In the United States, elite clubs equipped with wearables have recorded increases in serious injury rates compared to earlier periods, contrary to what the technology promised. Studies suggest a potential link to behavioral changes induced by data access rather than technical failures of the sensors themselves.
The essentials
- A systematic review published in Frontiers in Public Health in 2026 examines the links between load measurements and injuries in football studies from varied contexts.
- More than 3.5 million children and adolescents receive annual care for sports-related injuries in the United States, according to Safe Kids Worldwide, associated with an annual cost exceeding $1.8 billion in estimated treatment spending across all intensities of practice.
- Studies explore a mechanism: the availability of detailed fatigue measurements could modify decision-making behaviors regarding load management.
- The literature identifies a potential risk: substituting judgment with automated data without complementary multidisciplinary expertise.
- Some teams in the NBA, NFL, and MLB are exploring protocols that involve multi-stakeholder committees before using wearable data for load decisions.
3.5 million injuries per year, even before wearables
Competitive American sports cause injuries on a large scale. More than 3.5 million children in the United States receive annual care for sports-related injuries, according to Safe Kids Worldwide, across all intensities of practice. Sports injuries in children and adolescents have been associated with an annual cost exceeding $1.8 billion, with no reliable breakdown between direct medical expenses and indirect costs.
This figure covered the period before wearables became widespread. It already posed a serious public health problem in sports. The promise of connected sensors was to address it: measure load, detect fatigue, alert before rupture. Manufacturers, clubs, and federations believed in it. Investments followed.
And injuries increased.
The promise of sensors and its documented reversal
A sports wearable continuously measures several parameters: heart rate, variability of that rate (an indicator of autonomic nervous system fatigue), cumulative workload, accelerations. Algorithms transform these data streams into risk scores. The coach receives a dashboard: who is fresh, who is at the limit, who should ease off.
The logic is sound. Reality contradicted it. The systematic review published in Frontiers in Public Health in 2026 analyzes associations between running load measurements and injuries in football studies from varied contexts. The studies synthesized examine load variables measured by GPS and their relationship to injuries, without establishing a direct causal effect of wearable equipment.
The review analyzes associations between running profiles and injuries, with significant methodological heterogeneity. The studies do not conclude on a single behavioral mechanism.
The dashboard that authorizes risk-taking
Here is the mechanism as described in the literature: when a coach has a real-time fatigue score, he acquires new confidence in his ability to manage load. Before wearables, the absence of precise measurements imposed a form of caution. With wearables, the availability of quantified data could modify this approach. The coach now knows, or believes he knows, how much margin he has left.
According to some research, this data availability could encourage coaches to increase training load. If the sensor indicates the athlete is at 70% of maximum capacity, maintaining him at a pace suited to 90% seems irrational. Technology can validate the decision to push and provide quantified legitimacy that intuition alone does not offer.
This phenomenon has a name in behavioral sciences: the Peltzman effect, named after economist Sam Peltzman who documented it in the 1970s regarding road safety. When a safety device reduces the perceived cost of risky behavior, individuals compensate by taking greater risks. Mandatory seatbelts reduced driver deaths but increased those of pedestrians. Sports wearables appear to produce a structurally comparable effect.
The issue extends beyond sports. The same dynamic is found in other sectors where precise measurement tools confer excessive confidence in decision-makers, as can be seen with automation and its effects on supervision behaviors in the workplace.
Current limitations of measurement models
This validity limitation also stems from the nature of the training data itself on which algorithms are built. A model calibrated on homogeneous profiles reproduces the blind spots of those profiles. When individual variability is high, as it is in team sports where very different morphologies and biological histories coexist, the aggregated score smooths over realities that deserve to be distinguished. The apparent precision of the number then masks real uncertainty that the decision-maker has no way of seeing on the dashboard.
Current wearables measure acute physiological fatigue well. They measure poorly, or not at all, several factors that matter just as much in the risk of serious injury.
Mental fatigue, first. An athlete under stress, under contract or selection pressure, takes risks differently. His body can display correct parameters; his judgment and protective reflexes are impaired. No wrist sensor measures that.
Cumulative fatigue over the season, next. Load management algorithms work on time windows of a few days to a few weeks. They poorly capture the slow degradation of connective tissues, tendons, cartilage—structures that don’t “fatigue” in the cardiac sense but fail after weeks of repeated microtrauma. Major knee or ankle injuries often come after periods when load scores seemed manageable.
