Recovery Heart Rate

Recovery heart rate, usually called heart rate recovery (HRR), is how many beats per minute your heart rate falls in the first one or two minutes after you stop exercising. If your heart rate is 165 bpm when a hard effort ends and 138 bpm one minute later, your one-minute recovery is 27 bpm.

The number is simple to read and hard to measure well. Among patients sent for clinical exercise tests, a first-minute drop of 12 bpm or less predicted death over the following six years, independent of exercise capacity and of signs of coronary disease on imaging (Cole et al., 1999). The same drop in your own watch data can also come from a gentle finish, a sloppy wrist reading or a hard training block. This entry covers what the number measures, what the research found, what shifts it, how Apple Watch records it as Cardio Recovery, and how Titan uses it.

01How it is calculated

Heart rate recovery is a subtraction. Take your heart rate at the moment exercise stops, then subtract your heart rate a fixed time later.

Heart rates in the example are illustrative.

A larger number means a faster fall. The two windows measure slightly different things. During exercise your vagus nerve, the parasympathetic brake on the heart, is pulled back and sympathetic drive rises. When you stop, the brake comes back on quickly. Imai and colleagues (1994) blocked each branch of the nervous system with drugs and found that atropine, which blocks the vagus, slowed the heart rate decline in both the first 30 seconds and the first two minutes. The first 30 seconds was almost independent of exercise intensity and of sympathetic blockade. The two-minute decline depended on sympathetic activity and on how hard the exercise had been.

The first minute is dominated by that fast vagal phase, which makes it the cleaner read of vagal reactivation. The second minute carries more of the sympathetic side. Recovery is also held back by the metaboreflex, a signal from acidic, fatigued muscle that keeps heart rate raised. Buchheit (2014) puts it plainly: "The greater the relative exercise intensity, the greater the blood acidosis and metaboreflex stimulation, and the slower the HRR."

02Recovery heart rate and resting heart rate

Resting heart rate is a level, measured when you have been still for a while. Heart rate recovery is a rate of change, measured in the minutes after effort. Resting heart rate depends on the structure of your heart, your blood volume and your autonomic balance (Buchheit, 2014). Recovery reflects how quickly the vagal brake reengages after exercise has released it.

Both numbers carry independent risk information. Jouven and colleagues (2005) followed 5,713 working men aged 42 to 53 for 23 years after a graded exercise test. A resting heart rate above 75 bpm carried a relative risk of sudden death of 3.92. A drop of less than 25 bpm in the minute after exercise carried a relative risk of 2.20. Both stayed strongly associated with sudden death after adjustment for potential confounders.

In daily use they answer different questions. Resting heart rate is available every day. Heart rate recovery exists only on days you exercise, and it depends heavily on how the session ended. HRV is a third measure, the variation in time between beats. Its resting indices, such as RMSSD, reflect cardiac parasympathetic activity (Buchheit, 2014).

03What the mortality research found

The modern evidence starts at the Cleveland Clinic. Cole and colleagues (1999) followed 2,428 adults, mean age 57, who had a symptom-limited exercise test with thallium imaging. They defined heart rate recovery as the fall from peak exercise to one minute after stopping, and called 12 bpm or less abnormal. That applied to 26% of patients. Over six years, 213 died. An abnormal recovery carried a relative risk of death of 4.0. After adjustment for age, sex, medications, perfusion defects on imaging, standard cardiac risk factors, resting heart rate, the heart rate rise during exercise and workload achieved, the relative risk was 2.0.

Later studies tested whether the finding held in other groups and protocols.

All estimates except Jouven's are adjusted for other risk factors. Nishime's comes from the 8,549 patients not taking beta-blockers. Jouven's figure is before adjustment, and the association held after it. Shetler's abstract does not say whether its hazard ratio is adjusted. Shetler and colleagues also found that recovery had no diagnostic power for detecting coronary disease on angiography.

