Why Your Wearable's Readiness Score Is Wrong (And What to Track Instead)
Disclaimer: This content is for informational purposes only and is not medical advice. Consult your healthcare provider before starting any supplement.
A readiness score under 60 on a normal night's sleep, no illness, no unusual training load — and no clear explanation — is one of the most common frustrations reported by wearable users, and it's rarely a sign that something is actually wrong with recovery. It's usually a sign that the algorithm behind the score is reacting to something other than recovery: alcohol from two nights ago, a late dinner, a new medication, a loose ring fit, or simply the device's own baseline recalibrating after a change in routine. Readiness and recovery scores from Oura, Whoop, and Garmin are built on real physiological signals — heart rate variability (HRV), resting heart rate (RHR), and sleep staging — but the single number they output compresses those signals in ways that hide exactly what moved and why. Treating that number as a verdict on the day, rather than as an approximation with known blind spots, is the mistake worth fixing.
This guide covers what these scores actually measure, the specific things that skew them without reflecting real recovery status, and which underlying metrics are worth tracking directly instead of the single composite number.
What a Readiness Score Actually Averages Together
Every major wearable's readiness or recovery score is a weighted composite, not a direct measurement of anything. Oura's Readiness Score blends HRV balance, resting heart rate, body temperature deviation, sleep balance, previous day activity, and a few other sub-scores into one number. Whoop's Recovery score leans more heavily on HRV and RHR relative to your own rolling baseline, plus sleep performance. Garmin's Body Battery estimates energy reserves from HRV, stress data, and activity. The exact weighting is proprietary and differs by device, which is the first reason two wearables on the same body on the same night can disagree by 20+ points — they're not measuring the same thing, they're just each doing their own version of averaging similar inputs.
The practical implication: a low composite score tells you that something pulled the average down, not what. A night where HRV was fine but body temperature ran half a degree high from a late glass of wine can produce the same readiness number as a night where HRV genuinely dropped from overtraining — and those two situations call for opposite responses. One is "train as planned," the other is "back off." The composite score can't tell you which one you're looking at.
HRV Is the Most Informative Input and the Most Easily Confounded
Heart rate variability — the variation in time between heartbeats — is the input doing the most work in most readiness algorithms, and it's also the most sensitive to things that have nothing to do with training recovery. Research on HRV measurement consistently identifies several non-training factors that move HRV as much as or more than actual physiological strain: alcohol (even one drink can suppress HRV for the following one to two nights), a large or late meal before bed, dehydration, an irregular sleep schedule, illness onset before symptoms appear, and psychological stress unrelated to exercise. Menstrual cycle phase is another significant confounder for women — HRV naturally runs lower in the luteal phase, which is a normal hormonal pattern, not a recovery deficit, but most consumer algorithms don't adjust their baseline for cycle phase.
Ring and strap fit adds a purely mechanical source of noise. A loose-fitting ring or strap, or one worn on a different finger or wrist position than usual, degrades the optical sensor's ability to detect the heartbeat waveform cleanly, which shows up as noisier HRV data rather than an error message. Anyone who's seen an implausible overnight HRV swing — a jump or drop far outside their normal range with no clear cause — has usually just seen a fit or positioning problem, not a genuine physiological event.
Resting Heart Rate: Useful, But Slower to Reflect Real Change
RHR is the more stable of the two core inputs and, for that reason, often the more trustworthy one — but it responds more slowly, which means it's a poor same-day signal and a better multi-day trend signal. A single elevated RHR reading is common and often meaningless (room temperature, a late workout, a full bladder at the time of measurement). A RHR that trends upward for three or more consecutive days, especially paired with a downward HRV trend, is a considerably stronger signal of accumulating fatigue or an oncoming illness than either metric alone on a single night. The mistake is treating one night's RHR bump the way the app's readiness score does — as an immediate, actionable data point — when the more reliable use of RHR is as a slow-moving trend line checked weekly, not daily.
