Why Wearable Data Needs a Personal Baseline

A number from our wearable means nothing until something tells us how to read it. Compared to population averages, it says how we stack up against strangers. Compared to our own baseline, it tells us what’s actually changing in our body. The frame decides what the number means to us, and often, how we react to it.

HRV, 45. Resting heart rate, 58. Sleep score, 72. These are often the numbers we live by. They can feel like the information we need in order to function correctly; the stat that defines our day and determines our reactions. A low reading lands as bad news, while a high one lands as permission. How often do we ask the question that actually matters…72 compared to what?

That comparison is the whole game. A single measurement on its own says nothing. It becomes useful only when it’s held up against a reference, and the reference we choose decides what the number is allowed to tell us. Two people can wake up to an identical reading and walk away with opposite conclusions because they’re reading the same data through different models of themselves.

How Accurate is Wearable Data?

Accurate but rarely exact. Watches and rings usually read heart activity from light bounced off the skin using a technique called photoplethysmography, rather than from the electrical signal a clinical ECG measures directly. Plenty of studies have found these estimates to range in accuracy, and the brand of wearable we have largely determines how big the error is.

Accuracy, though, isn’t actually that important if we frame the data the right way. The raw number is the least interesting thing a wearable gives us. Even a perfectly accurate reading would sit there meaning nothing until someone or something told us how to read it. Many of us might be guilty of treating a soft estimate as a hard verdict, and then building our whole mood on top of it. That something is what we’ll call a frame, and many of us are running one we never chose.

Why Population Averages Describe a Person Who Doesn't Exist

A frame built on population averages can tell us how we stack up against a crowd of strangers. That’s almost never what we actually want to know, at least not if we want it to be actionable. It takes our reading, sets it beside everyone else's, sorted roughly by age and sex, and reports whether we came out high, low, or somewhere in the middle of the crowd. 

A key issue is how loosely any single person fits the crowd. A 2020 study that  followed nearly 92,000 people for two years found resting heart rates ranging from about 40 to 109 beats per minute across individuals, and that this enormous spread had little to do with age or sex. The average is real as a statistic but imaginary as an actual person. Nobody is actually walking around “being” the average, which is a strange thing to base our whole mood on.

Eric Topol, cardiologist and director of the Scripps Research Translational Institute, has spent years arguing that personal data matters precisely because it avoids this problem.

"You'll know much more about yourself, what's working, what's off track." — Eric Topol, cardiologist and director, Scripps Research Translational Institute

Interpreted through population norms, a naturally low resting heart rate can get flagged as a major issue when it’s actually normal for us, and a perfectly average-looking reading can hide a real problem because it still sits inside the band everyone else is in but is majorly out of range for us as an individual. We can end up reassured when we should be curious and rattled when we’re completely fine.

Why Our Own Baseline Gives Us More Value

A frame built on our own baseline measures change rather than rank. A baseline is just our own normal, calculated as the rolling average of our wearable data over the last “x” days. It’s built for us and no one else. Against that personal average, a single number stops being a grade and becomes a comparison to ourselves a week ago, providing much more usable data and avoiding the comparison trap we can fall into altogether.

Most wearables do this already, but it’s useful to understand why…and why the “compared to others my age” button on Garmin, Strava, and many other platforms can be a slippery slope.

Our body's stats drift constantly with sleep, stress, a coming illness, and even the time of year. A sudden jump well outside our personal range carries information the population band would never register because the population band is wide enough to swallow it and treat it as normal. An HRV of 45 reassures the person whose normal is 40 and worries the person whose normal is 60; only the personal frame can tell them apart. There are many confounding factors that affect our stats too, and we’re the only ones that truly know everything that’s going on in our lives.

When Population Norms and Personal Baseline Disagree

The gap between these two frames is most noticeable when they give opposite answers about the same morning. Someone recovering from a punishing week at work might see their HRV drop ten points below their own normal while still sitting inside the healthy population range. The population frame says we’re fine. The personal frame says our body is working harder than usual to hold things together.

With personalized baselines, the warning trigger can be different from one person to the next. The frame that knows our own history catches the shift where a comparison to population norms likely says everything’s fine. This is why most wearables already baseline our data against ourselves, but each one has it’s own approach and quirks. Some options offer population comparisons, which might help us orient ourselves for where we stand, but how useful is it really? If we’re trying to get healthier, fitter, or just understand our bodies more, the most important comparison is against ourselves. Are we improving vs. ourselves from two weeks ago? From a month ago? From a year ago? 

The instinct to chase a better device, a sharper sensor, or a higher-resolution number assumes the number itself is where the meaning lives. While accuracy is helpful, an inaccurate number that is consistently inaccurate still shows us where we’re trending. The number over time is where true value lives. A more accurate 45 is still just 45 until a model decides what “45” means for us specifically. Our wearable is much more useful the moment the question changes from whether the number is good to whether it’s good for us. Everything the device records passes through that choice of reference before it reaches us as a judgment about ourselves. 

References

1. Dial, M. B., Hollander, M. E., Vatne, E. A., Emerson, A. M., Edwards, N. A., & Hagen, J. A. (2025). Validation of nocturnal resting heart rate and heart rate variability in consumer wearables. Physiological Reports, 13(16), e70527. https://doi.org/10.14814/phy2.70527

2. Ginsburg, G. S., Picard, R. W., & Friend, S. H. (2024). Key issues as wearable digital health technologies enter clinical care. New England Journal of Medicine, 390(12), 1118–1127. https://doi.org/10.1056/NEJMra2307160

3. Nelson, B. W., Low, C. A., Jacobson, N., Areán, P., Torous, J., & Allen, N. B. (2020). Guidelines for wrist-worn consumer wearable assessment of heart rate in biobehavioral research. npj Digital Medicine, 3, 90. https://doi.org/10.1038/s41746-020-0297-4

4. Quer, G., Gouda, P., Galarnyk, M., Topol, E. J., & Steinhubl, S. R. (2020). Inter- and intraindividual variability in daily resting heart rate and its associations with age, sex, sleep, BMI, and time of year: Retrospective, longitudinal cohort study of 92,457 adults. PLOS ONE, 15(2), e0227709. https://doi.org/10.1371/journal.pone.0227709

5. Shaffer, F., & Ginsberg, J. P. (2017). An overview of heart rate variability metrics and norms. Frontiers in Public Health, 5, 258. https://doi.org/10.3389/fpubh.2017.00258