Recovery tracking. What the research actually supports.

What recovery means in the research

The field's working definition comes from a consensus statement published in 2018 in the International Journal of Sports Physiology and Performance, the output of an international expert symposium. Its framing is a balance: training and competition load on one side, and on the other everything the body does to absorb it — with the explicit note that *other life demands* sit on the load side too, alongside training.

The reason that balance matters is set out in a joint consensus statement from the European College of Sport Science and the American College of Sports Medicine, published in 2013. It describes a continuum. Training works by overloading you and then letting you recover; push a little past that and you get *functional overreaching*, a short-term drop in performance followed by improvement once you have rested. Push further, or rest less, and you move into non-functional overreaching and eventually overtraining syndrome, which are not a harder version of the same thing but a prolonged failure to adapt.

So recovery is a real concept with real consequences. What follows is about how badly it is measured.

There is no single validated marker of recovery.

This is the finding the whole topic rests on, and it is stated by the field itself rather than by sceptics outside it.

The 2013 joint consensus statement says plainly that no single marker currently satisfies the criteria for diagnosing overtraining syndrome, and that telling non-functional overreaching apart from overtraining syndrome is difficult — it depends on how things actually play out over time and on ruling other explanations out one by one.

A 2014 review in Sports Medicine went through the whole toolkit — power output and movement analysis, perceived effort, heart rate, blood lactate, training load calculations, heart-rate recovery after exercise, neuromuscular tests, blood-based measures, questionnaires and sleep — and reached the same conclusion in blunter language: very few of these markers have strong scientific evidence supporting their use, and no single definitive marker has yet been described in the literature. It also found that blood-based biochemical, hormonal and immune measures were not, on the evidence available, justified as indicators of internal training load.

None of that means recovery is fictional. It means a device showing you a single recovery number is showing you a company's answer to a question the research has not settled — which is a fair thing for a product to do, and an unfair thing for a product to hide.

The same signal, two opposite meanings.

Most consumer recovery numbers lean heavily on heart-rate measures, which makes one particular finding the most important thing on this page.

A 2016 systematic review with meta-analysis in Sports Medicine pooled studies of endurance-trained athletes to test whether autonomic heart-rate measures — resting variability, variability after exercise, and how quickly heart rate falls after exercise — could distinguish healthy adaptation from overreaching. Positive adaptation was associated with small increases in resting variability related to vagal influence, and moderate increases in the post-exercise measures.

The problem is that overreaching produced changes in the same direction. The measures moved similarly whether an athlete was adapting well or heading somewhere unhelpful, which led the authors to conclude that additional measures of training tolerance are likely to be needed alongside heart-rate indices.

Read that carefully, because it is easy to under-react to. It does not say heart-rate measures are useless. It says a rise or fall in them does not, by itself, tell you which of two opposite states you are in — and a single number on a screen cannot resolve an ambiguity that a laboratory could not.

One day is mostly noise

The 2014 Sports Medicine review reported that submaximal heart rate — measured under standardised conditions, which is more careful than anything happening on your wrist — varies day to day by up to around 6.5%. It found heart rate variability results inconsistent across studies, with high day-to-day variability, and concluded that weekly rolling averages are more valid than single-day values.

A 2014 review in Frontiers in Physiology makes the same case from the other direction: many of the contradictions in heart-rate-based monitoring come from methodological inconsistency rather than from the measures themselves, and an individual's change has to be interpreted against measurement error, against the smallest change that would actually matter, and against training context. Its practical recommendation is to combine heart-rate monitoring with training logs, questionnaires and performance tests, rather than relying on heart rate alone.

Which gives a rule that holds regardless of which device you own: a single day's movement in a recovery number is, more often than not, smaller than the noise. Copy that dramatises one day's change is dramatising measurement error.

Where sleep comes into it

Sleep is the recovery variable people most want a simple answer about, and the evidence is more mixed than either the wellness industry or common sense suggests.

A 2015 systematic review in Sports Medicine examined sleep loss in athletes and found the physical picture is genuinely split: some maximal efforts hold up despite sleep loss, while sport-specific performance typically declines. Cognitive performance becomes slower and less accurate. The review also describes possible mechanisms — an imbalance in the autonomic nervous system that resembles the symptom picture of overtraining syndrome, raised inflammatory markers, impaired immune function — and is careful to present these as possible rather than established, with the authors saying further research is needed.

