Your wearable’s best number is your own normal, not the average

Key takeaways

• In a study of 92,457 adults, average resting heart rates ranged from about 40 to 109 beats a minute, yet each person’s own rate usually moved only about 3 beats from day to day.
• In research studies, changes in heart rate, sleep and steps flagged many COVID-19 infections days before symptoms, but the methods couldn’t tell one virus from another.
• None of 14 “readiness” or “recovery” scores reviewed in 2025 disclosed its formula, and few had any independent testing.
• Travel, alcohol, short sleep and menstrual cycles can all move your numbers, so look for changes that last several days and show up in more than one measure.
• Chest pain, fainting or trouble breathing need care right away, whatever your device says.

In a study published in 2017, a Stanford team described one of its own members, a 58-year-old man who wore several health sensors for months. About 12 days after a trip to rural Massachusetts, his heart rate and skin temperature began running higher than usual. His blood oxygen readings dipped unusually low during and after a flight¹.

A fever followed, a doctor prescribed the antibiotic doxycycline, and a later blood test confirmed Lyme disease, an infection spread by ticks. The researchers wrote that the wearable signals had helped alert him that something was wrong¹.

The devices didn’t diagnose anything. What made the readings useful was that the team had months of his data, so they knew what normal looked like for him.

That idea, your own normal, is the key to reading a wearable. About one in five U.S. adults regularly wore a smartwatch or fitness tracker in 2019¹².

A population chart shows what’s typical for thousands of people. Your device can show what’s typical for you, and when that changes.

This guide explains how to find your baseline, which changes deserve attention, why “readiness” scores call for caution, and when a number plus a symptom means you should get care. It ends with a 10-minute review you can do every week.

A normal resting heart rate spans 70 beats, but yours spans far fewer

Resting heart rate is how many times your heart beats per minute while you’re still. For adults, a typical range is 60 to 100, according to Amy Sarma, a cardiologist at Mass General Brigham⁵. People who exercise a lot often run lower, and some medicines keep the rate below 60 on purpose⁵.

That range is broad, and a 2020 study shows just how broad. Giorgio Quer and colleagues at Scripps Research analyzed nearly 33 million daily readings from 92,457 Fitbit wearers in all 50 states, collected from 2016 to 2018⁴⁷.

The overall average was about 65 beats a minute. But individual averages ranged from about 40 to 109⁴. Age, sex, body mass index and sleep together explained no more than 10% of the differences between people⁴.

Within one person, though, the number held steady. The typical day-to-day wiggle was about 3 beats⁴. About 80% of people never saw their rate swing by 10 beats or more within a week, while the other 20% had at least one such week⁴.

There was even a small seasonal rhythm, with resting rates about 2 beats higher in early January than at the end of July⁴.

Here is why that matters: picture two people who each wake up with a resting rate of 72.

For one, it’s an ordinary morning. For the other, whose usual rate is 58, it’s a jump worth noticing. A population chart would call both of them normal.

In studies, wearables flagged illness days before symptoms

When COVID-19 arrived, research teams tested this baseline idea on a large scale.

At Stanford, Michael Snyder’s group recruited more than 5,000 adults, most of whom wore a smartwatch². Among 32 people with COVID-19 and usable data, 26 showed changes in heart rate, steps or sleep. When the signal came early, it typically showed up about four days before symptoms². Infected people also walked about 1,440 fewer steps a day and slept about 30 minutes longer⁶.

At Scripps Research, the DETECT study enrolled 30,529 people in the spring of 2020³⁷. For 333 people who had symptoms and got tested, the team compared each person’s resting heart rate, sleep and activity with their own baseline from the weeks before they got sick³.

The results held a surprise: resting heart rate alone did little better than a coin flip at sorting people with COVID-19 from people without it. Changes in sleep and activity did better, and sensor data plus symptoms did best³.

Researchers grade this with a score where 0.5 is a coin flip and 1.0 is perfect. Symptoms alone scored 0.71, and symptoms plus sensor data scored 0.80³. Resting heart rate alone scored 0.52³.

A third study used Oura rings, which measure skin temperature at the base of a finger. Of 50 people who reported COVID-19, 38, or 76%, showed fever-like temperature changes before they noticed symptoms⁸. The ring’s maker supplied the rings and helped pay for the research, and the infections were self-reported rather than confirmed by lab tests⁸.

Each method compared people with themselves. The ring researchers noted that one fixed fever cutoff wouldn’t work for everyone, because people differ so much⁸.

These were research tools, not home diagnoses. The Stanford method could not tell COVID-19 from other respiratory viruses, and the devices were not approved by the Food and Drug Administration (FDA) to detect illness². Detection also struggled when people lacked a stable stretch of normal data before getting sick⁹.