Qiu and colleagues (2017) pooled nine prospective cohorts covering 41,600 people from the general population, with 2,082 deaths. Attenuated recovery, compared with the fastest recovery group, carried a pooled hazard ratio of 1.68 for all-cause mortality and 1.69 for cardiovascular events. Each 10 bpm lower recovery carried a hazard ratio of 1.09 for death and 1.13 for cardiovascular events. The authors noted that only one included study clearly described an active cooldown, and none reported whether heart rate was counted lying down or standing.

Three limits apply before you hold your own number against this table. The risk estimates are associations in groups, mostly older patients referred for testing, and they do not tell you how long you personally will live. The tests ran in clinical exercise labs, most of them to maximal effort. And the Qiu meta-analysis found about a 9% higher hazard per 10 bpm, a modest effect next to the session-to-session noise in a watch reading, covered below.

04What counts as normal

There are no reliable age and sex percentile tables for heart rate recovery from wearables. What exists are clinical cutoffs for an abnormal result, and each is tied to its protocol.

For one-minute recovery after a maximal treadmill test, more than 12 bpm is the conventional normal line (Cole et al., 1999, Nishime et al., 2000). Between a fifth and a quarter of patients referred for testing fell at or below it. Watanabe and colleagues (2001) noted that those earlier studies measured recovery during a light cooldown. For patients who stopped and lay on their left side, they set the line at more than 18 bpm. For two-minute recovery the published lines range from 22 bpm in male veterans (Shetler et al., 2001) to 42 bpm after a submaximal test in healthy adults (Cole et al., 2000). The spread shows how much protocol shapes the number.

The practical reading for your own data follows from that spread. A one-minute value above 12 bpm after a hard finish sits outside the range that predicted risk in clinical cohorts. Beyond that line, the population figures say little about whether 25 or 35 bpm is better for you. Your own history, recorded after comparable sessions, is the better reference. The baseline entry explains how to build one.

05What changes recovery heart rate

Aerobic fitness. Well-trained endurance athletes recover faster. In Imai's study, cross-country skiers had a faster early decline than age-matched controls, and patients with chronic heart failure had a slower one. Daanen and colleagues (2012) reviewed 13 studies. Three of five cross-sectional studies found faster recovery in trained people, and most longitudinal studies found recovery improved along with power output. Buchheit (2014) cites work in which recovery improved within the first week of an endurance program, weeks before resting HRV changed.

How hard you were working when you stopped. Harder efforts stimulate the metaboreflex more, which slows recovery (Buchheit, 2014). Easier efforts leave less room to fall, since a walk that ends at 110 bpm cannot drop 40 bpm. A drop after an all-out interval and a drop after a steady jog are different tests.

Cooldown and posture. The cutoff for an abnormal result rose from 12 bpm with a light walking cooldown to 18 bpm when patients stopped and lay down (Watanabe et al., 2001). The higher line for passive recovery means that stopping and lying down produces a larger drop than walking on. Standing, sitting, walking and lying down each give different values, and mixing them in one trend adds noise.

Fatigue and overreaching. Fatigue can make heart rate recovery faster. Bellenger and colleagues (2016) pooled 24 studies of endurance athletes. Recovery got faster when training improved performance, with a moderate effect size of 0.63. It also got faster during overreaching, when performance fell, with an effect size of 0.46. In a follow-up with 12 cyclists, recovery after a 5-minute time trial increased after two weeks of heavy training and decreased again during the taper (Bellenger et al., 2017). A rising recovery number during a hard block may reflect fitness or fatigue, and only performance, sleep and how you feel will tell them apart. The overreaching entry covers the other signs.

Heat, age and session length. Daanen and colleagues (2012) list age, ambient temperature, and the intensity and duration of the preceding exercise as confounders that make studies hard to compare. They make your own sessions hard to compare too.

Measurement noise. In Buchheit's review, the typical error of one-minute recovery was about 25%, expressed as a coefficient of variation. For a 30 bpm recovery that is roughly 7 or 8 bpm of noise between repeated tests. A single low reading after one session means little.

06How Apple Watch measures it

Apple Watch records heart rate recovery as Cardio Recovery, stored in Apple Health as the HealthKit type heartRateRecoveryOneMinute. Apple's developer documentation defines it as "the reduction in heart rate from the peak exercise rate to the rate one minute after exercising ended." The value is in beats per minute and is always positive. The type became available to apps in iOS 16 and watchOS 9.

The watch needs heart rate after the workout ends. For several years Apple's support page stated that when you use the Workout app, "Apple Watch measures your heart rate continuously during the workout and for 3 minutes after the workout ends to determine a workout recovery rate." Apple's documentation does not say how the watch picks the peak.

The heart rate itself comes from photoplethysmography. Green LEDs flash hundreds of times per second and photodiodes measure how much light the blood in your wrist absorbs with each beat. Apple states a supported range of 30 to 210 bpm. Accuracy during exercise is good, with larger errors in recovery. Dooley and colleagues (2017) compared an early Apple Watch with a chest strap in 62 adults and found mean absolute percentage errors of 1.14% to 6.70% across stages. The watch read lower than the strap during seated recovery. Kim and colleagues (2023) tested an Apple Watch Series 7 in 44 patients with coronary artery disease during a cardiopulmonary exercise test. Error stayed under 2% in every phase, and the watch's error was larger in recovery than during exercise. Neither study tested the Cardio Recovery value itself.

Apple Watch Series 12 and Ultra 4 also show a Recovery HRV in the Heart Rate app. That is heart rate variability sampled while you are still, a separate measure from Cardio Recovery.

To get Cardio Recovery values you can compare, the watch needs a clean start and a clean minute.

1. End the workout right after a hard, steady effort.
2. Stop the workout on the watch as soon as the effort ends.
3. Do the same thing in the next minute every time, either standing still or walking slowly.
4. Wear the watch snug. Apple's heart rate guidance starts with a good fit.
5. Compare values from the same kind of workout, such as the last interval of the same weekly session.

07How Titan uses recovery heart rate

Titan reads Apple's one-minute Cardio Recovery samples from Apple Health. It uses them in two places.

Trends. Trends has a Heart-rate recovery chart. Each day's point is the average of that day's one-minute recovery samples, in bpm. A menu filters the chart by activity type, including running, cycling, walking, hiking, swimming and rowing. With a filter on, Titan keeps a day's value only if you logged a workout of that type that day. This filter is the easiest way to keep easy walks from diluting the values from your runs.

Titan Age. Titan Age counts two-minute heart rate recovery as one of its 11 weekly signals. It carries 10% of the model and sits in the Recovery group. Apple Watch records only the one-minute value, so Titan multiplies it by 1.65 to estimate a two-minute drop and keeps the result between 10 and 70 bpm. A 27 bpm one-minute recovery becomes a 45 bpm estimate. Titan counts only readings from days with a recorded workout and needs at least 2 such days in the week, as the Titan Age Recovery group explains. The neutral reference is 10 bpm, the floor of that range, and higher estimates move Titan Age younger. When you expand the row, Titan shows a rating for the estimate, from Poor below 12 bpm through Average, Good and Excellent to Superior from 40 bpm.

The 1.65 factor is Titan's own conversion, and Titan's documentation cites no study for it. The research above shows why a fixed factor is an approximation. The ratio of two-minute to one-minute recovery depends on how hard you stopped and what you did afterward. Titan's rating bands are also Titan's own scale and do not come from the clinical cutoffs in the tables above.

Titan's Recovery score does not use heart rate recovery. Recovery compares the HRV recorded during your sleep and your sleeping heart rate with your own baselines. Both are available every day, including rest days.

08Reading your own number

Treat heart rate recovery as a slow fitness marker. Compare values from matched sessions over weeks, and expect a single reading to move by several bpm for no physiological reason. A rising trend alongside better pace or power and normal sleep is a good sign. A rising trend alongside worse performance and heavy legs can be overreaching, per the Bellenger data.