So: sleep matters to recovery, and the honest version of that sentence is less sweeping than the version you usually read. It is also worth noting what the review does not establish, which is that improving a sleep score improves recovery or performance. No study cited anywhere on this page tested that.

Who this research was done on

This is the most common over-claim in the whole topic, so it gets a section rather than a footnote.

Almost all of the literature above studies trained or competitive athletes, under standardised measurement conditions, often in a laboratory. The 2016 meta-analysis is specifically about endurance-trained athletes. The 2013 consensus statement is about competing athletes and a diagnosable syndrome. The measurement protocols involve controlled posture, controlled timing, and equipment that does not sit on a wrist.

Extending any of it to a general population wearing a consumer device overnight is an extrapolation, and should be described as one. That is not a reason to ignore the research — it is the best evidence there is — but it is a reason to be sceptical of any product that quotes sport science at you as though the study population were you.

Where Omnisio fits

Omnisio shows a Recovery reading each morning, so the honest thing to do on this page is say what it is and what it is not.

It is a wellness reading derived from consumer-grade sensors, read against your own recent weeks rather than against other people, and it is not validated by any of the research described above — because nothing in that research validates any product's recovery score, from any company. Two design decisions on it follow the evidence on this page rather than working against it: the comparison is always against you, because between-person comparison is not supportable; and on a night that could not be read honestly, there is no score at all, because a number assembled from a thin recording is not a low reading of you but a reading of nothing.

How the reading is produced is not published, here or anywhere else. What it contains, what sits around it and what happens on an unreadable night are described in full on the Recovery page, linked in the rail beside this section.

A low number is not overtraining

Overtraining syndrome is a diagnosis of exclusion, arrived at by ruling other causes out over time. It is not what a wearable is describing when it shows you a low figure, and it is not the right word for ordinary tiredness, a heavy week, or a bad night's sleep.

Omnisio is a wellness product for general lifestyle use. Neither the app nor the devices used with it hold FDA, CE-MDR or TITCK medical device clearance, and nothing in it is intended to diagnose, treat, cure or prevent any disease. It cannot detect overtraining, it does not screen for illness, and a low reading is not evidence of either.

Persistent, unexplained fatigue that does not lift with rest is worth taking to a physician rather than to an app — and a recovery number that stays low for weeks is a reason to have that conversation, not a substitute for it.

Where this comes from

Six published sources, named plainly. Each was confirmed live on 8 September 2026.

• Source: European College of Sport Science and the American College of Sports Medicine, 2013. Joint consensus statement on overtraining: defines the overload, functional overreaching, non-functional overreaching and overtraining syndrome continuum, and states that no single marker currently satisfies the criteria for diagnosing overtraining syndrome.
• Source: International Journal of Sports Physiology and Performance, 2018. Consensus statement on recovery and performance: the stress-and-recovery balance including non-training life demands, and explicit acknowledgement of variability between and within individuals. Its statement that monitoring aims at preventing under-recovery is an aim, not demonstrated efficacy.
• Source: Sports Medicine, 2014. Review of training-load and fatigue monitoring: very few markers have strong evidence supporting their use, no single definitive marker has been described, submaximal heart rate varies day to day by up to about 6.5%, heart rate variability findings are inconsistent, and weekly rolling averages are more valid than single-day values.
• Source: Sports Medicine, 2016. Systematic review with meta-analysis in endurance-trained athletes: positive adaptation associated with small increases in resting vagal-related variability and moderate increases in post-exercise measures, but overreaching produced overlapping changes, leading the authors to conclude additional measures of training tolerance may be required.
• Source: Frontiers in Physiology, 2014. Methodological review: contradictions in heart-rate monitoring often come from inconsistent method, individual changes must be read against measurement error and the smallest important change, and heart rate should be combined with training logs, questionnaires and performance tests.
• Source: Sports Medicine, 2015. Systematic review of sleep loss in athletes: mixed effects on physical performance, typical decline in sport-specific performance, slower and less accurate cognitive performance, and possible autonomic, inflammatory and immune mechanisms presented as requiring further research.

Questions