Most bumps in your data have an everyday cause

Even healthy people got alerts. In the Stanford study, people without infection set off about 0.66 alarms a month, or about two every three months, many during the holiday season².

The Stanford team named common triggers: long flights, high altitude, menstrual cycles, heavy drinking and some medicines⁶. Mass General Brigham adds caffeine and lack of sleep⁵. In the Fitbit study, women of childbearing age had the most day-to-day variation of any group⁴.

Blood oxygen readings have their own trap. In the 2017 Stanford study, oxygen levels fell during flights, and about 15% of readings at cruising altitude on one device were 90% or lower¹. The dips showed up across the passengers studied, so a low reading on a plane is not, by itself, a sign of illness¹.

This creates a real risk: when alerts fire often for ordinary reasons, it gets easy to tune them all out, including the one that matters.

Readiness scores are recipes with a secret ingredient list

Many devices now roll your data into one daily score, with names like readiness, recovery or strain. In 2025, Cailbhe Doherty and colleagues examined 14 of these scores from 10 companies, including Fitbit, Garmin, Oura, Polar, Samsung and WHOOP¹⁰.

Most scores drew on similar ingredients. Heart rate variability, the small changes in time between heartbeats, was used by 86% of them. Resting heart rate appeared in 79%, and sleep amount and activity in 71% each. Body temperature appeared in 29%¹⁰.

None of the companies disclosed its exact formula or how much each ingredient counted¹⁰. The authors found little to no peer-reviewed testing of the scores themselves. One study they cited found that a recovery score didn’t match how recovered a group of elite swimmers said they felt¹⁰.

The team spotted other problems. Sleep and heart rate variability are linked, so one bad night can be counted twice. “Readiness” also means different things from brand to brand, and some scores sit behind a paywall¹⁰.

The practical move is to read the ingredients. When your score drops, open the details and see which number moved. A low score driven by one short night calls for an early bedtime, not alarm.

Federal rules help explain why devices stay vague. Under the FDA’s General Wellness guidance, reissued in January 2026, a wellness wearable may show your values, trends and baselines and may suggest seeing a health professional. To stay in that category, though, the notice can’t name a disease or call a reading “abnormal, pathological, or diagnostic”¹¹.

The same guidance says being a wellness product does not mean the FDA has found it safe or effective¹¹. Interpretation is left to you and, when needed, your clinician.

A 10-minute weekly review

Daily scores pull you toward reacting, while a weekly look shows the pattern. Pick a regular time, such as Sunday morning, and use your app’s weekly or monthly view.

1. Write down three averages. Note resting heart rate, sleep time and daily activity. Add heart rate variability or skin temperature if your device tracks them.
2. Compare with your recent past. Look at the last three or four weeks, not at a chart of other people. The DETECT team used each person’s readings from three weeks to one week before illness as their baseline³.
3. Note the context. List travel, alcohol, illness, hard training, your cycle, new medicines and stress. Many of these move resting heart rate⁵⁶.
4. Look for runs, not blips. A one-day spike after a party is noise. A rise that lasts several days, especially in more than one measure, is worth watching³⁶.
5. Check in with your body. Ask whether you feel run-down, achy or short of breath. In DETECT, symptoms plus sensor data beat either one alone³.
6. Choose one small action. Try an earlier bedtime, an easy training day or fewer drinks. If any red flag below applies, call your clinician instead.
7. Keep a one-line log. Write the date, what changed, the likely cause and how you felt. Over months, this becomes a summary a clinician can use.

When a number plus a symptom means get care

Your body’s signals outrank your device’s. Mass General Brigham advises against overanalyzing tracker data. Sarma, though, says to tell your doctor about any sudden, dramatic change in heart rate⁵.

What you notice	What to do
Chest pain, or other signs you may be having a heart attack	Call 911, whatever your watch shows5
Fainting, nearly fainting, or sudden dizziness or lightheadedness	Get medical care right away5
A racing or pounding heartbeat that won’t stop	Get medical care right away5
Trouble breathing, with or without a low oxygen reading	Get medical care right away; the symptom matters more than the reading15
A low heart rate with lightheadedness, or a rate that won’t rise when you exercise	See your primary care provider5
A sudden, large change in resting heart rate with no clear cause, or feeling unwell with an unusual rate	Call your doctor5
New heart symptoms when you already have heart disease	Contact your cardiologist5
A higher resting rate for a day after drinking, travel or a hard workout, and you feel fine	Note it and recheck in a few days46

When you do talk with a clinician, bring a summary rather than a phone full of charts. For example: “My resting heart rate usually runs about 58. It has been near 70 for five days, and I’ve been short of breath on the stairs.” Add your device’s name and the dates.

