In partnership with
Most wrist wearables display pulse, steps, and energy expenditure beside one another, but they do not measure them in the same way. Optical sensors estimate pulse from changing light at the skin. Accelerometers record movement. Software then combines those signals, sometimes with other inputs, to calculate activity or calorie estimates. The display can make unlike measurements look equally direct.
The 10 Best Stocks for Fall 2026
MarketBeat analysts just updated their 10 Best Stocks to Own for Fall 2026 report. The list includes AI growth names, a high-yield dividend payer, a steady portfolio diversifier and companies positioned for continued growth into year-end. Get all 10 stocks and the research behind them for free now.
Photoplethysmography, usually shortened to PPG, shines light into tissue and detects changes in returned light as blood volume changes with each pulse. Software processes this waveform to estimate pulse rate. An electrocardiogram records electrical activity through electrodes. PPG instead observes an optical signal at the skin surface. The distinction matters because the sensors measure different physical phenomena.
Movement makes the optical signal harder to separate from noise. The device can shift against skin, the wrist can flex, and contact pressure can change during exercise. A loose strap may weaken or distort the reading. Algorithms filter movement artifacts, but each filter also decides which intervals to smooth, reject, or display. The result depends partly on those software choices.
A 2026 systematic review indexed in PubMed assessed 47 studies of wrist-worn trackers. The studies compared devices with different references, including ECG, indirect calorimetry, doubly labelled water, manual step counts, and research-grade accelerometers. The review reported better performance over time, while finding continuing variation across brands, activities, and populations. It also noted that study settings ranged from laboratories to simulated daily-life conditions.
Introducing Fair Game from Vanity Fair
Join Mark Guiducci for a series of in-depth and far-reaching conversations with the most interesting characters from the worlds of Hollywood, Politics, Style, Silicon Valley, and beyond. It's all Fair Game.
The review found that energy-expenditure estimates had improved least. Recent trackers still showed 10% to 40% error against indirect calorimetry in the studies it summarized. That reference measures oxygen use and carbon-dioxide production. A wrist device does not directly sample those gases. It estimates energy use from available signals and a model, so the output depends on assumptions.
Heart-rate results were stronger in some modern devices, but the review did not support a universal accuracy label. It reported typical exercise validity scores above 0.80 for some models, alongside variation by activity. A correlation or validity score can summarize how measurements move together. It does not by itself show how far a reading differs from a reference for one person at one moment.
A separate 2025 study tested 28 healthy adults using six monitors during rest, warm-up, burpees, and treadmill exercise. Agreement was high at rest, with concordance values above 0.99. During burpees, it fell below 0.50 across devices. Placement also mattered: upper-arm or forearm devices performed better than wrist placements in some study conditions. These findings describe a small sample and specific models, not every device.
Contact is another engineering variable. A 2026 scoping review of 21 wearable PPG studies found wide differences in how researchers measured or reported sensor pressure. The included samples ranged from single-participant tests to a dataset with 1,142 people. Its authors called for standardized contact reporting and real-world validation. Strap fit therefore affects the optical interface, not simply comfort.
Different metrics need different references. Step counts can be checked against manual counts or research accelerometers. Pulse readings can be compared with ECG or validated chest sensors. Energy estimates require a metabolic reference. Comparing each output with its own reference avoids treating all watch numbers as one accuracy problem. It also makes limitations visible across devices and studies.
For anyone evaluating a wearable, the useful question is what the sensor measures, what the software infers, and under which conditions it was tested. A daily check-in cannot settle every device question, but it can keep method and context beside the number.
Arnold Schwarzenegger. Jay Shetty. And You.
Codie Sanchez, Colin & Samir, Joanna Stern, and Oliver Darcy are here too. They all built newsletters (and podcasts and communities and websites) on beehiiv.
Some came with massive followings. Others built theirs along the way. That’s the thing about beehiiv: with so many ways to earn and grow, the platform is ready for whatever you’re building next.
