When a good HRV reading is not good news

Short answer

Nearly every app treats higher heart rate variability as better recovery, and for most people most of the time that holds. But if you live with ME/CFS, Long COVID, POTS or fibromyalgia, a calm-looking reading can appear during a crash rather than after recovery from one — and acting on it can cost you the next few days. This is not a claim that high HRV means you are unwell. It is a caution that a single number, read through a "higher is better" frame, is not a safe basis for deciding whether to push.

Note: this article is for general wellness education. It is not medical advice, and heart rate variability cannot diagnose any condition.

Do you need a wearable for this?

No. These readings can be taken with a phone camera alone, which matters when buying hardware is not a given — see measuring HRV without a wearable. A wearable adds overnight data, which is genuinely useful here, but it is not the entry ticket.

Where "higher is better" comes from, and where it stops

Heart rate variability reflects the balance between the two branches of your autonomic nervous system. Broadly, more variability is associated with parasympathetic activity — the rest-and-recover side — and less is associated with sympathetic activation, load, illness or poor sleep. That relationship is well established and it is why every consumer app builds a readiness or recovery score on top of it.

The framing that gets bolted on top — higher is better, lower is worse — is a simplification that works for training load in healthy bodies. It was developed largely in sports physiology, on athletes, and validated in that population.

Energy-limiting conditions break the assumption underneath it. In ME/CFS, reduced heart rate variability has been shown to track with fatigue severity, and abnormally blunted heart rate responses to exertion are a documented feature — the heart does not rise the way it should when effort is applied. In Long COVID, wearable HRV monitoring has been used to identify autonomic dysfunction and to look for thresholds associated with post-exertional malaise. Across these conditions the autonomic system is not simply "more stressed" or "more recovered". It is responding differently.

When a system's response is blunted, a quiet number can mean two very different things: a body that has settled, or a body with less left to respond with. Those look similar on a chart and feel nothing alike.

The pattern people actually report

Our own guidance to Welltory users says this plainly, and it is worth repeating here:

For some people with ME/CFS, POTS, Long COVID or fibromyalgia, a spike in HRV — or an unusually good-looking reading — can show up right in the middle of a crash, rather than during real recovery.

Several mechanisms are plausible, and we want to be clear that they are proposed explanations rather than settled findings. The heart's response to effort can be blunted, so the expected rise does not appear. The parasympathetic branch can reactivate slowly and incompletely after a load, producing an unusual pattern rather than a clean recovery curve. A stress-hormone response that has already been spent can flatten out afterwards.

A related observation: a low resting heart rate during a crash is not automatically the athletic kind of low. In an energy-limiting condition it can mean several systems — cardiac response, stress hormones, autonomic reset — have each hit their own ceiling and have less left to give.

None of this is a diagnostic sign, and none of it should be read as "high HRV is bad". It is a reason not to treat a good-looking number as permission.

Why almost no app tells you this

Consumer HRV products are built on a readiness metaphor: one number, one direction, one instruction. Green means go. That design is genuinely useful for its original audience — people managing training load who want a fast, unambiguous signal.

It becomes a problem in three specific ways for anyone pacing an illness.

A single number hides the shape. "High HRV" can describe a well-recovered system or a blunted one. Splitting a reading into how activated the system is, how readily it settles, and how well the two are coordinated keeps that distinction visible instead of averaging it into one score — that is what a Nervous System Snapshot shows.

Overnight-only measurement misses the day. Post-exertional malaise typically arrives hours to a day after the effort that caused it. A number computed from last night describes last night. A reading you can take after an activity, and again two hours later, describes the thing you actually want to know about.

Fixed morning protocols were designed for other bodies. Measuring the instant you wake assumes you wake in a comparable state. Many people with ME/CFS describe needing thirty to ninety minutes to come online. In POTS, body position alone changes autonomic measures enough that the postural change is used diagnostically in clinical settings. A protocol that assumes a stable, standardised morning excludes exactly the people it would help most.

