Future Wellness Briefing · Issue 002 · Thursday 17 September 2026
Issue No. 002: Resting State
The readiness score on your members’ wrists is not a measure of their nervous system, and the people who write the measurement guidelines have now said so in print.
Plus: a wrist that reads movement and calls it sleep, and a cold plunge trial that found nothing then reported something.
Readiness is sold as physiology. The guidelines say it is not one.
It is not measuring what it says
The guidelines. The American Physiological Society has published consensus guidelines on heart rate variability, the beat-to-beat timing signal every readiness score is built from. Their verdict on consumer devices is blunt: wearables are “neither valid measures of cardiac/vagal physiology nor direct proxies of sympathetic outflow”.
What survives. Time-domain measures and high-frequency power do reflect the breathing-linked, parasympathetic side of the signal, handled carefully. The low-frequency band does not isolate the sympathetic side, and the LF:HF ratio, still sold as autonomic balance, fails both physiologically and mathematically.
What that leaves. Most wearables report RMSSD or SDNN, or a branded score computed from them. On the marketing of those as recovery or training-readiness indicators, the guidelines are direct: “the evidence base for these recommendations is lacking”.
The comparison has never been run
The implicit claim. Every readiness score asserts that it tells you something you could not get more cheaply.
Never been tested. The guidelines state there is “a lack of data that device based HRV measures outperform self-report measures” or submaximal exercising heart rate in a performance context. The trial that would justify the category has not been run.
Their words, not ours. The authors call it a modern version of the McNamara fallacy, focusing on what is easily measured by wearables rather than on what is meaningful.
The trend line is the weakest part
Rank is stable, change is not. Users are told to ignore the absolute number and watch their own trend instead. That is backwards. HRV shows moderate to excellent reliability for ranking one person against a group, while absolute reliability, the kind that tells you whether your own number really moved, is generally poorer, with substantial within-person variability.
The reading is coachable. Slower, deeper breathing raises the number, peaking at about six breaths a minute, so a breathing exercise before the morning reading moves the measurement rather than the physiology.
Length decides the number. Ten-second snatches give reliability of 0.41 to 0.57, rising to 0.70 to 0.91 at two to six minutes. Most wrist-worn scores are built from windows at the short end of that range.
Punchline: the score is a real signal, measured badly, interpreted well beyond its evidence, and never once shown to beat asking the person how they feel.
| Variable | Specify | Verify with |
|---|---|---|
| Recording length | 5 min minimum, or a full sleep period | Device export showing window length |
| Time and posture | Same time of day, same posture, on waking | Timestamp and posture logged per reading |
| Breathing | No breathwork in the 10 min before | Protocol stated in staff training |
| Metric reported | RMSSD or HF power. No LF:HF, no balance score | Dashboard field list, audited |
| Comparator | One-question self-reported readiness, every session | Recorded alongside the device value |
Source: Future Wellness Group specification standard, 2026 revision, after Carter et al. (2026).
What to do this week
Ask first. Add a one-question readiness check at session start. It is the comparator nobody has beaten, and it costs nothing.
Fix the protocol. If you are going to record it, record it properly: same time, same posture, no breathwork immediately before, five minutes minimum or a full sleep period.
Drop the ratio. Strike LF:HF and autonomic balance from dashboards, reports and staff training. It does not mean what it is used to mean.
Say less. Cross out any line describing a wearable score as a measure of the nervous system, recovery state or readiness. Keep what you can defend: a lifestyle signal, useful for engagement.
Evidence Check
Claim: a wrist-worn device measures sleep
What is claimed. That wrist-worn actigraphy reports total sleep time, sleep efficiency and sleep stages accurately enough to act on for an individual.
What was measured. A research-grade wrist monitor was compared against overnight polysomnography, the clinical standard, in 107 patients at a London sleep centre across eight nights.
Where it breaks. The wrist read 27 minutes more total sleep a night than the clinical test and overstated sleep efficiency by six percentage points. For any single patient, agreement ran from 97 minutes under to 151 minutes over. A group average that looks respectable hides a per-person error larger than most people’s entire sleep debt. Overestimation grew with age, which is the population most sleep offers are priced at.
What would settle it. Per-person agreement limits published alongside the group mean, for the specific device and firmware in use, in the population being sold to. Manufacturers hold this data and do not publish it.
Position for the Register: wrist-worn sleep staging is an adjunct to clinical measurement, not a substitute for it. No consumer sleep score has been shown to support individual triage or pricing. Verdict: adjunct, not a measurement.
Also this week
Position
A readiness score is a measurement problem wearing the costume of a physiological one, and the cheapest comparator in the building has never been beaten.
For the sceptics
- Consensus guidelines on heart rate variability measurement and interpretation. Review, no primary sample, covering human cardiovascular research and wearable-derived HRV. Carter, J.R. et al. (2026) ‘Heart rate variability guidelines’, American Journal of Physiology: Heart and Circulatory Physiology, 331(3), pp. H918–H943. doi:10.1152/ajpheart.00041.2026
- Wrist actigraphy against simultaneous inpatient polysomnography. 107 sleep centre patients, eight nights. Drakatos, P. et al. (2026) Journal of Sleep Research. doi:10.1111/jsr.70456
- Cold water immersion and next-day endurance. 11 collegiate rugby players, randomised crossover, eight minutes at 15 °C. Dobashi, K. and Hirai, T. (2026) The Journal of Sports Medicine and Physical Fitness, 66(9), pp. 993–1001. doi:10.23736/S0022-4707.26.17891-8
- Stepping intensity and mortality. Prospective cohort, 64,743 UK Biobank participants with wrist accelerometry, 1,697 deaths. Wei, L. et al. (2026) British Journal of Sports Medicine. doi:10.1136/bjsports-2025-111173
- Sensor packs against laboratory equipment. Method-agreement pilot, 16 female Division 1 athletes, single session. Huff, D. et al. (2026) International Journal of Sports Physiology and Performance, pp. 1–4. doi:10.1123/ijspp.2026-0016
- Wearables for insomnia. Systematic review and meta-analysis, 16 randomised trials, 910 participants. Zhu, W. et al. (2026) Journal of Medical Internet Research, 28, e93496. doi:10.2196/93496
- Sleep extension and driving. Randomised controlled trial, 99 completers aged 18 to 24, six-month follow-up. Rossa, K.R. et al. (2026) Accident Analysis and Prevention, 238, 108767. doi:10.1016/j.aap.2026.108767
Verification note: verified 16 September 2026 via PubMed, PubMed Central and Scite. Editorial notices were checked on every paper cited, and none carries a retraction, correction or expression of concern.
Written by Phil Learney. Phil Learney is founder of HMN24, and the Briefing does not rule on HMN24 products.
Next week: what a firmness rating measures, and why two mattresses at the same number behave differently under the same body.
Subscribe
Thursday mornings. One claim tested properly, what to do about it, and the position that goes on the Register.
It is the only wellness email I read where I can check the working.Estates director, five-property European hotel group
Every position lands on the Register
No brand is named in a negative finding, and where no trial exists we say never been tested rather than dressing up an absence as caution.