A persistent fall in recovery at a fixed effort, especially with chest pain, unusual breathlessness, dizziness or palpitations during or after exercise, is a reason to see a doctor. The clinical cutoffs above came from supervised exercise tests with ECG monitoring. A wrist reading cannot diagnose anything. Even in the clinic, Shetler and colleagues found that recovery did not identify who had coronary disease, so a normal-looking number does not rule it out.

09References

• Apple Inc. (2022, archived). Monitor your heart rate with Apple Watch. Apple Support. https://web.archive.org/web/20220606175635/https://support.apple.com/en-us/HT204666
• Apple Inc. (2026). Monitor your heart rate with Apple Watch. Apple Support. https://support.apple.com/en-us/120277
• Apple Inc. (n.d.). heartRateRecoveryOneMinute. Apple Developer Documentation, HealthKit. https://developer.apple.com/documentation/healthkit/hkquantitytypeidentifier/heartraterecoveryoneminute
• Bellenger CR et al. (2016). Monitoring athletic training status through autonomic heart rate regulation, a systematic review and meta-analysis. Sports Medicine 46(10):1461-1486. https://doi.org/10.1007/s40279-016-0484-2
• Bellenger CR et al. (2017). The effect of functional overreaching on parameters of autonomic heart rate regulation. European Journal of Applied Physiology 117(3):541-550. https://doi.org/10.1007/s00421-017-3549-5
• Buchheit M (2014). Monitoring training status with HR measures, do all roads lead to Rome? Frontiers in Physiology 5:73. https://doi.org/10.3389/fphys.2014.00073
• Cole CR et al. (1999). Heart-rate recovery immediately after exercise as a predictor of mortality. New England Journal of Medicine 341(18):1351-1357. https://doi.org/10.1056/NEJM199910283411804
• Cole CR et al. (2000). Heart rate recovery after submaximal exercise testing as a predictor of mortality in a cardiovascularly healthy cohort. Annals of Internal Medicine 132(7):552-555. https://doi.org/10.7326/0003-4819-132-7-200004040-00007
• Daanen HA et al. (2012). A systematic review on heart-rate recovery to monitor changes in training status in athletes. International Journal of Sports Physiology and Performance 7(3):251-260. https://doi.org/10.1123/ijspp.7.3.251
• Dooley EE et al. (2017). Estimating accuracy at exercise intensities, a comparative study of self-monitoring heart rate and physical activity wearable devices. JMIR mHealth and uHealth 5(3):e34. https://doi.org/10.2196/mhealth.7043
• Imai K et al. (1994). Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure. Journal of the American College of Cardiology 24(6):1529-1535. https://doi.org/10.1016/0735-1097(94)90150-3
• Jouven X et al. (2005). Heart-rate profile during exercise as a predictor of sudden death. New England Journal of Medicine 352(19):1951-1958. https://doi.org/10.1056/NEJMoa043012
• Kim C et al. (2023). Validation of wearable digital devices for heart rate measurement during exercise test in patients with coronary artery disease. Annals of Rehabilitation Medicine 47(4):261-271. https://doi.org/10.5535/arm.23019
• Nishime EO et al. (2000). Heart rate recovery and treadmill exercise score as predictors of mortality in patients referred for exercise ECG. JAMA 284(11):1392-1398. https://doi.org/10.1001/jama.284.11.1392
• Qiu S et al. (2017). Heart rate recovery and risk of cardiovascular events and all-cause mortality, a meta-analysis of prospective cohort studies. Journal of the American Heart Association 6(5):e005505. https://doi.org/10.1161/JAHA.117.005505
• Shetler K et al. (2001). Heart rate recovery, validation and methodologic issues. Journal of the American College of Cardiology 38(7):1980-1987. https://doi.org/10.1016/S0735-1097(01)01652-7
• Watanabe J et al. (2001). Heart rate recovery immediately after treadmill exercise and left ventricular systolic dysfunction as predictors of mortality, the case of stress echocardiography. Circulation 104(16):1911-1916. https://doi.org/10.1161/circ.104.16.1911