Which Number Tracks Your Recovery?
We're going deep into HRV, and solving the confusion of two HRV numbers once and for all.
Key Takeaways
- Your new watch reports two HRV numbers. Recovery HRV (RMSSD) is Apple's signal for day-to-day stress and recovery. Overall HRV (SDNN) feeds Apple's broader health features.
- RMSSD reads the vagus nerve more directly. It tracks quick beat-to-beat changes and holds up in short recordings. SDNN collects every influence in the window, and it grows as the recording gets longer.
- Same unit, different question. An SDNN of 40 ms and an RMSSD of 40 ms can't be compared, so an older Apple Watch reading won't line up against a friend's Oura or Whoop.
- The clock moves your HRV. Vagal activity peaks at night. A 2 p.m. reading and a 6 a.m. reading describe two different states of the same nervous system.
- Overnight is the easiest window to keep consistent, provided your evenings are consistent too. A late dinner or a glass of wine shows up in the number.
- Read the weekly average. Night-to-night scatter is large. The trend carries the signal.
What Changed With Apple Watch HRV in September 2026?
Apple split HRV into two numbers. On September 9, Apple introduced the Series 12 and Ultra 4, which check heart rate every five seconds and HRV as often as every five minutes, which Apple says is up to 24 times more often than before (Apple Newsroom).
Since HRV first appeared on the watch in 2017, Apple reported it with a single calculation called SDNN. The new watches add a second measurement called Recovery HRV. Apple's press release doesn't name the formula, but Apple told Ray Maker of DC Rainmaker that Recovery HRV, and every HRV value shown in the Vitals app, uses RMSSD (DC Rainmaker). Apple's HealthKit also added a dedicated RMSSD data type this release (Gadgets & Wearables). SDNN stays on as Overall HRV, which Apple positions for broader health insights.
A few details matter for how you read the number:
- Your nightly Recovery HRV is a simple average of the readings taken while you sleep.
- The watch needs seven nights of data, with at least four hours of sleep each night, before HRV trends and the new Readiness score appear.
- That baseline is built per watch. Years of data from a previous Apple Watch don't carry over.
Apple says the Readiness scoring was developed with data from the Apple Heart and Movement Study, working with its own exercise scientists and physicians.
What Is the Difference Between SDNN and RMSSD?
SDNN measures how widely all your beat intervals spread across a recording. RMSSD measures how much each beat differs from the one right before it.
Heart rate variability (HRV): the variation in time from one heartbeat to the next, measured in milliseconds.
Your heart at rest might average one beat per second, but the individual gaps, called interbeat intervals, might run 0.95 seconds, then 1.06, then 0.98. Much of that shifting comes from your autonomic nervous system, the automatic regulator that balances your sympathetic branch, which mobilizes you for effort and threat, against your parasympathetic branch, which governs repair and recovery.
SDNN: the standard deviation of normal-to-normal intervals. How widely every interbeat interval in a recording spreads around the average.
That spread collects everything acting on your heart during the window. Breathing, blood pressure regulation, body temperature, movement and both branches of your nervous system all end up in it. The 1996 Task Force that set the measurement standards the field still follows noted that SDNN grows as a recording gets longer, and cautioned against comparing SDNN values taken over different durations (Task Force of the ESC and NASPE, Circulation).
RMSSD: the root mean square of successive differences. How much each interbeat interval differs from the one before it.
Vagus nerve: the main parasympathetic nerve connecting your brain to your heart. It can lengthen the very next beat within a fraction of a second.
Those quick beat-to-beat changes come mainly from the vagus nerve. Fred Shaffer and J.P. Ginsberg describe RMSSD as the primary time-domain estimate of vagal influence on heart rate (Shaffer & Ginsberg, 2017, Frontiers in Public Health). RMSSD also holds up in short recordings. In 23 collegiate athletes, Michael Esco and Andrew Flatt found a one-minute RMSSD reading agreed closely with the standard five-minute reading (Esco & Flatt, 2014, Journal of Sports Science and Medicine).
For recovery tracking, that combination carries weight. The parasympathetic branch does the repair work after hard training, short sleep or a demanding week, and RMSSD reads that branch more directly than SDNN does. Martin Buchheit's 2014 review and Daniel Plews and colleagues' monitoring work with elite endurance athletes built the daily recovery practice used across sports science on RMSSD for that reason (Buchheit, 2014; Plews et al., 2013).
