Are wrist heart rate monitors accurate?

For most people at low-to-moderate intensity, yes. In a study of 199 people, wrist optical sensors had a mean absolute error no larger than about 3 beats per minute and a concordance of 0.98, which the researchers called clinically acceptable for a number of applications. Accuracy falls, however, as exercise intensity and arm movement rise, so a wrist reading is most trustworthy during steady effort and least trustworthy during fast, high-intensity work.

Is a chest strap more accurate than a watch?

Yes, against an ECG. A chest strap measures the heart's electrical activity directly, the same signal an electrocardiogram records, and validation studies put its agreement with the ECG at 0.98 in an athlete study and 0.996 in an American College of Cardiology study, where 1 is perfect agreement. Wrist optical sensors were lower in both, from about 0.89 to 0.96 in the athlete study and from 0.67 to 0.92 in the cardiology study, depending on the device and the activity. The chest-strap advantage is largest during intense or arm-heavy exercise.

Why is my watch heart rate wrong during exercise?

A wrist watch reads heart rate optically, by shining light through the skin, so anything that disturbs that signal can throw it off. Studies found that wrist accuracy decreases as intensity increases, with the worst readings during fast arm movement such as an elliptical with hand levers, and that absolute error is on average about 30 percent higher during activity than at rest. A loose fit, sweat, cold hands, or a watch that slides on the wrist can all degrade the reading. A chest strap, which measures the electrical signal instead, avoids the optical problem.

Do I need a chest strap, or is a watch enough?

It depends on the goal. For general fitness, daily activity, and steady cardio, a wrist sensor is generally accurate enough. A chest strap is the better choice for interval and sprint training, where heart rate changes fast, and for holding a precise heart-rate zone, where the target depends on a trustworthy number. Neither tool is a medical device, so the choice is about training precision, not diagnosis.

Can a smartwatch detect heart problems?

A consumer heart-rate reading is not a diagnosis. The American College of Cardiology notes that these wrist-worn devices are not intended to be medical devices, and a heart-rate number cannot confirm or rule out an irregular rhythm or any heart condition. Symptoms such as palpitations, fainting, chest pain, shortness of breath, or an unusually fast or slow rate at rest warrant assessment by a clinician rather than interpretation from a watch.

Does skin tone affect heart rate monitor accuracy?

The evidence is mixed, so the honest answer is that it might. Skin type is one of the documented factors in optical sensing. One systematic study that tested wrist optical accuracy across the full range of skin tones found no statistically significant difference across skin tones (Bent 2020), but that null has been critiqued as underpowered for the darkest skin tones (Colvonen 2021), and a 2025 PLoS One study found that wrist-sensor error grew with exercise intensity for darker skin tones, reaching a mean of 16.5 beats per minute above 60 percent of heart-rate reserve, more than four times the 3.5 beats per minute in the light skin-tone group (Hung et al. 2025). Motion and fit clearly affect any wrist reading too. Anyone who wants to remove the optical question entirely, for any reason, can use a chest strap, which does not rely on a light signal through the skin.

Gear

Heart-Rate Monitors: How Chest Straps and Wrist Sensors Compare

A calm, sourced comparison of chest-strap and wrist heart-rate monitors: chest straps measure the heart's electrical signal and are the most accurate against an ECG, while wrist optical sensors are good enough for most people at low-to-moderate effort but lose accuracy as intensity and arm movement rise. None of these consumer devices is a diagnostic medical instrument.

Written by Michael Harley, Independent Health & Nutrition ResearcherLast reviewed: Jun 8, 2026

Heart-rate monitors come in two common forms, and they read the heart in fundamentally different ways. A chest strap sits against the skin around the torso and measures the heart's electrical activity, the same signal an electrocardiogram (ECG) records. A wrist-based watch or band instead uses optical sensing, known as photoplethysmography or PPG: it shines light into the skin and reads the small changes in blood volume that happen with each beat. Both can report a heart rate in beats per minute, but the underlying methods are not equally accurate in every situation.

This guide is general and educational, and it is brand-neutral: it explains what the category of equipment does and does not do, not which product to buy. It covers how the two technologies work, what the accuracy studies actually found, where each tool is the better choice, and an important boundary that applies to every consumer device here, namely that none of them is a diagnostic medical instrument. The accuracy figures below come from published validation studies and are reported as data, not as endorsements of any specific device.