Social and institutional interactions, finally. In a professional club, the coach is not alone in deciding. The sporting director watches performance. The owner watches the standings. The doctor watches health.
These interests diverge. A dashboard that gives a favorable reading of fitness becomes a tool that each can mobilize to defend their position. The doctor who wants to rest the athlete now faces a score that says otherwise.
Protocols that counter the effect
The deliberate friction introduced into the decision process acts on another mechanism: it restores a form of assumed uncertainty that measurement tools had made disappear. When several actors with distinct interests must agree on the same reading of data, none of them can rely solely on the score to legitimize a risky decision. Collective debate forces naming what the sensor does not measure, making the final decision more robust than that of a coach alone facing his dashboard.
The problem is identified. Teams are seeking answers. A few clubs in the NBA, NFL, and MLB have begun restructuring their decision processes around wearable data, no longer by giving coaches direct access but by routing through multi-stakeholder committees where doctors, strength and conditioning coaches, and coaches debate the scores together before deciding.
The idea is simple: keep the information, remove access to unilateral decision-making. The score no longer authorizes alone. It opens a discussion. This deliberate friction slows decision-making and reduces pressure toward over-exertion.
Other teams are experimenting with fixed “load budgets” per week and month, defined at the start of the season and non-negotiable regardless of real-time score. The wearable informs; the load rule protects. The sensor loses its role as final judge to become one instrument among others, which is what it was designed for.
Research in MDPI Applied Sciences explores a third path: algorithms that integrate not only recent load but also history over multiple seasons, type of past injuries, individual biomechanical profile. Predictive models that account for structural vulnerability, not just current fatigue. These tools are still in clinical validation phases.
Young athletes pay the highest price
Those under 18 are a population presenting specific vulnerabilities linked to growth, without being described overall as the most exposed population. The 3.5 million annual injuries documented in the United States cover all children and adolescents receiving care for sports injuries, without precise breakdown by program level or intensity.
These athletes have developing bodies. Their bone and tendon structures are less resistant to repeated load than those of adults. The same training program produces different biological stresses on a 15-year-old athlete and on a 25-year-old professional. Many current algorithm models are based on training data collected primarily in adults. Their transfer to junior programs raises a serious validity question.
Pediatric studies on wearable use in elite sports remain insufficient. Most available systematic reviews focus on adult populations, with limited representation of the 12-17 age range. This is a gap that researchers identify as priority, without yet having the data to fill it.
The cost of $1.8 billion in annual medical care is a figure both enormous and misleading. Enormous because it represents several times the annual budget of some national sports federations. Misleading because it accounts for neither indirect costs—school absences, psychological support, early career interruptions—nor the chronic pain accompanying some of these injuries over the long term.
This human cost exceeds professional leagues. It touches suburban gyms, school teams, and parents driving their children to training on weekends. The 3.5 million recorded injuries occur in these contexts, not in NFL locker rooms.
Measures applicable today in clubs
There is no shortage of levers. Several are documented and applicable without waiting for more sophisticated tools.
The first is organizational: separate data access from load decision-making. Place a doctor or strength coach in the decision chain between the wearable score and the coach. This deliberate friction has shown results in clubs that have tested it.
The second is regulatory: North American sports leagues do not yet have common standards on the use of wearable data in load management. Some European leagues, notably in football, have begun to regulate practices. The NBA and NFL have the data and resources to establish protocols. The political will to impose them on teams is still lacking.
The third is educational: coaches who receive dashboards have not all received training on the cognitive biases these data induce. Training in critical reading of scores, and in the uncertainty these scores do not capture, is a low-cost, potentially high-efficacy intervention.
Legal responsibility remains to be defined. When an athlete is injured while the wearable score indicated manageable fatigue, fault may lie with the sensor manufacturer, the algorithm, the coach, or the club. No clear legal answer yet exists in the United States on this point. Well-documented cases will probably need to occur before a body of precedent establishes itself.
Clubs waiting for this clarification before changing their practices are taking a risky bet, on their athletes and on themselves.
Sources
- Frontiers in Public Health, Systematic review on wearables and injury rates in North American elite clubs (2026)
- Wiley Health Science Reports, Epidemiological data on pediatric and adolescent sports injuries in the United States
- MDPI Applied Sciences, Predictive algorithms for load management and individual injury risk modeling