You can also check a reading yourself. Mass General Brigham suggests pressing two fingers on the inside of your wrist, counting beats for 15 seconds, and multiplying by four⁵.

What the evidence supports

Claim	Evidence level	What the research shows
Each person has a fairly stable resting heart rate of their own	Strong evidence	In 92,457 adults, the typical daily wiggle was about 3 beats4
Comparing with your own baseline catches changes that population ranges miss	Promising but early	Research methods relied on personal baselines, but mostly in small studies348
Wearable changes can flag infections before symptoms	Promising but early	Small COVID-era studies; methods couldn’t name the virus238
Resting heart rate alone reliably signals infection	Evidence says no, or unsafe	In DETECT, it scored close to a coin flip3
Symptoms plus sensor data beat symptoms alone	Promising but early	The detection score rose from 0.71 to 0.803
Finger temperature rises can come before fever symptoms	Promising but early	Seen in 76% of a 50-person study supported by the ring’s maker, with self-reported infections8
Readiness and recovery scores reflect how recovered you are	Not enough evidence	14 scores, undisclosed formulas, little independent testing10
A low oxygen reading during a flight signals illness	Evidence says no, or unsafe	Dips were common at cruising altitude in healthy travelers1

Numbers can sharpen your body sense, not replace it

Natural health has long asked people to notice their own bodies: how they sleep, how they recover, when they’re coming down with something. A wearable can put numbers on that attention.

In the Lyme case and the COVID-19 studies, the value came from comparing a person with their own past¹². That is self-knowledge, the kind natural-health traditions prize, with a little help from a sensor.

But a score built on a formula no one outside the company can see should not outrank how you feel¹⁰. If your device says you’re “ready” and you feel awful, believe your body. If the watch is quiet and you have chest pain, call 911.

Questions readers ask

The bottom line

The most useful thing a wearable knows is your own normal. Population ranges are wide, but your resting heart rate, sleep and activity tend to stay in a narrow band. A change that lasts several days and shows up in more than one measure is worth your attention.

Treat readiness scores as rough summaries, read the numbers underneath, and review your data weekly instead of hourly. Above all, let symptoms lead. Chest pain, fainting or trouble breathing call for care, whatever the screen says.

Keep exploring: AI health information and health app privacy.

Related reading

• Fitness trackers add steps, but their calorie counts are rough guesses
• Why being out of range often means nothing at all
• The symptoms that mean call 911 now
• The health record nobody else is keeping for you
• “FDA registered” isn’t “FDA approved”: a buyer’s map to wellness devices

Sources

1. Li X, Dunn J, Salins D, et al. “Digital Health: Tracking Physiomes and Activity Using Wearable Biosensors Reveals Useful Health-Related Information.” PLOS Biology, 2017. journals.plos.org
2. Mishra T, Wang M, Metwally AA, et al. “Pre-symptomatic detection of COVID-19 from smartwatch data.” Nature Biomedical Engineering, 2020. nature.com
3. Quer G, Radin JM, Gadaleta M, et al. “Wearable sensor data and self-reported symptoms for COVID-19 detection.” Nature Medicine, 2021. nature.com
4. Quer G, Gouda P, Galarnyk M, Topol EJ, Steinhubl SR. “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, 2020. journals.plos.org
5. Mass General Brigham. “What Your Heart Rate Is Telling You.” Mass General Brigham, June 24, 2024. massgeneralbrigham.org
6. Stanford Medicine News Center. “Smartwatch can detect early signs of illness.” Stanford Medicine, December 4, 2020. med.stanford.edu
7. Scripps Research. “Early results from DETECT study suggest fitness trackers and smartwatches can predict COVID-19 infection.” Scripps Research, October 29, 2020. scripps.edu
8. Smarr BL, Aschbacher K, Fisher SM, et al. “Feasibility of continuous fever monitoring using wearable devices.” Scientific Reports, 2020. nature.com
9. Bahmani A. “Early detection of COVID-19 at scale using wearables” (Behind the Paper). Springer Nature Research Communities, December 16, 2020. communities.springernature.com
10. Doherty C, Baldwin M, Lambe R, Burke D, Altini M. “Readiness, recovery, and strain: an evaluation of composite health scores in consumer wearables.” Translational Exercise Biomedicine, 2025. doi.org
11. U.S. Food and Drug Administration. “General Wellness: Policy for Low Risk Devices.” Final guidance, January 2026. fda.gov
12. Pew Research Center. “About one-in-five Americans use a smart watch or fitness tracker.” Pew Research Center, January 9, 2020. pewresearch.org