Overtraining Syndrome: The Diagnosis Elite Sport Still Gets Wrong
This is part of our series on the factors actually shaping elite athlete performance in 2026. As ever, this is not a manual written from the touchline. It is a map of where the evidence is strong, where it is thin, and what a performance department should ask before trusting a claim that a device can see something the medical literature says it cannot.
Few terms in elite sport are used as loosely as “overtraining.” It is invoked to describe everything from a tired week to a career-threatening collapse in function, and that imprecision is not harmless. It scares developing athletes away from the training loads they need, it leads to misdiagnosis in both directions, and it creates a market for technology that promises to detect a condition the medical literature says cannot yet be detected by any single measurement. This article sets out what overtraining syndrome actually is, why it remains so difficult to diagnose, and why most claims to predict it should be treated with caution.
A spectrum, not a switch
The foundational document remains the joint consensus statement of the European College of Sport Science and the American College of Sports Medicine, authored by Meeusen and colleagues in 2013 (Meeusen et al., 2013). Its most important contribution was terminological: what is loosely called “overtraining” is in fact a spectrum of three distinct states, separated mainly by how long recovery takes.
The first is functional overreaching. This is the intended consequence of a deliberate training overload: a short-term decrement in performance that, after adequate recovery, leads to a supercompensation and an improvement. Functional overreaching is not a problem to be avoided; it is, in the consensus framing, part of how training works. Recovery is measured in days.
The second is non-functional overreaching, which occurs when the balance between training and recovery is not sufficiently respected. Here the performance decrement is more pronounced and recovery takes weeks to months. Crucially, there is no supercompensation; the athlete simply loses time and function.
The third is overtraining syndrome itself, where recovery extends to months or even years, and which the consensus describes as involving a prolonged maladaptation not only of the athlete but of multiple biological, neurochemical and hormonal regulatory mechanisms. The distinction between non-functional overreaching and overtraining syndrome, the consensus notes, is very difficult to draw and depends on the clinical outcome and an exclusion diagnosis.
That phrase, exclusion diagnosis, is the heart of the problem.
Why it cannot be detected by a single marker
The defining diagnostic feature of overtraining syndrome is that it is diagnosed by exclusion, and this is not a quirk of the 2013 statement; it is the settled position of the field. A 2022 scoping review of every biomarker and tool proposed as diagnostic for the syndrome reached the same conclusion, noting that because of the lack of a gold-standard diagnostic test, overtraining syndrome remains a diagnosis of exclusion (Carrard et al., 2022). Before a clinician can attribute an athlete’s prolonged performance decrement, fatigue and mood disturbance to overtraining syndrome, they must rule out the other possible causes: underlying illness, infection, nutritional deficiency, iron status, thyroid dysfunction, depression and other clinical conditions that can produce the same presentation. There is no confirmatory test. The diagnosis is reached only by eliminating everything else, which is why it takes time and why it cannot, even in principle, be delivered by a wearable on a given morning.
This matters because the marketing of monitoring technology frequently implies the opposite. A device or platform that claims to “detect overtraining” is making a claim the diagnostic framework does not support, because overtraining syndrome is not a state with a single measurable signature. The consensus statement and subsequent reviews have repeatedly noted that the hormonal and biochemical markers proposed as candidates, the cortisol-to-testosterone ratio being the most cited, are frequently normal in athletes who are genuinely overtrained, and lack the diagnostic specificity to confirm the condition. A marker that is often normal in the presence of the disease cannot be used to rule it in or out.
This does not mean the markers are useless. Tracked longitudinally against an individual’s own baseline, some may contribute to the broader clinical picture. But that is a very different claim from “this device detects overtraining,” and the gap between the two is exactly where over-claiming lives.
How thin the evidence base actually is
It is worth pausing on just how underpowered the underlying science is, because it reframes every confident commercial claim built on top of it. A 2022 systematic review set out to find studies that objectively documented the physiological and psychological changes an athlete goes through as they cross from healthy into an overtrained state, with performance suppressed for more than four weeks. It found none. Not a single study met that bar (Weakley et al., 2022). The authors attributed this to vague terminology, the difficulty of monitoring athletes prospectively for long enough, and the ethical impossibility of deliberately overtraining people to study them.
The implication is stark. The field does not yet have a well-characterised, prospectively documented picture of how overtraining syndrome develops in the very population it most concerns. A commercial tool that claims to detect the onset of a condition whose onset has never been cleanly documented in the literature is, at best, pattern-matching against an incomplete template. This is not an argument that the syndrome is not real. It is an argument for humility about how well anyone, human or algorithm, can currently identify it in progress.
