Heart rate variability (HRV): what a low reading means
HRV stands for heart rate variability: the changes in the time between one heartbeat and the next. A healthy heart is not a metronome, and those intervals keep shifting to answer whatever the body asks of it (Shaffer & Ginsberg, 2017). A low value says today's variation is smaller than yours usually is. The cause has to be looked for separately.
RMSSD and SDNN are different numbers
Both come in milliseconds and both get called HRV. SDNN is the standard deviation of the intervals between normal beats: how spread out they are around their average. RMSSD is the square root of the mean of the squared differences between each interval and the next: how much one beat changes from the one before.
Two made-up series of six intervals, with the same average and the same range, give very different results.
| Series | Intervals | Average | SDNN | RMSSD |
|---|---|---|---|---|
| A, alternating | 800, 850, 800, 850, 800, 850 | 825 | 27 | 50 |
| B, rising slowly | 800, 810, 820, 830, 840, 850 | 825 | 19 | 10 |
A real recording has hundreds or thousands of beats, but the mechanism is the same: RMSSD reacts to change from one beat to the next, while SDNN also picks up slow drift. So a 40 from an app that reports RMSSD and a 40 from one that reports SDNN do not mean the same thing. Apple Watch stores SDNN (Apple Developer, SDNN), and the guide to HRV on Apple Watch has the detail.
Why to compare with yourself and not with other people
Shaffer & Ginsberg, 2017, gather published reference values and warn about what matters: your baseline depends on the measurement context, including recording length, age and sex, and norms for 24 hours, for about 5 minutes and for under 5 minutes are not interchangeable. If you come across a chart of ‘normal HRV by age’, check which metric it uses, how long the recording was and who was measured. This guide leaves such a chart out on purpose.
Plews et al., 2013, reviewed HRV in elite endurance athletes. They conclude that each athlete has to be followed over time to understand their individual ‘fingerprint’. In that group, both rises and falls in HRV have been linked to poor adaptation, and gains in fitness have been seen alongside falling HRV. We do not know how much of this carries over to someone training in an ordinary gym, but it makes plain that the direction of a change cannot be read on its own.
Flatt & Esco, 2016, look at another piece: whether the reading is comparable. In 20 collegiate cross-country athletes, lying down and on an ECG, each of the first five minutes gave an lnRMSSD close to the reference (ICC 0.92 to 0.97), which supports settling for just one minute. It does not compare people with each other, and an ECG is not a watch. The practical conclusion is ours: measure the same way each day, in the same position and at the same time.
A worked example with your own average
Seven mornings for one person, with made-up numbers: 44, 41, 46, 38, 43, 45 and 31 ms. The average of the first six days is 43 ms. Day seven drops to 31, yet the seven-day average only falls to 41 ms.
A 31 is low for this person, though someone averaging 28 would call it a good day. And one morning barely moves the average. Plews et al., 2013, propose appropriate averaging for daily monitoring. If the 31 repeats for several days and the average keeps falling, then there is something to look at.
What can move it
This is what we could check at the source. Nearly all of it is association, not cause.
| Factor | What the source says | Source |
|---|---|---|
| Alcohol | 4,098 Finnish employees, each compared with themselves. A recovery index derived from HRV, over the first 3 hours of sleep, fell on average by 9.3, 24.0 and 39.2 percentage units with low, moderate and high doses. This is an observational study. | Pietilä et al., 2018 |
| Psychological stress | A review of 37 publications: in most, HRV variables changed in response to induced stress. | Kim et al., 2018 |
| Short sleep | Less clear than people tend to say: in 43 healthy young adults, HRV did not differ significantly between the groups with and without sleep deprivation. A small study. | Vierra et al., 2022 |
| Medication, altitude, illness | Apple lists these among the things that can affect the metrics in its Vitals app. It does not give a direction. | Apple Support, Vitals |
| Pregnancy | Across 757 pregnancies followed with Apple Watch, the median went from 39.9 ms before pregnancy to 29.9 ms in the third trimester. The authors present this as data for understanding normal pregnancy physiology. | Mishra et al., 2024 |
What a low reading does and does not tell you
What it tells you
- That today, on this measurement, you are below your own level.