A long exposure and a burst of frames
Think of SDNN as a long-exposure photograph. Leave the shutter open and everything that moved during the window blurs into one image: your breathing, a shift in position, a thought about tomorrow's meeting. The longer the shutter stays open, the more blur you collect, which is why two SDNN values from different-length recordings can't be compared.
RMSSD works like a burst of fast frames. Each frame is compared only to the one just before it, so it catches the quick adjustments your vagus nerve makes beat by beat. A short burst still tells you something useful.
| SDNN (Overall HRV) | RMSSD (Recovery HRV) | |
|---|---|---|
| What it measures | Total spread of all interbeat intervals in a recording | Change from each beat to the next |
| Main influence | Everything in the window: breathing, blood pressure, temperature, movement, both nervous system branches | Mostly the vagus nerve, the parasympathetic "repair" branch |
| Effect of recording length | Grows as the recording gets longer; don't compare different durations | Stable in short recordings; one minute tracks five minutes closely |
| Best use | Long-term, population-style health measures | Day-to-day recovery tracking |
| On Apple Watch | Every model; labeled Overall HRV on Series 12 and Ultra 4 | Series 12 and Ultra 4 only |
| Other wearables | Rare as the headline number | The standard for Oura, Whoop and Garmin |
Why Does Apple Keep Both HRV Numbers?
Apple keeps SDNN because it carries a longer research history in population health, much of it drawn from 24-hour heart recordings. Apple uses it for Overall HRV and as an input to features such as the Health Age estimate coming in its redesigned Health app.
Independent testing of Recovery HRV is only beginning. In his early Ultra 4 review, Ray Maker found Garmin's readiness score a bit quicker to respond than Apple's new Readiness score, while pointing out that Garmin had years of his data and Apple had a single week. He also noted there is no industry standard for how a readiness score should be built (DC Rainmaker). Fair enough. It is early.
Can You Compare Apple Watch HRV With Oura, Whoop or Garmin?
Only Recovery HRV from a Series 12 or Ultra 4 uses the same formula as those devices, and even then the comparison is loose.
If you own an older watch, you have SDNN only, whether or not you install watchOS 27. It is captured opportunistically when you happen to be still, roughly every two to four hours in Engadget's testing (Engadget). That number can't be lined up against an RMSSD value from your training partner's Whoop or your spouse's Oura ring. Both read out in milliseconds. The calculations answer different questions about your nervous system.
Even two devices using RMSSD will differ in when they sample, how they filter noisy beats and how they average the night. The comparison that holds up is you against yourself, on one device, over weeks.
Does Time of Day Change Your HRV?
Yes, a lot. Your HRV follows a daily rhythm. Parasympathetic activity climbs through the night and falls after you wake, and daytime values shift with posture, meals, caffeine, conversation and movement.
Hendrik Bonnemeier and colleagues documented this pattern in 2003 with 24-hour heart monitoring in healthy adults. Vagal markers of HRV ran highest at night in every age group, and the nighttime rise shrank with age (Bonnemeier et al., 2003, Journal of Cardiovascular Electrophysiology). A reading taken at 2 p.m. and one taken at 6 a.m. catch your nervous system in two different states. Compare them, and the time of day can account for more of the gap than your recovery does.
This is also why Apple's new daytime HRV view is harder to read than the overnight one. Daytime values swing minute to minute, and Apple compares each one against your own history for that same hour of the day.
The athlete monitoring protocols Buchheit and Plews developed fix the conditions for exactly this reason: same time, same position, same routine, day after day. Plews and colleagues also recommend reading a seven-day rolling average instead of any single day, because the day-to-day scatter in HRV is large and the weekly trend carries the signal (Plews et al., 2013). Their later work found at least three valid readings a week were needed for that average to hold up (Plews et al., 2014, IJSPP).
Is Overnight HRV Better Than a Morning Reading?
For most people, overnight is the easier window to keep consistent. There is a real caveat, and it comes from a respected critic.