The essentials at a glance

Chest straps are the most accurate against an ECG: a study of athletes found the Polar H7 chest strap had the greatest agreement with the ECG (concordance 0.98), and an American College of Cardiology study put chest-strap agreement with an EKG at 0.996 (Pasadyn 2019; ACC/Gillinov 2017).
Wrist optical sensors are good enough for most people at low-to-moderate intensity: in a mixed sample of 199 people, wrist PPG had a mean absolute error no larger than about 3 beats per minute and a concordance of 0.98, which the researchers called clinically acceptable for a number of applications (Hettiarachchi and Stahl 2018).
Accuracy of wrist devices falls as exercise intensity and arm movement rise: in the athlete study no wrist watch reached acceptable accuracy at the two highest treadmill speeds, and the worst readings come during fast arm movement such as an elliptical with hand levers (Pasadyn 2019; ACC/Gillinov 2017).
Motion and fit are documented factors in optical accuracy: a systematic study found absolute error about 30 percent higher during activity than at rest, and that same study found no statistically significant difference in accuracy across skin tones (Bent 2020). The evidence on skin tone is not settled, however: some validation work, especially at higher exercise intensity, has found reduced wrist-sensor accuracy for darker skin tones (Hung et al. 2025).
The chest-strap advantage matters most for interval and sprint training and for staying inside a heart-rate zone; for general fitness and steady cardio a wrist sensor is generally good enough.
None of these consumer monitors is a diagnostic medical device: the American College of Cardiology notes they are not intended to be medical devices, so symptoms such as palpitations, fainting, chest pain, or a suspected irregular rhythm warrant a clinician, not a watch (ACC/Gillinov 2017).

How the two technologies work

A chest strap is an electrical sensor. Worn against the skin around the chest, it picks up the small voltage changes the heart produces with each beat, which is the same kind of signal an electrocardiogram measures. Because it reads the electrical event directly, a chest strap tends to track even rapid changes in heart rate closely. The American College of Cardiology describes the chest strap as working like the EKG, which measures the electrical activity of the heart.

A wrist monitor works by light. The technique is called photoplethysmography, or PPG: green or infrared light shines into the skin from the back of the watch, and a sensor reads how much light is reflected as blood pulses through the vessels under the wrist with each heartbeat. This is an indirect measurement, and that is where its weaknesses come from. Anything that disturbs the optical signal can reduce accuracy: movement that shifts the watch against the skin, a loose fit that lets light leak in, sweat, cold or reduced blood flow at the wrist, and the general challenge of reading a faint pulse signal through tissue. The trade-off is convenience. A wrist sensor is always on the body, comfortable, and needs no extra strap, which is exactly why it has become the everyday default even though a chest strap reads the heart more directly.

What the accuracy studies show

The comparison is consistent across studies: chest straps are the more accurate reference, and wrist optical is close enough at lower effort but slips at higher effort. A 2019 study in Cardiovascular Diagnosis and Therapy tested 50 healthy, athletic adults wearing a three-lead ECG, a Polar H7 chest strap, and assorted wrist monitors while running at treadmill speeds from 4 to 9 miles per hour. Agreement with the ECG was summarized with a concordance correlation coefficient, where 1 is perfect agreement. The Polar H7 chest strap had the greatest agreement with the ECG at 0.98, followed by one wrist watch at 0.96, with three others at 0.89. The key pattern came on the treadmill: the researchers reported that the accuracy of the wrist-worn devices decreased as intensity increased, and at 8 and 9 miles per hour none of the wrist-worn devices reached an acceptable level of agreement.

An American College of Cardiology study reached the same conclusion in different exercise modes. It found the chest strap closely matched the EKG, with agreement of 0.996, while the wrist-worn devices were less accurate on average, ranging from 0.67 to 0.92. The wrist monitors became less accurate the more intense the activity, and the lead author noted they were most accurate on the treadmill at low intensity and worst on the elliptical at high intensity. The single worst case was fast arm movement: when participants used the elliptical with hand levers, the wrist and forearm monitors generally failed to give correct readings.

The other side of the balance is just as important. A 2018 study in BMC Sports Science, Medicine and Rehabilitation tested wrist optical against a chest strap in a heterogeneous sample of 199 people, including pregnant women and patients with coronary artery disease, over 371 hours of everyday and exercise activity. There the wrist sensor performed well: a concordance of 0.98, a mean absolute error no larger than about 3 beats per minute, and limits of agreement between roughly minus 12 and plus 13 beats per minute. The authors concluded that this level of error can be considered clinically acceptable for a number of applications. Read together, the studies say wrist optical is a capable everyday tool whose weak spot is high-intensity, high-movement exercise, exactly where a chest strap holds its accuracy.