Sleep Staging Accuracy Is the Weakest Link
Sleep stage classification (light, deep, REM) from wrist- or finger-based wearables is the least accurate component feeding most readiness scores, and it's worth knowing that going in. Validation studies comparing consumer wearables against polysomnography (the clinical gold standard, which uses EEG to directly measure brain activity) consistently find wearables are reasonably good at distinguishing sleep from wake, and much less accurate at distinguishing sleep stages — deep and REM sleep are the categories most often misclassified, because the wearables are inferring stage from movement and heart rate patterns rather than measuring brain activity directly. A readiness score that weights "insufficient deep sleep" heavily is, in part, weighting a measurement with a real error rate — which is a reason to treat a single night's deep sleep percentage as a rough estimate, not a precise clinical number, and to look at how it trends over weeks rather than reacting to one low night.
What to Track Instead of the Single Score
None of this means the underlying data is useless — it means the single composite number is the wrong level to act on. A more reliable approach:
- Track HRV and RHR trends over 7-14 days, not single nights. A gradual drift in either direction is meaningful; a one-night swing usually isn't.
- Log the confounders alongside the data — alcohol, late meals, unusual stress, menstrual cycle phase, illness symptoms — so a low score has an obvious explanation instead of looking like an unexplained recovery deficit.
- Weight RHR trend more heavily than HRV for same-week decisions, since it's less noisy night to night, and reserve HRV for longer trend confirmation.
- Treat one bad night's score as noise until it repeats. Two or three consecutive low readings with no obvious explanation is a real signal worth acting on; one is usually not.
- Check ring or strap fit before troubling over an outlier reading, especially if the number is wildly outside your normal range.
Where Diet Consistency Fits In
Day-to-day diet variability is an underappreciated source of readiness noise that's easy to rule out cheaply. Late, heavy, or alcohol-paired meals measurably suppress overnight HRV independent of training load, and inconsistent micronutrient intake — skipping meals on busy days, running low on magnesium or electrolytes — shows up in RHR and sleep quality data without an obvious cause in the app. AG1 is a reasonable way to keep a baseline nutritional floor consistent on the days a real meal doesn't happen, which removes one common, non-training explanation for a noisy readiness number before assuming it reflects real recovery status.
Affiliate Disclosure: This article may contain affiliate links. If you make a purchase through these links, we may earn a small commission at no extra cost to you. We only recommend products we genuinely believe in. This helps support our work and allows us to continue providing free content.
Common Questions
Why does my readiness score sometimes contradict how I actually feel? Because the algorithm is averaging inputs — temperature deviation, sleep staging estimates, HRV, RHR — some of which are measurement-error-prone (sleep staging) or confounded by non-training factors (HRV). Subjective feel is itself a real data point and, for same-day training decisions, often at least as reliable as the composite score.
Should I ignore my readiness score entirely? No — the trend over one to two weeks is genuinely useful for spotting accumulating fatigue or an oncoming illness before it's obvious otherwise. The issue is treating a single night's number as a precise, standalone verdict rather than one noisy input to a slower-moving trend.
Does alcohol really affect HRV that much? Yes — research on alcohol and HRV consistently shows measurable suppression the night of consumption and often into the following night as well, roughly proportional to amount consumed. It's one of the most reliable single-night explanations for an unexpectedly low score.
Why is my HRV lower than my friend's even though we train similarly? HRV is highly individual — normal ranges vary widely between people based on age, genetics, and baseline autonomic nervous system tone. Comparing your absolute HRV number to someone else's is not meaningful; comparing your own number to your own rolling baseline is what the algorithms are actually designed to do, and what's worth paying attention to.
Do readiness scores adjust for menstrual cycle phase? Most mainstream algorithms do not meaningfully adjust their baseline for cycle phase as of 2026, despite HRV reliably running lower during the luteal phase for many women. Logging cycle phase alongside readiness data is currently the more reliable way to avoid misreading a normal hormonal pattern as a recovery problem.
Last updated: 2026-07-16
Want the evidence-based breakdown before you buy anything? Subscribe to the VitalStack newsletter — we cut through the marketing and tell you what the research actually supports: