Biohacking Kompakt

Tip · Tracking

Measuring HRV daily

Heart rate variability is the best marker of recovery.

Measuring HRV daily means recording the intervals between heartbeats in the morning and calculating the rMSSD value from them. This tip is here because hardly any biohacking tool has been tested as cleanly against an ECG — and because precisely these test figures decide which device can support a training decision.

In short

The measurement is as good as its sensor, and that is known in numbers. Against ECG, the mean absolute percentage error for rMSSD was 2.16 % for the Polar chest strap and 17.49 % for a smartphone PPG app; the Apple Watch underestimated HRV by 8.31 ms with a MAPE of 28.88 % and missed the predefined equivalence margin of ± 10 ms. HRV-guided training works, but differently than advertised: the vagally mediated HRV values improved markedly (SMD+ = 0.50), performance did not improve significantly (SMD+ = 0.20). And a single morning’s value does not support a decision.

What lies behind it

The vagus nerve slows the heartbeat from beat to beat, especially in the rhythm of breathing; the sympathetic nervous system acts more sluggishly and in the opposite direction. The rMSSD value — the square root of the mean of the squared differences between successive R-R intervals — therefore mainly reflects parasympathetic activity. Practically all the studies cited here worked with it or its natural logarithm.

Strain lowers vagal activity acutely, recovery raises it again. After a marked increase in time below the first lactate threshold, weekly Ln rMSSD in nine elite rowers changed by a standardized +0.20, after more time above the second threshold by −0.20.

How accurately the devices measure

This is where the substance of this tip lies. In 37 trained participants, ECG, Polar H10 and a smartphone PPG app ran simultaneously. Repeatability within a session was good for all three (ICC for rMSSD 0.83–0.90), agreement with the ECG was not: 2.16 % mean absolute percentage error for the chest strap versus 17.49 % for the app.

For the wrist, the verdict is clearer. In 316 measurements over 14 days, the Apple Watch underestimated HRV by 8.31 ms (p = 0.025), MAPE 28.88 %. The same watch hit resting heart rate almost perfectly: mean difference −0.08 beats per minute, MAPE 5.91 %. The watch is not a bad device, just unsuitable for this one quantity.

What guiding training by it achieves in everyday life

The central controlled study divided 40 recreational runners into an HRV-guided group and a plan-based group. A hard session was only programmed if the morning value was within an individually determined threshold; otherwise training was easy. Result: considerably fewer hard sessions with an at least equivalent result over 3,000 meters.

That is the realistic expectation. The tool does not make you faster; it prevents hard sessions on days when your body is not up to it. Those who have little time or train too much gain from it. Those looking for a shortcut to more performance do not.

What is well supported

The hardest figures are the validation numbers against ECG: chest strap 2.16 % MAPE, smartphone PPG app 17.49 %, Apple Watch 28.88 % with systematic underestimation by 8.31 ms — a difference of more than a factor of ten between devices that appear side by side as equals in recommendation lists. A meta-analysis of 23 studies with 301 effect estimates found a small, significant and highly heterogeneous deviation from ECG for wearable devices overall (ES = 0.23; 95% CI 0.05–0.42; I² = 78.6 %). Equally robust is the question of measurement frequency: the standardized Ln rMSSD changes rose from 0.20 ± 0.28 with one day to 0.49 ± 0.33 with three days and no further after that.

What the studies show

Johansson 2025, Frontiers in Physiology

37 trained participants (17 female, 21.95 ± 3.69 years) with simultaneous recording by ECG, Polar H10 and a smartphone PPG app, two runs on the same day. Repeatability within a session for rMSSD across all three devices ICC 0.83–0.90. Compared with ECG, the mean absolute percentage error for rMSSD was 2.16 % for the chest strap and 17.49 % for the app.

O’Grady 2024, Sensors

The test of the smartwatch over several days: 39 healthy adults, 316 HRV measurements over 14 days, reference Polar H10 with Kubios software. Underestimation by 8.31 ms (p = 0.025), MAPE 28.88 %, mean absolute error 20.46 ms; the equivalence test against ± 10 ms was negative. Resting heart rate was accurate: mean difference −0.08 beats per minute, MAPE 5.91 %.

Vesterinen 2016, Med Sci Sports Exerc

40 recreational runners, eight-week intensive phase, HRV-guided versus fixed plan. The HRV group completed fewer hard sessions (13.2 ± 6.0 versus 17.7 ± 2.5; P = 0.021, effect size 0.98). 3,000 m performance improved by 2.1 % ± 2.0 % versus 1.1 % ± 2.7 %; the group difference remained small (effect size 0.42).