Night asks the least of you
A measurement you have to remember to take will eventually get skipped. Sleep removes that problem. You lie still for hours, you aren't talking, eating or answering email, and the watch records whether you think about it or not. For a parent with a 6 a.m. carpool, or a grandparent whose mornings vary by the day, the night is a window that repeats itself with no effort on your part.
Laboratory data support the consistency of that window, with limits. David Herzig and colleagues recorded 15 healthy young men across three nights in a sleep lab. During slow-wave sleep, the deepest stage of non-REM sleep, RMSSD showed an intraclass correlation of 0.84, meaning most of the variation in the data reflected real differences between people rather than noise within the same person. SDNN held up reasonably well in deep sleep but poorly during lighter sleep and REM (Herzig et al., 2018, Frontiers in Physiology). The authors also advised averaging across several segments of the night, which is what Apple does. The sample was small, young and male, so the finding describes a controlled lab more than your bedroom.
Field studies point the same direction. Olli-Pekka Nuuttila and colleagues in Finland followed 24 recreational runners, aged 28 to 48, through three weeks of normal training and two weeks of overload. Nighttime HRV dropped after an all-out 3,000-meter test while morning readings didn't, partly because the night came sooner after the test. Across the overload block, only the nighttime changes tracked which runners improved their 3,000-meter time (Nuuttila et al., 2024, Sports Medicine – Open). Christina Mishica and colleagues followed 11 endurance athletes around 16 years old for three weeks and found overnight and morning RMSSD closely correlated, with coefficients near 0.9. That is a small sample, suggesting both windows track the same underlying state (Mishica et al., 2022, PLOS One).
The case for mornings
Marco Altini, who has spent more than a decade building HRV measurement tools and publishing on them, argues that a short reading right after waking is more useful for training decisions. His case is that nighttime numbers carry the fingerprints of your evening, a late dinner, a glass of wine, a hard evening session, and can look suppressed even when you wake up recovered (Altini, Substack). He also points to research in which overtrained athletes showed nighttime HRV similar to healthy controls but clear differences in morning measurements (Hynynen et al., 2006, Medicine & Science in Sports & Exercise).
Both positions rest on the same principle. A measurement tracks recovery only when the conditions around it stay constant, and the night standardizes itself only when your evenings do.
How Should You Read Recovery HRV on Your Apple Watch?
Pick one consistent window, read the weekly average against your own baseline, and keep your evenings steady enough that the night means the same thing each time.
Across decades of coaching, the athletes who gained the most from any monitoring tool shared one habit. They measured the same way every day and read the trend instead of the morning.
| Your watch | What it reports | What I would do |
|---|---|---|
| Series 12 or Ultra 4 | Recovery HRV (RMSSD) overnight and during the day, plus Overall HRV (SDNN) | Make overnight Recovery HRV your primary number. Wear the watch to bed for the first seven nights so the baseline can form, then watch the weekly average against it. Treat the daytime values as context. |
| Series 11 or earlier, Ultra 3 or earlier, SE | Overall HRV (SDNN) only, captured at scattered times | Compare readings taken in a similar window. Or start a one-minute Mindfulness session at the same time each morning, which prompts the watch to record an HRV reading under matched conditions. |
| Either | Compare yourself to yourself, on one device. Don't line up your number against someone else's ring or band. |
Whichever watch you wear, give your evenings a routine that makes your nights comparable. Keep your last meal at a regular time, keep alcohol well away from bedtime, and finish hard training several hours before sleep. When the night stays consistent, a change in your number belongs to your recovery.
One more thing. Apple describes its Vitals app as a wellness feature, not a medical one. If your numbers shift sharply and you also feel unwell, that is a conversation for your doctor rather than your watch.
Your daily habits are the thermostat. HRV is the thermometer. Apple has now put a thermometer tuned for recovery on millions of wrists, and it reads your nervous system accurately only when you take the reading the same way each time.
Go Deeper: The HRV Book
I wrote The HRV Book for people who already wear the device and want to know what the number means. It covers what your score is calculating, why a single night is mostly noise, how many weeks of data it takes before your baseline is worth trusting, and why a ring and a watch can hand you two different numbers for the same heart.
Frequently Asked Questions
What is Recovery HRV on Apple Watch? Recovery HRV is the heart rate variability measure Apple added on the Series 12 and Ultra 4 for tracking daily stress and recovery. Apple told reviewers it uses RMSSD, the same beat-to-beat calculation used by Oura, Whoop and Garmin, and it appears in overnight Vitals against your personal baseline.
