How to measure HRV without a wearable

Three ways to get the signal

1. Phone camera

Several apps read pulse by having you rest a fingertip over the rear camera lens with the flash on. The camera sees the blood-volume pulse in the fingertip and the app extracts beat intervals from it. Validation work has found that at rest, with the hand still, smartphone photoplethysmography agrees closely with chest-strap and ECG RMSSD. It is the cheapest option, costs nothing if you already own the phone, and is accurate enough for personal tracking. Its weakness is motion: a fidgety minute produces garbage, and the app may not tell you.

2. Bluetooth chest strap

A chest strap reads the heart's electrical signal directly, which is what clinical HRV is measured from. Paired with any of the common HRV apps, it gives the most reliable consumer reading available, and it is far cheaper than a ring or a smartwatch. If you are going to buy one thing for this purpose, buy the strap.

3. Your own pulse and a clock

You cannot compute RMSSD by hand, but you can observe the phenomenon behind it. Sit quietly, find your pulse, and breathe slowly — in for five seconds, out for five. In most healthy people the pulse perceptibly quickens on the in-breath and slows on the out-breath. That swing is respiratory sinus arrhythmia, the main component of resting HRV. It is not a measurement, but it is a useful demonstration that the number on everyone's dashboard is describing something you can feel.

The morning protocol

HRV is exquisitely sensitive to conditions. Two readings taken under different conditions are not comparable, which means the entire value of self-measurement depends on holding the conditions fixed. The standard protocol used in research and by most apps:

1. Immediately on waking, after using the bathroom, before coffee, food, phone-scrolling or conversation.
2. Same posture every day. Seated is the most common choice. Supine readings run higher, standing readings lower; pick one and never change it.
3. One to two minutes of settling, then a recording of at least one minute — most apps use one to three. Five minutes is the research standard but rarely necessary for tracking.
4. Breathe normally. Do not pace or deepen your breathing during the recording; slow breathing inflates RMSSD and you will be measuring your breathing, not your recovery.
5. Still hands, still body, closed or softly focused eyes. No talking.
6. Every day, or at least five days a week. The value is in the trend, and a trend needs points.

Six things that ruin a reading

• Measuring at different times. HRV has a strong circadian rhythm; an evening reading and a morning reading are different quantities.
• Moving. Motion artefacts almost always push RMSSD up, sometimes by a factor of two. A single implausibly high day is usually this.
• Controlling the breath. See above.
• Measuring after a stimulus. Caffeine, a cold bathroom, an alarm you lunged for, an alarming email — all shift the reading for many minutes.
• Reading single days. Normal day-to-day variation within one person is around 10–20%. Use a seven-day rolling average.
• Switching sensors or apps. Different hardware and different artefact filters give different numbers. Any change starts a new dataset.

Reading the result

After two or three weeks you will have a baseline — your own typical range under your own protocol. From then on the useful questions are about deviation from it. A morning reading well below your baseline, with no obvious cause, often precedes illness or signals under-recovery; a sustained drift downward over several weeks, with the protocol held constant, is worth taking seriously as a sign of accumulating load. A reading above baseline is generally reassuring and otherwise uninformative.

For locating yourself against the population, see the reference ranges by age — with the caution that seated morning readings are what those tables describe, which makes this method, unlike overnight wearables, directly comparable to them.

If you would rather not measure at all

You do not need HRV to assess stress. The PSS-10 has forty years of validation behind it and asks the question HRV only infers: how overloaded, unpredictable and uncontrollable has the last month felt? HRV adds a physiological cross-check — useful, non-specific, and optional. The Calmspan test works without it, and the questionnaire half of the result does not change whether you enter an HRV value or not.