What we are not claiming. That Welltory detects crashes, predicts post-exertional malaise, or can tell you whether a reading reflects recovery or depletion. It cannot. What it can do is show you more than one dimension of a reading, let you measure when it is useful rather than only at dawn, and be honest that a good-looking number is not a green light.

How to use HRV when you are pacing, carefully

This is not a protocol and it is not medical advice. It is how to avoid the specific mistake of trusting a number over your body.

Never let a reading override your symptoms. If the number looks good and you feel terrible, the number is the less reliable of the two. It is one physiological signal measured over a minute; you have the whole rest of the evidence.

Watch your own pattern, not population norms. Healthy HRV values span an enormous range between individuals — from under 20 to over 200 milliseconds depending on the metric and the person. Comparing yourself to a normal range tells you almost nothing. Comparing today to your own last three weeks tells you something.

Measure around events, not only at fixed times. Before and after a shower, an errand, a conversation, a short walk. Pairs of readings around an activity say more about that activity than any single daily value.

Expect the delay. Post-exertional malaise commonly appears hours or a day after the trigger. A reading taken immediately after activity may look unremarkable and mean nothing about what tomorrow holds.

Log what you did. The value of any of this comes from the pairing with context. A number without a note is a number you cannot interpret later.

Treat unfamiliar readings as questions, not answers. A surprising value is a prompt to look at the rest of the week, not a verdict about today.

What this looks like in each condition

These conditions are not interchangeable, and the reasons a reading can mislead differ between them. None of the following is diagnostic, and none of it replaces a clinician who knows your case.

ME/CFS. The defining feature is post-exertional malaise: a disproportionate worsening after exertion that is often delayed by hours or a day. Reduced heart rate variability has been shown to track with fatigue severity in this population, and blunted heart rate responses to exertion are documented — the heart does not rise the way it should when effort is applied. A measurement taken immediately after activity can therefore look unremarkable and say nothing about what the next two days will cost. Many people also describe needing thirty to ninety minutes after waking before their body is representative of anything, which makes a strict wake-up protocol actively misleading.

POTS. Here posture is not a confounder to be controlled, it is the central phenomenon. The change from lying to standing produces heart rate and autonomic differences large enough to be used diagnostically in clinical settings. That has a practical consequence for anyone tracking at home: a reading is only comparable to another reading taken in the same position, and recording the position matters as much as recording the value.

Long COVID. Wearable heart rate variability monitoring has been used to identify autonomic dysfunction in this group and to look for thresholds associated with post-exertional malaise. That work is early. It is a reasonable basis for taking autonomic measurement seriously in Long COVID, and not a basis for any app claiming to warn you.

Fibromyalgia. Pain, non-restorative sleep and autonomic symptoms overlap, and morning stiffness and pain can themselves drive sympathetic activation. A morning reading in a body that is in pain is measuring the pain as much as anything else.

What is common to all four. The autonomic system is not simply shifted up or down. It is responding differently, and a framework built to grade recovery in healthy athletes does not map onto that cleanly.

Why the delay is the hardest part to track

The single feature that makes post-exertional malaise difficult to measure is that it is delayed. The effort happens on Tuesday; the consequence lands on Wednesday evening or Thursday morning. Almost every tracking habit people adopt is built around same-day feedback, and same-day feedback is exactly what this does not give.

Three practical consequences follow.

A reading taken right after activity is not a verdict on that activity. It describes the minute it covered. Whether the walk was affordable is a question answered two days later, by symptoms.

Your log needs the lag built into it. If you note only what you did and how you felt that evening, you will keep failing to connect causes to effects that are forty-eight hours apart. Recording both the activity and the following two days is what makes the pattern legible at all.

Averages across a week can hide it. A weekly mean smooths precisely the peaks and troughs that carry the information. Look at the sequence, not the summary.