- That it is worth going over how you slept, what you drank, how much you trained and how you feel.
What it does not tell you
- The cause: very different reasons can give the same reading.
- That you are ill, or that you are well.
- That you must rest: in elite athletes, better fitness has been seen with HRV going down (Plews et al., 2013).
A rule of thumb, which is a convention and not a study finding: look at the seven-day average before today's figure. If a low day comes with feeling unwell, reduce the load. If you feel fine, do your warm-up and decide on the first sets, for instance with RPE and RIR.
GoForIt, which is not out yet, uses HRV as one of the signals in its daily estimate and always compares it with your own range, never a generic average. The guide to recovery scores covers how it combines with sleep and load.
Frequently asked questions
- What does low HRV mean during sleep?
- That your heartbeats varied less that night than they usually do across your own nights. Alcohol is one of the explanations that has been studied: in 4,098 people, drinking was linked to poorer regulation in the first hours of sleep, more so at higher doses (Pietilä et al., 2018). One night alone does not point to a cause, so look at several.
- What are normal heart rate variability values?
- No single figure fits everyone. Published values depend on recording length, age and sex, and norms for 24 hours and for a few minutes cannot be swapped (Shaffer & Ginsberg, 2017). The useful reference is your own average over several days, always measured the same way.
- Is a higher HRV always better?
- Not always. In elite endurance athletes, both rises and falls have been linked to poor adaptation, and falls in HRV have been described alongside a lower resting heart rate (Plews et al., 2013). These are very particular athletes, but it is a useful warning: read a change in its context.
- Should I stop training if my HRV is low today?
- One reading does not force anything. If you feel fine, you can train and adjust according to how the first sets go. If you feel unwell or have symptoms such as dizziness, palpitations or chest pain, do not train, and speak to a doctor.
Sources
- Shaffer & Ginsberg, 2017. An Overview of Heart Rate Variability Metrics and Norms. Frontiers in Public Health, 2017. 10.3389/fpubh.2017.00258
- Plews et al., 2013. Training adaptation and heart rate variability in elite endurance athletes: opening the door to effective monitoring. Sports Medicine, 2013. 10.1007/s40279-013-0071-8
- Flatt & Esco, 2016. Heart rate variability stabilization in athletes: towards more convenient data acquisition. Clinical Physiology and Functional Imaging, 2016. 10.1111/cpf.12233
- Pietilä et al., 2018. Acute Effect of Alcohol Intake on Cardiovascular Autonomic Regulation During the First Hours of Sleep in a Large Real-World Sample of Finnish Employees: Observational Study. JMIR Mental Health, 2018. 10.2196/mental.9519
- Kim et al., 2018. Stress and Heart Rate Variability: A Meta-Analysis and Review of the Literature. Psychiatry Investigation, 2018. 10.30773/pi.2017.08.17
- Vierra et al., 2022. Effects of sleep deprivation and 4-7-8 breathing control on heart rate variability, blood pressure, blood glucose, and endothelial function in healthy young adults. Physiological Reports, 2022. 10.14814/phy2.15389
- Mishra et al., 2024. Trends in sensor-based health metrics during and after pregnancy: descriptive data from the apple women's health study. AJOG Global Reports, 2024. 10.1016/j.xagr.2024.100388
- Apple Support, Vitals. Track your overnight and daytime vitals with Apple Watch. Apple Support, 2026. support.apple.com/en-us/120142
- Apple Developer, SDNN. heartRateVariabilitySDNN. Apple Developer Documentation, read 2026-11-02. developer.apple.com/documentation/healthkit/hkquantitytypeidentifier/heartratevariabilitysdnn