Does Apple Watch use RMSSD or SDNN? Both, on the newest models. Series 12 and Ultra 4 report Recovery HRV using RMSSD and Overall HRV using SDNN. Every older Apple Watch reports SDNN only, including models updated to watchOS 27.
Why is my Apple Watch HRV different from my Oura or Whoop number? Older Apple Watches report SDNN while Oura and Whoop report RMSSD, so the two numbers measure different things even though both are in milliseconds. Even devices using the same formula differ in when they sample and how they average the night, so compare each device only with itself.
What is a good Recovery HRV on Apple Watch? There is no universal good number. HRV varies widely between healthy people and declines with age, so the useful comparison is your weekly average against your own baseline. A sustained move away from your normal range means more than any single value.
How long does Apple Watch take to show HRV trends? Seven nights. The Series 12 and Ultra 4 need a week of sleep data, with at least four hours per night, before HRV trends and the Readiness score appear. The baseline is built per watch, so data from a previous Apple Watch doesn't count toward it.
Should I track HRV at night or in the morning? Either works if you keep the conditions constant. Overnight readings need no effort and research shows they respond well to training load, but they reflect your evening habits. A short morning reading avoids that, at the cost of remembering to take it every day.
Sources
- Apple Newsroom. Introducing Apple Watch Series 12, with the all-new Health Sensing System. September 9, 2026.
- Maker R. Apple Watch Ultra 4 In-Depth Review. DC Rainmaker, September 17, 2026.
- Low C. Apple Watch Series 12 review. Engadget, September 2026.
- Apple Watch Recovery HRV and Overall HRV are not the same thing. Gadgets & Wearables, September 11, 2026.
- the5krunner. Apple Watch Series 12 vs Series 11: Full Comparison. September 9, 2026.
- Task Force of the ESC and NASPE (1996). Heart rate variability: standards of measurement, physiological interpretation, and clinical use. Circulation, 93, 1043–1065.
- Shaffer F, Ginsberg JP (2017). An overview of heart rate variability metrics and norms. Frontiers in Public Health.
- Esco MR, Flatt AA (2014). Ultra-short-term heart rate variability indexes at rest and post-exercise in athletes. Journal of Sports Science and Medicine, 13, 535–541.
- Buchheit M (2014). Monitoring training status with HR measures: do all roads lead to Rome? Frontiers in Physiology, 5, 73.
- Plews DJ, Laursen PB, Stanley J, Kilding AE, Buchheit M (2013). Training adaptation and heart rate variability in elite endurance athletes. Sports Medicine, 43, 773–781.
- Plews DJ, et al. (2014). Monitoring training with heart rate variability: how much compliance is needed for valid assessment? International Journal of Sports Physiology and Performance, 9, 783–790.
- Bonnemeier H, et al. (2003). Circadian profile of cardiac autonomic nervous modulation in healthy subjects: differing effects of aging and gender on heart rate variability. Journal of Cardiovascular Electrophysiology, 14, 791–799.
- Herzig D, Eser P, Omlin X, Riener R, Wilhelm M, Achermann P (2018). Reproducibility of heart rate variability is parameter and sleep stage dependent. Frontiers in Physiology, 8, 1100.
- Nuuttila OP, Kyröläinen H, Kokkonen VP, Uusitalo A (2024). Morning versus nocturnal heart rate and heart rate variability responses to intensified training in recreational runners. Sports Medicine – Open, 10, 120.
- Mishica C, et al. (2022). Evaluation of nocturnal vs. morning measures of heart rate indices in young athletes. PLOS One, 17, e0262333.
- Hynynen E, Uusitalo A, Konttinen N, Rusko H (2006). Heart rate variability during night sleep and after awakening in overtrained athletes. Medicine & Science in Sports & Exercise, 38, 313–317.
- Altini M. Heart Rate Variability (HRV) Measurement Timing: Morning or Night? Substack.
This article is for educational purposes and is not medical advice. Apple Watch, HealthKit and watchOS are trademarks of Apple Inc. Mode + Method is not affiliated with or endorsed by Apple.
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.