Why Smartwatch Heart Rate Readings Struggle on Dark Skin and Tattoos
Your watch says 72 beats per minute on the couch. Halfway through a hard run it says 118 — while a chest strap says 152. Or it shows nothing at all over the tattoo on your wrist.
The short answer: at rest, your skin tone barely affects the reading. During hard exercise, the error on darker skin can grow to around 12 beats per minute — and tattoos can block the sensor's light altogether.
The cause is physics, not a software bug. And the fix depends on which problem you actually have: a wrong number, a missing reading, or a blocked sensor. We covered the general accuracy problems — motion, fit, and cold hands — in our guide to how accurate your smartwatch really is. This one is about the two factors most guides gloss over: skin tone and tattoos.
How the Sensor Works — and Why Green Light Is the Weak Point
Every wrist sensor uses the same trick: shine light into your skin and watch how much comes back.
The technique is called photoplethysmography, or PPG. A green LED flashes light into your wrist hundreds of times per second, and a photodiode measures the reflection. Blood absorbs green light, so each heartbeat — a surge of blood through the capillaries — shows up as a dip in the reflected light. Count the dips, and you have a heart rate.
Green is the industry's default for two reasons: blood absorbs it strongly, giving a clear pulse signal, and it resists motion noise better than other colors. But green has a weakness that matters here:
- Melanin absorbs green light too. The more melanin in the skin, the less light reaches the blood and returns to the sensor — so the pulse signal gets weaker before the watch ever sees it.
- Red and infrared light go deeper. They are absorbed less by melanin, which is why newer watches add red and infrared LEDs alongside green. The improvement is real, but it does not remove the problem entirely.
- Tattoo ink is a different obstacle. Ink sits in the skin and absorbs or scatters the green light before it can reach blood at all — and unlike melanin, its effect depends on color, density, and pattern.
One measurement note: most studies sort participants on the six-point Fitzpatrick scale, which was designed to classify sun sensitivity, not skin color. It is an imperfect tool, and the newest studies have started measuring melanin directly instead.
What the Studies Actually Found: Fine at Rest, Shaky at Full Effort
The honest summary is a split verdict: at rest, skin tone barely matters; during hard exercise, the gap opens up.
A 2023 systematic review in the Journal of Racial and Ethnic Health Disparities pooled 10 studies, 469 participants, and 26 devices — and could not reach a clean verdict. Four studies found no skin-tone effect, four found lower accuracy on darker skin, and two were mixed. The authors called the evidence inconclusive.
That split is the story. The effect depends on the device and on how hard you are working:
The clearest numbers come from a 2025 study in PLOS ONE. Researchers compared a Fitbit Charge 5 against a Polar H10 chest strap in 25 people cycling on a recumbent bike. At rest, there was no difference in error across skin-tone groups. But as exercise intensity rose, the gap widened: at 40–60% of heart rate reserve, the average error for darker skin ran 7.6 beats per minute higher than the lightest-skin, low-intensity baseline; above 60%, it ran 11.7 beats per minute higher. The watch also tended to underestimate heart rate rather than overestimate it.
Other studies point the same way. A 2021 study found some WearOS watches underestimated heart rate by 10–15 beats per minute during moderate to vigorous exercise in darker-skinned users — and by more than 20% at the highest intensities. By contrast, Apple Watch models in the same review stayed within 5 beats per minute of variation across skin tones.
The practical read: the watch is not "wrong for dark skin." It is usually right at rest and increasingly unreliable as effort rises — and some watches handle it far better than others.
The Newest Finding: Readings Go Missing, Not Just Wrong
A February 2026 study found a different failure mode — and it is the one most guides miss.
It was the first study to measure epidermal melanin directly, using a colorimeter and the Individual Typology Angle (ITA°), instead of sorting people by eye. Twenty-eight healthy adults wore an Apple Watch Series 8, a Garmin vivosmart 5, and a SlateSafety BAND V2 while a Polar H10 chest strap recorded the true heart rate during rest, cycling, and recovery.
The headline result was not about wrong numbers. It was about missing ones:
- Half of the Apple Watch's missing data came from participants with the darkest skin (ITA° below 10°), and 85% of the SlateSafety band's missing data did too.
- Outliers — readings far from the true heart rate — were also concentrated in darker skin, accounting for 56–62% of them across all three devices.
- But the error itself barely moved. Skin pigmentation increased measurement error by only about 1 beat per minute, and only on one of the three devices.
That distinction matters. A wrong number is visible — you can spot a heart rate that does not match how you feel. A missing reading is silent: the watch simply has no data for part of your workout, and the training load, calorie estimate, and recovery score built on that workout quietly lose a chunk of their input. If your watch shows gaps in its heart-rate trace during hard sessions, that is the failure to look for.
Tattoos Block the Light in a Different Way
With tattoos, the problem is not a weaker signal. It is often no signal at all.