What the evidence does not support, and which tool fits which goal

The evidence does not support treating a wrist monitor as useless. In the largest mixed-population comparison, wrist optical was clinically acceptable for many applications, and for general fitness, daily activity tracking, and steady cardio it is generally good enough. What the evidence does support is matching the tool to the goal. When the heart rate is changing fast or the arms are moving hard, a chest strap is the more reliable choice. That makes the chest strap most valuable for interval and sprint training, where a few seconds of laggy or wrong readings can mislead the effort, and for anyone trying to hold a specific heart-rate zone, where the target depends on a trustworthy number.

Heart-rate zones are simply intensity bands, often defined as percentages of a maximum heart rate, used to keep an easy session easy and a hard session hard. The companion guide on why cardiorespiratory fitness may be one of the strongest predictors of a long life explains why training across those intensities matters, and the heart rate that calorie and energy estimates lean on is part of the broader picture covered by the site's TDEE calculator. The practical takeaway is unglamorous: for most people most of the time, a wrist sensor is convenient and accurate enough, while a chest strap earns its place during fast, high-intensity, arm-heavy work and for precise heart-rate-zone training. Neither reading should be confused with a clinical measurement, a point the next section makes explicit.

About skin tone, fit, and motion

Because wrist monitors read a light signal through the skin, several physical factors can influence the quality of that signal. A systematic study in npj Digital Medicine identified three documented sources of optical inaccuracy: diverse skin types, motion artifacts, and signal crossover. That same study set out specifically to test accuracy across the full range of skin tones, and in its sample it found no statistically significant difference in accuracy across skin tones; the differences it did find were between devices and between activity types, with absolute error during activity on average about 30 percent higher than during rest.

That single result, however, is not the settled answer, and the evidence on skin tone is genuinely mixed. The npj Digital Medicine null has been critiqued in the same journal as underpowered for the darkest skin tones, with too few participants in the highest Fitzpatrick category to rule out an effect (Colvonen 2021). More importantly, a 2025 PLoS One study that compared a wrist sensor against a chest strap across light, medium, and darker skin-tone groups during graded exercise found a statistically significant interaction between skin tone and intensity: heart-rate measurement error was greater with increasing exercise intensity for people with darker skin tones. At the highest intensity tested, above 60 percent of heart-rate reserve, the mean error in the darker skin-tone group was 16.5 beats per minute, more than four times the 3.5 beats per minute seen in the light skin-tone group at the same intensity (Hung et al. 2025). The fair reading is that motion and fit clearly degrade any wrist reading, and that skin tone may also reduce wrist-sensor accuracy, particularly during hard exercise, even though not every study has found this.

The practical steps are the same regardless. A snug band that does not slide, worn a little above the wrist bone, and a sensor kept clean and reasonably dry give the optical method the best chance of a clean signal; cold hands and reduced blood flow at the wrist can also weaken the reading. For anyone concerned about accuracy during hard effort, for any reason, a chest strap sidesteps the optical signal entirely and stays steady when a wrist reading does not.

Not a diagnostic device. Consumer heart-rate monitors, whether chest strap or wrist watch, are fitness and training tools, not diagnostic medical instruments. The American College of Cardiology states plainly that these wrist-worn devices are not intended to be medical devices. A heart-rate number on a watch cannot diagnose an irregular rhythm, and it should not be relied on to rule a heart problem in or out. Symptoms such as palpitations, an unusually fast or slow rate at rest, fainting or near-fainting, shortness of breath, or chest pain warrant assessment by a clinician rather than interpretation from a device, and chest pain, severe breathlessness, or fainting are reasons to seek urgent care. This is general guidance on when a reading is worth raising, not a diagnosis. For what a resting pulse can and cannot indicate, the companion guide on resting heart rate covers the topic in more depth; anyone uncertain about their own readings should ask a healthcare professional who can interpret a number in the full context of their health.