Manresa-Rocamora 2021 and Düking 2021

Two meta-analyses, one picture. For the vagally related HRV indices, the standardized mean difference in favor of the HRV-guided group was SMD+ = 0.50 (95% CI 0.09–0.91), for maximal aerobic capacity 0.20 (−0.07–0.47). The second analysis, across 8 studies with 198 participants, found significance only for submaximal physiological parameters (g = 0.296; p = 0.028), not for performance (g = 0.079) or V̇O2peak (g = 0.171).

What the measurement does not show

The data base for guiding training is small: 8 studies with 198 participants in one meta-analysis, a comparably small number in the other. On hard endpoints such as injury or infection rates, there is no study. The claim that HRV is the best recovery marker also lacks head-to-head evidence: in an 18-week study of 35 endurance-trained men and women, running speed at 90 % of maximal heart rate in a submaximal test was the most sensitive variable, while heart rate recovery, HRV and jumping power showed no significant relationship with performance development.

How to do it

In the morning right after waking up, before getting up, lying down or sitting — the position must stay the same from day to day, because it is itself one of the moderators of measurement deviation. 5 minutes is the standard of the validation studies; one validation found no significant differences between 1 minute and 5 minutes. Device: chest strap. Frequency: at least 3 valid measurements per week as a weekly average; for nightly wearable values, at least 5 out of 7 nights. Before the first decision, you need your own baseline — in the study with sedentary adults, measurement alone was carried out for two weeks before the eleven-week intervention. And the decision rule of the only larger controlled study was not rest but swapping intensity.

Safety

The measurement itself is risk-free; risks arise solely from interpretation. HRV is not a diagnostic tool, and a deviating HRV on its own says nothing about the heart or circulation. Anyone who notices an irregular pulse, palpitations, shortness of breath or chest pain should have it checked by a physician, not by an app — an arrhythmia distorts the values and can be misinterpreted as poor recovery. Two groups should treat the number with particular caution: highly endurance-trained people, in whom HRV can fall despite rising parasympathetic activity — measured in three out of four world champion rowers — and people prone to self-tracking, for whom a low morning value itself becomes a stressor. Beta blockers, alcohol the evening before, infections and the time of day change the value.

BK-Score Thin human evidence

Human evidence5
Mechanism7
Safety data10
Hype gap4
Track record of use8

Heart rate variability demonstrably reflects autonomic balance, and HRV-guided training shows advantages in small sports studies. That a single daily value from the wrist provides useful guidance, on the other hand, is not established – the day-to-day fluctuation is large.

The score rates the state of knowledge, not the effect. “Safety data 9” means well studied – not harmless.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)

Frequently asked questions about measuring HRV daily

Is my Apple Watch enough, or do I need a chest strap?

For resting heart rate the watch is enough; in the validation it was off by 0.08 beats per minute on average. For HRV it is not: the same study found a systematic underestimation of 8.31 milliseconds with a mean error of almost 29 percent, and the predefined equivalence margin of ten milliseconds was missed. A chest strap was at 2.16 percent against ECG. If you want to base a decision on the value, use the strap.

How long do I have to measure before I can read anything?

Two weeks of daily measurement as a baseline is what the intervention studies put in front before applying any decision rule at all. After that: the single value is noise, the weekly average of at least three valid measurements is the signal. Anyone who measures once a week has, according to the analysis by Plews and colleagues, practically no information about their training status.

My value is markedly lower today than yesterday – should I take a break?

The study that came closest to examining this question did not take breaks but swapped: with a value outside the individual normal range, there was an easy session instead of a hard one. The result was an equally good or slightly better 3,000-meter time with considerably fewer hard sessions. You should ignore a single outlier, because alcohol, lack of sleep, an incipient infection and the measurement itself can produce it.

Does a high HRV mean I am fit?

Not necessarily. The absolute value depends strongly on age, sex, body weight and measurement method. In highly endurance-trained people it can even fall while parasympathetic activity rises; this was measured in three out of four world champion rowers. What is meaningful is the change relative to your own baseline, not the comparison with others or with an app scale.

Why does my value fluctuate so much?

Because fluctuation is the normal state. In the analysis of around two million nights, the night-to-night fluctuation was itself a meaningful indicator: it was higher with more alcohol, less exercise, and shorter and more irregular sleep. So the fluctuation is not a measurement error; it reflects how irregular your everyday life is.

Does HRV-guided training bring more performance than a normal plan?

According to two meta-analyses: not measurably more performance, but the same or slightly better development with fewer hard sessions. The only statistically established advantage concerned submaximal physiological parameters with g = 0.296 and the vagally mediated HRV values themselves with SMD 0.50. Maximal oxygen uptake and competitive performance did not differ significantly.

Related

Sources

Open in the database – with all tips, filters and BK-Score (German app)

Information only, not medical advice and not a usage recommendation. If you have pre-existing conditions or before major changes, consult a physician. Last updated: 2026-09-13.