Apple says so directly in its support documentation: "the ink, pattern, and saturation of some tattoos can block light from the sensor." Dark, solid ink causes the most trouble — black and red in particular — while lighter colors and sparse patterns interfere less. The effect is specific to the patch of skin under the sensor: the same watch can read fine on one part of your wrist and fail a centimeter away.
What does not exist yet is a number for how wrong tattooed readings get. No manufacturer has published controlled accuracy data for tattooed skin, and independent validation studies generally exclude tattooed participants. Apple's own 2026 accuracy study for the Apple Watch Series 12 excluded people with wrist tattoos, according to an independent analysis of the white paper — so even the newest manufacturer claims say nothing about tattooed wrists.
Tattoos can also trip the watch's wrist detection, which is why some users find themselves repeatedly asked for a passcode. Turning wrist detection off stops the prompts, but it also stops background heart-rate readings — a trade that makes the accuracy problem worse, not better.
What Actually Works
None of this means giving up the watch. It means matching the tool to the job.
1. Tighten the fit and move the watch up
Wear the watch snug — about one finger above the wrist bone — and slide it higher up the forearm during workouts, where the skin is flatter and the sensor sits more steadily. A loose band lets light leak in and the sensor bounce, which is the same failure mode as a weak signal.
2. Try the other wrist — or the un-inked patch
If a tattoo sits under the sensor, move the watch to the other wrist or to a spot the ink does not cover. This is the fastest test of whether the tattoo is the cause: if readings improve immediately, it was.
3. For training that matters, use a chest strap
A chest strap measures the heart's electrical signal, not reflected light, so melanin and ink cannot touch it. The Polar H10 is the reference standard researchers use to judge every wrist device in the studies above. Pair it with the watch over Bluetooth and the watch records the strap's numbers instead of its own.
4. Or try an upper-arm optical sensor
If a strap feels uncomfortable, an upper-arm band is the middle ground. In a 2025 validation study in JMIR Cardio, the Polar Verity Sense upper-arm sensor averaged 1.43 beats per minute of error across nine activities, against 6.41 for a wrist watch tested the same way — both measured against a chest strap.
5. When buying, ask what the maker actually tested
Look for multi-wavelength sensors — green plus red and infrared — and check whether the manufacturer publishes accuracy results broken down by skin tone. Most do not. That silence is itself useful information: a watch whose accuracy was never tested across skin tones is a watch whose accuracy on your skin is unknown.
One last boundary: a smartwatch is a wellness device, not a medical one. If a reading worries you — a resting heart rate that has jumped, an irregular-rhythm alert, a number that does not match how you feel — check it another way and talk to a clinician rather than trusting or dismissing the watch on its own.
Why Regulators Are Stepping In — but Not for Your Watch
The same physics affects blood-oxygen readings, and there the consequences forced regulators to act.
A 2020 study in the New England Journal of Medicine found that pulse oximeters missed low blood oxygen about three times more often in Black patients than in White patients. In January 2025, the FDA issued draft guidance requiring pulse oximeter makers to test on 150 or more participants, with at least 30% of the data from each skin-tone group, and to meet strict accuracy limits across those groups.
But that guidance covers medical devices. Consumer wearables sold as wellness products are not reviewed by the FDA before sale, and no equivalent rule requires a smartwatch maker to prove its heart-rate sensor works across skin tones. Until that changes, the burden falls on the buyer: the studies above are, for now, the only public check.
Key Takeaway
Resting readings are trustworthy on any skin. The gap opens when you work hard, tattoos can shut the sensor down entirely, and the newest research says the most common failure on darker skin is a missing reading, not a wrong one. Check your heart-rate trace for gaps, not just for numbers that look off.
If the number matters — a training zone, a health question — measure it with a chest strap, and treat the wrist reading as an estimate.
Sources
- Hung, S.H. et al. — "Validity of heart rate measurements in wrist-based monitors across skin tones during exercise," PLOS ONE 20(2): e0318724 (2025) — doi.org/10.1371/journal.pone.0318724
- Mulholland, A.M. et al. — "Influence of skin pigmentation on the accuracy and data quality of photoplethysmographic heart rate measurement during exercise," European Journal of Applied Physiology (2026) — pubmed.ncbi.nlm.nih.gov/40968160
- Koerber, D. et al. — Systematic review of wearable heart-rate accuracy by skin tone, Journal of Racial and Ethnic Health Disparities (2023)
- Bent, B. et al. — Wearable optical heart-rate sensor accuracy across Fitzpatrick skin types, npj Digital Medicine (2020)
- Asif, S. et al. — "Photoplethysmography in Diverse Skin Tones: Evaluating Bias in Smartwatch Health Monitoring," Cureus 17(10): e94074 (2025) — pmc.ncbi.nlm.nih.gov/articles/PMC12592569
- Shcherbina, A. et al. — Stanford validation study of wrist-worn heart-rate and energy-expenditure accuracy (2017)
- Apple Support — "Your heart rate. What it means, and where on Apple Watch you'll find it"
- U.S. Food and Drug Administration — Draft guidance, "Pulse Oximeters for Medical Purposes" (January 7, 2025)
Comments
Post a Comment