Most heart rate sensors in smartwatches and fitness trackers work by shining light into your skin and measuring how much of that light bounces back. This technique, called photoplethysmography (PPG), detects tiny changes in blood volume with each heartbeat. Chest straps use a completely different approach, picking up the electrical signals your heart generates. Both methods track heart rate, but they measure fundamentally different things.
Optical Sensors: Light In, Pulse Out
Every time your heart beats, a small wave of blood pushes through your arteries and into the tiny blood vessels near your skin’s surface. That pulse of blood changes how much light the tissue absorbs. An optical heart rate sensor exploits this by pairing a small LED with a photodetector. The LED shines light into your wrist, and the photodetector measures how much light returns. When more blood is present (the peak of a heartbeat), more light gets absorbed and less reaches the detector. When blood volume drops (between beats), more light bounces back. The sensor reads these fluctuations as a waveform, and each dip in returning light corresponds to one heartbeat.
Hemoglobin, the oxygen-carrying molecule in red blood cells, is the key player here. It absorbs light at specific wavelengths, so the concentration and movement of hemoglobin during each cardiac cycle directly shapes the signal the sensor picks up.
Why Green Light Works Best on Your Wrist
If you’ve ever looked at the back of a smartwatch, you’ve probably noticed green LEDs. Green light (around 570 nm wavelength) penetrates just deep enough to reach the small arteries in the dermis layer of skin without going much further. That shallow penetration is actually an advantage: it picks up a cleaner pulsatile signal with less interference from deeper, non-pulsating tissue.
Infrared light, by contrast, passes through the full thickness of the skin and picks up information from a wider range of tissue. Traditional medical pulse oximeters use infrared paired with red light because those two wavelengths behave differently depending on whether hemoglobin is carrying oxygen or not, which is how they calculate blood oxygen levels. But for simply counting heartbeats on a wrist, green light produces a clearer signal. Research has also shown that green light performs reliably across various skin types at rest, and green or blue wavelengths hold up better during exercise when motion makes the signal noisier.
How Chest Straps Differ
Chest strap monitors don’t use light at all. They detect the electrical impulses that trigger each heartbeat, essentially performing a simplified version of a hospital electrocardiogram (ECG). Electrodes in the strap sit against your chest and pick up the voltage changes that spread across your body each time the heart’s electrical system fires. Because they capture the signal at its source rather than measuring a downstream effect, chest straps provide more precise timing between beats.
That precision matters for heart rate variability (HRV), which measures the subtle differences in timing from one beat to the next. Optical wrist sensors estimate those intervals indirectly by detecting when a pulse of blood arrives at your wrist. But the time it takes blood to travel from your heart to your wrist isn’t perfectly constant. It shifts with blood pressure, blood vessel stiffness, and other circulatory factors. This is why ECG-based chest straps remain the preferred tool for detailed HRV analysis, while wrist sensors work well for general heart rate tracking.
Filtering Out Motion Noise
The biggest engineering challenge for wrist-based sensors is movement. When you swing your arms while running, the sensor shifts against your skin, blood sloshes around in your wrist’s vessels, and the optical signal gets buried in noise. Wearable manufacturers solve this by pairing the optical sensor with an accelerometer and sometimes a gyroscope. These motion sensors track exactly how your wrist is moving, and algorithms use that data to identify which parts of the optical signal come from your heartbeat and which come from arm swing or vibration.
This motion artifact cancellation happens in real time and has improved dramatically in recent years. Multiple LEDs and photodetectors arranged in different positions also help, because the software can compare signals from several spots on your wrist and triangulate the true pulse. Still, during very intense or erratic movement, even the best algorithms occasionally lose the signal for a few seconds before locking back on.
Why Accuracy Varies by Skin Tone
Melanin, the pigment that determines skin color, absorbs light in the visible spectrum. In darker skin, more of the sensor’s green light gets absorbed by melanin before it ever reaches the blood vessels underneath. This reduces the strength of the returning signal and can affect accuracy. A study using the Fitzpatrick skin classification scale found a significant correlation between darker skin tones and heart rate discrepancies during high-intensity exercise, with PPG sensors on darker skin sometimes reporting readings that diverged from ECG reference measurements.
Manufacturers have worked to address this by increasing LED brightness, using multiple wavelengths, and improving their signal processing. Some newer devices combine green LEDs with infrared and red LEDs to get a more reliable reading across a wider range of skin tones. But the physics of melanin absorption means that optical sensors still tend to perform best on lighter skin, and this remains an active area of improvement across the industry.
Cold Weather and Blood Flow
Temperature also plays a significant role. When your hands get cold, blood vessels near the skin’s surface constrict to conserve heat. This pulls blood away from exactly the area where the sensor is trying to read it. Research on cold digits found that pulse signal amplitude dropped by roughly 46% to 60% compared to normal skin temperature, with the effect being statistically significant. In practical terms, this means your watch may struggle to get a reliable reading during a cold winter run or if your hands are chilled.
Warming your hands back up reverses the effect quickly. Signal amplitude actually improved by 59% to 70% above baseline when skin was warmed to about 33°C (91°F). If your sensor seems to be giving erratic readings in cold conditions, warming your wrist or wearing the watch slightly higher on your forearm where skin stays warmer can help.
From Raw Signal to Beats Per Minute
The raw data from an optical sensor is a continuous waveform, not a simple count. To turn it into a heart rate number, the device’s processor first filters the signal to remove noise outside the expected heart rate range (typically 0.5 to 15 Hz, which covers roughly 30 to 250 beats per minute). It then identifies the peaks in the waveform, measures the time between them, and calculates beats per minute.
For HRV measurements, the timing needs to be more precise. Consumer wrist sensors typically sample 25 times per second, but the software upsamples that data (interpolating between measurements to create a higher-resolution signal, sometimes up to 200 data points per second) to more accurately pinpoint each pulse peak. Medical-grade ECG monitors, by comparison, sample at 1,000 times per second. This gap in raw resolution is one reason clinical HRV assessments still rely on chest-based ECG rather than wrist optical sensors.
ECG Features in Smartwatches
Some smartwatches now include actual ECG capability alongside their optical sensors. The Apple Watch, for example, received regulatory clearance for an ECG app that records a single-lead electrocardiogram by using electrodes built into the watch’s back crystal and crown. When you place your finger on the crown, a small electrical circuit completes across your chest, and the watch captures a signal similar to one lead of a standard 12-lead hospital ECG.
This feature is designed specifically to detect atrial fibrillation (AFib), an irregular heart rhythm that increases stroke risk. The regulatory threshold for clearance required the algorithm to achieve at least 90% sensitivity (correctly identifying AFib when present) and 92% specificity (correctly identifying normal rhythm). These watches aren’t replacements for full cardiac monitoring, since they capture only one electrical perspective compared to the 12 views a hospital ECG provides, but they can flag irregularities worth investigating.
The optical PPG sensor and the ECG sensor serve different roles in the same device. The PPG runs continuously in the background, tracking your heart rate throughout the day. The ECG activates only when you initiate a reading, requires you to stay still for 30 seconds, and provides a snapshot that can reveal rhythm abnormalities the optical sensor alone would miss.