Trackers wildly off

A calorie number on your wrist can look precise the way a gas pump looks precise, with digits rolling to the decimal. The catch is that Stanford researchers found the calorie estimates on seven popular wrist trackers missed by 27% on the best device and 93% on the worst. (med.stanford.edu) The basic job of a fitness tracker is two different jobs hiding in one screen. Heart rate is a pulse signal your watch can see with light, but calories burned are an estimate built from formulas about your body, your movement, and your effort. (mdpi.com) Stanford’s team tested seven devices on 60 volunteers while the volunteers sat, walked, ran, and cycled. The researchers compared each watch against continuous heart monitoring and indirect calorimetry, which measures oxygen use and carbon dioxide output to estimate energy burn in a lab. (mdpi.com) The heart-rate side held up fairly well. Six of the seven devices stayed within 5% median error during cycling, which is close enough that Stanford said most of them measured heart rate adequately in lab-based activities. (med.stanford.edu) The calorie side did not. The paper says no device got energy-expenditure error below 20%, and device error was highest during walking, which is one of the most common things people buy trackers to measure. (mdpi.com) That gap exists because “calories burned” is not something a watch directly sees. The device watches clues like motion and pulse, then guesses the invisible part the way a bank guesses fraud from spending patterns instead of watching the purchase happen. (mdpi.com) The Stanford paper also found bigger errors for men, for people with higher body mass index, and for people with darker skin tone. That means the same workout can produce different quality estimates depending on who is wearing the device, not just which brand they bought. (mdpi.com) The practical problem shows up after the workout, not during it. If a watch says you burned 700 calories when the real number was far lower, it becomes easy to “eat back” exercise with a snack or meal that wipes out the deficit you thought you created. (med.stanford.edu) That risk can be worse for fitter people because efficient bodies often do the same work with less energy cost. A tracker that leans on generic formulas can turn that personal difference into a false permission slip to eat more than the workout actually covered, which is an inference from how these devices estimate energy expenditure rather than a direct finding of the Stanford trial. (mdpi.com) The useful part of a tracker is not the single calorie number at 6:14 p.m. It is the trend line over weeks: whether you are walking 8,000 steps instead of 3,000, whether your resting heart rate is falling, and whether you are training more consistently than last month. (med.stanford.edu) Stanford’s senior author, Euan Ashley, put it bluntly in 2017: people are making life decisions from these numbers. Nine years later, the viral post is old research getting fresh attention because the core warning still lands: use the watch for patterns, not for exact calorie math. (med.stanford.edu)