The rarity problem
A further complication, often missed, is that genuine overtraining syndrome appears to be far less common than the casual use of the term implies, although its precise prevalence is difficult to establish because the diagnosis itself remains uncertain. Most of what is described as “overtraining” in practice is non-functional overreaching, or simply under-recovery driven by poor sleep, inadequate nutrition and accumulated life stress, rather than the full syndrome with its months-to-years recovery horizon.
This rarity has the same statistical consequence that affects injury prediction: when a condition is genuinely rare, any tool claiming to detect it faces a severe base-rate problem. A detector for a rare condition will, unless extraordinarily specific, generate far more false alarms than true cases, and each false alarm carries a cost, in this case the unnecessary deloading or withdrawal of an athlete who was not actually overtrained. The same arithmetic that undermines individual-level injury prediction applies here, and for the same reason.
What monitoring can and cannot contribute
None of this means monitoring is irrelevant to the management of training load and fatigue. It means the contribution is different from what is often advertised.
What monitoring can do is track an individual athlete’s own trends over time: their resting physiology, their sleep, their subjective wellness, their performance markers, each interpreted against that athlete’s established baseline rather than a population norm. A sustained, multi-system departure from baseline, especially when several independent markers move together and align with the athlete’s lived experience, is a legitimate prompt to investigate and to consider whether recovery is being adequately respected. That is real, useful information.
What monitoring cannot do is deliver a diagnosis. It cannot distinguish, on its own, between functional overreaching that will resolve favorably with recovery and the early stages of something more serious, because that distinction is defined by the eventual clinical outcome, not by a contemporaneous reading. And it cannot substitute for the exclusion process that genuine diagnosis requires. A device that flags “overtraining risk” is, at best, flagging a departure from baseline that warrants a human looking more closely, which is valuable, but is not the same as detecting the syndrome.
The position a department can hold
Overtraining syndrome is real, serious, and, properly understood, comparatively rare. The errors elite sport makes around it are errors of precision: using the term loosely, treating non-functional overreaching as the full syndrome, and trusting technology to detect a condition that is defined by exclusion and lacks any single confirmatory marker.
The position the evidence supports is this. Use the consensus terminology precisely, and distinguish functional overreaching (intended, days to recover) from non-functional overreaching (unintended, weeks to months) from overtraining syndrome (months to years, diagnosed by exclusion) (Meeusen et al., 2013). Treat any marketing claim to “detect overtraining” with skepticism, and ask which validated diagnostic criteria the claim is measured against; if the honest answer is none, treat the output as a prompt to investigate rather than a diagnosis. Use monitoring for what it does well, tracking individual departures from baseline across multiple markers, and route any genuine concern to a clinician who can conduct the exclusion process the diagnosis actually requires. And remember that for most athletes, most of the time, the issue is under-recovery rather than the syndrome, and the most effective intervention is the least technological: more sleep, better nutrition, and respect for the balance between load and recovery that the consensus identified as the whole point.
The diagnosis elite sport gets wrong is not usually a missed case of overtraining syndrome. It is the routine over-use of a serious clinical term, and the trust placed in tools that promise to detect what the medical literature says cannot yet be detected by measurement alone.
How this series is made, and how to read it: this is editorial analysis, not a practitioner’s memoir and not a systematic review. PERFORM’s pieces are researched and drafted with the assistance of AI tools, then reviewed, edited and fact-checked by our editorial team against primary sources, peer-reviewed literature, clearly labelled preprints, industry reports, league and company announcements, and practitioners’ own published work. Where the evidence is strong we say so; where it is limited we treat it as limited; where a claim comes from a vendor or corporate announcement we treat it as a hypothesis, not proof. The views here are our editorial position, drawn from the published record rather than first-hand experience inside an elite performance department. Where practitioners are named or quoted, those words are their own. Where we couldn’t verify a claim, we left it out. And where you have the hands-on experience we’re writing about, we’d rather hear from you than pretend to it.
References
Carrard, J., Rigort, A.-C., Appenzeller-Herzog, C., Colledge, F., Königstein, K., Hinrichs, T., & Schmidt-Trucksäss, A. (2022). Diagnosing overtraining syndrome: A scoping review. Sports Health, 14(5), 665–673. https://doi.org/10.1177/19417381211044739
Meeusen, R., Duclos, M., Foster, C., Fry, A., Gleeson, M., Nieman, D., Raglin, J., Rietjens, G., Steinacker, J., & Urhausen, A. (2013). Prevention, diagnosis, and treatment of the overtraining syndrome: Joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Medicine & Science in Sports & Exercise, 45(1), 186–205. https://doi.org/10.1249/MSS.0b013e318279a10a
Weakley, J., Halson, S. L., & Mujika, I. (2022). Overtraining syndrome symptoms and diagnosis in athletes: Where is the research? A systematic review. International Journal of Sports Physiology and Performance, 17(5), 675–681. https://doi.org/10.1123/ijspp.2021-0448