have a direct knock-on effect on performance as it will enable players, practitioners, and stakeholders to make better adjusted risk decision making.

Over the next horizon, our industry needs to look carefully what is going on outside our industry and potentially anticipate further advancements in other fields.

What are the potential limitations of using surrogate measures in invisible monitoring, and how can they be mitigated?

Using surrogate measures of a construct might draw you away from the actual phenomenon you are trying to measure, potentially leading to a wrong interpretation or decision. Surrogate measures have to be approached within the proper context and with due regard for their limitations.

These approaches need to be validated before integrating any new monitoring tools. Considering the frequency of measurement (potentially daily), protocols might need to align to ensure a sound validation, which might be logistically challenging, e.g., daily measurement of fatigue or fitness. In other words, compare your model with a gold standard measure for a sufficient period to minimise potential errors.

Some guidelines should come together regarding the most appropriate way to validate such new approaches. Additionally, we need to share the uncertainty we experience around these measures. Practitioners need to acknowledge that we do not have all the answers: data informs but does not replace expert judgment.

How can sports organizations improve athlete trust and transparency regarding the use of invisible monitoring technologies?

It feels like you must redo the job of 15 years ago when GPS turned up and players thought it would affect contract decisions. It was a long process, sometimes still ongoing, to persuade them that these devices and practices help and inform, and are not the basis for judging!

The key first step, then, is to inform athletes on why we are monitoring, how it might (or not) benefit them, and what data we are collecting. Transparency builds trust and engagement.  This must start early and should be part of the player development pathway at the academy. At the senior level, informational meetings, infographics, and informal discussions should be part of the process.

Players also need to see the added value of the approach. Staff must ensure they have the capacities to provide regular individual reports to the player, as well as the depth and breath to discuss the outcomes they observe.

From an organisation standpoint, clear policies about data ownership, privacy, and ethics are needed. Communications with internal legal departments will help ensure good practices are in place.