What to bring to an appointment

If your own tracking is telling you something, the way to make it useful clinically is to bring the pattern rather than the numbers.

Bring a timeline, not a spreadsheet. What you did, when, and what followed — including the delay. A clinician can work with "I walked for twenty minutes on Tuesday and was in bed Thursday and Friday" far more easily than with a chart of RMSSD values.

Note your positions. If you are tracking anything postural, say which position each reading was taken in. Without that, the numbers are not comparable even to each other.

Say what the app is and is not. It is a consumer wellness measurement from an optical sensor, not an ECG. Being upfront about that makes the rest of what you bring more credible, not less.

Flag the things that need a person. Fainting or near-fainting, chest pain, an irregular heartbeat that does not settle, breathlessness at rest, or a resting heart rate that has changed sharply and stayed changed. These are not tracking questions.

What we would need before claiming more

We would rather be explicit about the evidence gap than quietly imply it is closed.

Validation in this population. Essentially all consumer HRV validation has been done on healthy or athletic adults. Nobody has established how well optical measurement performs specifically in people with dysautonomia, tremor or poor peripheral circulation — the very features that degrade an optical signal. Until someone does, published accuracy figures should be read as an upper bound.

Prospective thresholds. Identifying a pattern after a crash is far easier than predicting one before it. A warning system would need a prospective study with proper handling of false alarms, and that does not currently exist in any consumer product.

Condition-specific interpretation. The interpretation layers in every app on the market, ours included, were built on population data dominated by healthy users. Interpreting a reading differently for someone pacing an illness is a real research problem, not a copy change.

Until those exist, the honest use of a physiological measurement here is as one signal among several, and never as the one that overrules how you feel.

What heart rate variability cannot do here

• It cannot diagnose ME/CFS, Long COVID, POTS, fibromyalgia or anything else. There is no HRV pattern that establishes any of these conditions.

• It cannot predict a crash. Research is looking for thresholds associated with post-exertional malaise; that work is early, and nothing in a consumer app constitutes a validated warning system.

• It cannot separate a settled system from a depleted one on its own. That is the whole point of this article.

• It is not an electrocardiogram, it does not detect arrhythmias, and it does not replace clinical assessment.

• Accuracy is weaker in exactly this population — the detail is in how accurate HRV from a phone camera is. Most consumer HRV validation was done on healthy, often athletic adults. Dysautonomia, tremor and poor peripheral circulation all degrade optical signals, and lower absolute HRV values mean a smaller dynamic range and a larger relative error.

When to see a clinician rather than an app: fainting or near-fainting, chest pain, a racing or irregular heartbeat that does not settle, breathlessness at rest, or a resting heart rate that changes sharply and stays changed. A worsening pattern of crashes is also a reason to speak to someone, not to measure harder.

Where a measurement fits inside pacing

Pacing is a behavioural practice, not a physiological one. The core of it is staying inside your energy envelope — the amount of activity you can do without triggering a worsening — and that envelope is defined by what happens afterwards, not by any number taken during.

Which means the honest role of a physiological reading here is narrow and specific.

It cannot set your envelope. Only the pattern of what you did and what followed can do that, and it takes weeks of notes rather than measurements.

It can add a second, non-subjective record. On days when your own perception is unreliable — and brain fog makes that common — having an objective log alongside your notes is worth something, even if neither alone is conclusive.

It can make a pattern visible sooner. A drift across two weeks is easier to spot in a series than in memory, particularly when each individual day felt unremarkable.

It should never be the deciding vote. If the reading says one thing and your body says another, the body is the more reliable witness. This is the opposite of how these products are usually framed, and it is the single most important thing on this page.

A reasonable way to hold all of that: keep the measurement as a column in your log next to what you did and how you felt, look at it weekly rather than daily, and let it raise questions rather than settle them.

Discounts for blog readers: up to 36% off

See what affects your energy, stress, sleep, and daily state with Welltory

Explore your data