Fit, Placement and Skin Characteristics in Wearable Accuracy

Why the skin matters

Most wrist and armband heart-rate sensors use photoplethysmography (PPG): a light shines into the skin and the sensor reads how the reflected light changes with blood volume. [1] Anything that changes the light's path, such as movement, a loose fit, pigment or ink, can add error. The researchers behind one study list three main sources of PPG inaccuracy: diverse skin types, motion artifacts (movement noise) and signal crossover. [1]

Evidence labels: mixed (studies disagree or differ by device), insufficient evidence, established.

Skin tone: what studies found

• A 2020 study of consumer and research-grade wearables found no statistically significant difference in heart-rate accuracy across skin tones, but did find significant differences between devices and between activity types; absolute error during activity was on average 30% higher than at rest. [1]
• A 2025 comparison of a Garmin Forerunner 45 against a Polar H10 chest strap in adults of varied Fitzpatrick skin types found no significant effect of skin tone on heart rate, with differences appearing when activity intensity changed. [2] The authors stress replicating this regularly because devices change quickly. [2]
• Another study of an armband monitor found skin tone was the only characteristic associated with accuracy during rest and walking, in a sample of 25. [3]
• Label: mixed. No study here covers every device or every skin tone, and results for one model should not be applied to another.

A related but separate issue is medical pulse oximeters (blood-oxygen readers). FDA says current evidence suggests some accuracy differences and lists use on patients with darker skin pigmentation as a circumstance with risk of inaccuracy. [4] It also says pulse oximeters sold as general wellness or sporting products are not reviewed by the agency before sale. [4] That statement is about oximeters, not about heart-rate watches.

Tattoos

In 25 people wearing a Polar Verity Sense armband, readings over tattooed skin were not accurate at rest (mean absolute percent error 22.9%), walking (7.5%) or running (5.1%), while non-tattooed skin stayed within acceptable accuracy; effects were largest at rest. [3] Tattoo age and intensity did not statistically explain the error. [3] One device, one small study: label insufficient evidence beyond that.

Motion and activity

Error rises during exercise, and devices differ in how quickly they respond to changes in effort. [1][2] Reviews find heart-rate wearables are more variable than step counters, and accuracy depends on brand and model. [5]

Fit, placement and body factors

Guidelines for researchers recommend recording factors that may influence wrist heart-rate measurement, including body mass index, wrist dominance and wrist circumference. [6] An analysis of many validation studies found larger error in studies with overweight samples and in studies with more women than men for heart rate and energy expenditure. [7] These are patterns across studies, not predictions for you. Proper evidence not found in the sources read for an effect of strap tightness on accuracy, so no fit instructions are given here.

What this means for you

• If a reading looks odd, compare with another method, such as feeling your pulse, before concluding anything.
• Do not use a wearable reading to rule out a problem. New palpitations, fainting or chest pain need medical care (911 in the US).
• See Chest Strap vs Wrist Heart-Rate Tracking and Optical Heart-Rate Sensing (PPG) Explained.
• For population differences see Wearable Validation in Different Populations.