Frequently asked questions

Are wrist heart rate monitors accurate?: For most people at low-to-moderate intensity, yes. In a study of 199 people, wrist optical sensors had a mean absolute error no larger than about 3 beats per minute and a concordance of 0.98, which the researchers called clinically acceptable for a number of applications. Accuracy falls, however, as exercise intensity and arm movement rise, so a wrist reading is most trustworthy during steady effort and least trustworthy during fast, high-intensity work.
Is a chest strap more accurate than a watch?: Yes, against an ECG. A chest strap measures the heart's electrical activity directly, the same signal an electrocardiogram records, and validation studies put its agreement with the ECG at 0.98 in an athlete study and 0.996 in an American College of Cardiology study, where 1 is perfect agreement. Wrist optical sensors were lower in both, from about 0.89 to 0.96 in the athlete study and from 0.67 to 0.92 in the cardiology study, depending on the device and the activity. The chest-strap advantage is largest during intense or arm-heavy exercise.
Why is my watch heart rate wrong during exercise?: A wrist watch reads heart rate optically, by shining light through the skin, so anything that disturbs that signal can throw it off. Studies found that wrist accuracy decreases as intensity increases, with the worst readings during fast arm movement such as an elliptical with hand levers, and that absolute error is on average about 30 percent higher during activity than at rest. A loose fit, sweat, cold hands, or a watch that slides on the wrist can all degrade the reading. A chest strap, which measures the electrical signal instead, avoids the optical problem.
Do I need a chest strap, or is a watch enough?: It depends on the goal. For general fitness, daily activity, and steady cardio, a wrist sensor is generally accurate enough. A chest strap is the better choice for interval and sprint training, where heart rate changes fast, and for holding a precise heart-rate zone, where the target depends on a trustworthy number. Neither tool is a medical device, so the choice is about training precision, not diagnosis.
Can a smartwatch detect heart problems?: A consumer heart-rate reading is not a diagnosis. The American College of Cardiology notes that these wrist-worn devices are not intended to be medical devices, and a heart-rate number cannot confirm or rule out an irregular rhythm or any heart condition. Symptoms such as palpitations, fainting, chest pain, shortness of breath, or an unusually fast or slow rate at rest warrant assessment by a clinician rather than interpretation from a watch.
Does skin tone affect heart rate monitor accuracy?: The evidence is mixed, so the honest answer is that it might. Skin type is one of the documented factors in optical sensing. One systematic study that tested wrist optical accuracy across the full range of skin tones found no statistically significant difference across skin tones (Bent 2020), but that null has been critiqued as underpowered for the darkest skin tones (Colvonen 2021), and a 2025 PLoS One study found that wrist-sensor error grew with exercise intensity for darker skin tones, reaching a mean of 16.5 beats per minute above 60 percent of heart-rate reserve, more than four times the 3.5 beats per minute in the light skin-tone group (Hung et al. 2025). Motion and fit clearly affect any wrist reading too. Anyone who wants to remove the optical question entirely, for any reason, can use a chest strap, which does not rely on a light signal through the skin.

References

Accuracy of commercially available heart rate monitors in athletes: a prospective study (Pasadyn SR et al.; Polar H7 chest strap rc=0.98; wrist watches rc=0.89-0.96; accuracy of wrist-worn devices decreased as intensity increased) · Cardiovascular Diagnosis and Therapy, 2019; 9(4):379-385. Accessed 2026-06-08.
Wrist-worn optical and chest strap heart rate comparison in a heterogeneous sample of healthy individuals and in coronary artery disease patients (Hettiarachchi IT, Stahl et al.; 199 participants, 371 h; concordance 0.98, mean absolute error not larger than 3 bpm, clinically acceptable; LoA -12.3 to 13.3 bpm) · BMC Sports Science, Medicine and Rehabilitation, 2018; 10:10. Accessed 2026-06-08.
Wrist-worn Heart Rate Monitors Less Accurate Than Standard Chest Strap (Gillinov M et al.; chest strap vs EKG rc=0.996; wrist devices rc=0.67-0.92; not intended to be medical devices) · American College of Cardiology (press release). Accessed 2026-06-08.
Investigating sources of inaccuracy in wearable optical heart rate sensors (Bent B et al.; PPG inaccuracy from skin types, motion artifacts, signal crossover; no statistically significant difference across skin tones; error ~30% higher during activity) · npj Digital Medicine, 2020; 3:18. Accessed 2026-06-08.
Validity of heart rate measurements in wrist-based monitors across skin tones during exercise (Hung SH, Serwa K, Rosenthal G, Eng JJ; significant skin tone x intensity interaction; mean error 16.5 bpm in darker skin tones vs 3.5 bpm in light skin tones above 60% heart-rate reserve, more than four times higher; error greater with increasing intensity for darker skin tones) · PLOS ONE, 2025; 20(2):e0318724; Public Library of Science. Accessed 2026-06-08.