How the Stress Monitor Actually Works

Every wearable now sells a "stress" number, and most of them quietly imply something no sensor can deliver: that a watch knows how you feel. RepTrack's Stress monitor is built on a narrower, more honest claim, and this article walks through exactly what it measures, how the score and timeline are computed, and the rules that keep it from lying to you.

What the score actually claims (and what it never will)

Here is the entire claim, in one sentence: your heart is running above your own calm baseline, at a time when movement doesn't explain it.

That's it. Not "you are stressed." Not "anxiety detected." A heart-rate sensor measures physical activation, arousal, in the physiological sense, and arousal has many sources: work pressure, caffeine, heat, a fever coming on, excitement about good news. The number can't tell those apart, so RepTrack doesn't pretend it can. What it can do well is tell you when your body was activated and how far above your own normal it ran, which, watched over days, is genuinely useful.

This is a hard rule in the product, not a tone preference: no sentence on the Stress page diagnoses, grades, or names a mental state. And the whole feature is hideable in one tap: no confirmation dialog, no dark pattern, and the switch to bring it back is named right there.

The engine: your day in 5-minute bins

RepTrack slices the day into 288 five-minute bins and classifies each one before anything is scored:

• Still and awake: scoreable. This is the signal.
• Asleep: scoreable overnight (more below), shown under a moon-shaded wash.
• Working out, and the 60 minutes after a workout: excluded. Exercise elevates heart rate for reasons that aren't stress, and the recovery tail after a session would otherwise read as a panic attack.
• Moving (walking, stairs, detected from motion context and step cadence): excluded, for the same reason.
• No data: the watch wasn't on or didn't sample. Excluded, and honestly labeled.

Only bins that pass this filter get a score. Everything else renders as a gap or a shaded window on the timeline, never as a low reading: because "we couldn't measure you" and "you were calm" are entirely different facts, and drawing the first as the second is the classic wearable lie.

Scored against your calm, not a textbook's

Each scoreable bin's heart rate is compared against your personal calm baseline: the median of your sedentary heart rate over the last 14 days. The deviation maps to a 0–100 score:

• Calm (under 34): at or near your own settled normal.
• Moderate (34–66): running somewhat above it, common on busy days.
• Elevated (67+): well above your calm baseline for a sustained stretch.

Until you have 7 days of readings, the score is visibly labeled "calibrating" and leans on a population estimate, shown as what it is, never disguised as personal.

When your watch also records heart-rate variability, it corroborates the heart-rate story (suppressed HRV alongside elevated HR is a stronger signal than either alone). On days with no usable HRV, the score still works. It just carries a visibly lower confidence, labeled "heart rate only today."

Overnight: the line keeps going while you sleep

Sleeping bins with usable heart rate are scored too, so the timeline runs continuously through the night under its moon wash, the same behavior you'd see in Bevel or WHOOP. Two details keep this honest:

• Sleep readings never enter your calm baseline. Your sleeping heart rate is naturally lower than your waking calm; letting it drag the baseline down would make ordinary afternoons read as "elevated."
• Sleep never counts toward daily coverage. The daily average is a claim about your waking day, and sleeping bins can't pad it.

The timeline: 24 hours, zoomable, scrubbable

The Stress chart is a rolling 24-hour window ending at now: it slides with the clock and never resets to an empty pane at midnight. A magnifier button zooms into the trailing 12 hours. Touch and drag anywhere on the line and the gauge above follows your finger: the score at that instant, its band, and the clock time, snapped to actual measured readings only, never interpolated into a gap.

Below it, the Total day card compares today's time in each band against your typical day: preferring your history for the same weekday (your typical Saturday is not your typical Tuesday), built entirely from days that were actually measured. Days with no data are excluded from "typical" rather than averaged in as fake calm.

The daily average, and when there isn't one

Your daily stress average only exists when at least 30% of your waking hours were actually measured. Below that floor, RepTrack says "not enough data today", because an average of three bins isn't your day, it's three bins wearing your day's name.

What it needs, and what it refuses to fake

Stress reads all-day heart rate, which in practice means an Apple Watch worn during the day. Third-party wearables work exactly to the extent they sync heart rate into Apple Health. With no wearable at all, the page says "Requires Apple Watch" and explains why. There is no honest watchless arousal model, so RepTrack doesn't ship a dishonest one.

One medical caveat worth knowing: medications that suppress heart rate, beta blockers especially, remove the very signal this feature reads. If that's you, the score may not be valid, and the page's fine print says so explicitly. None of this is a medical assessment, and it detects no condition.

How to actually use it

• Look for patterns, not points. One elevated hour is a meeting. The same elevated block every weekday at 3 PM is information.
• Cross-reference the causes it can't see. Elevated all afternoon after a double espresso is caffeine doing exactly what caffeine does.
• Watch the overnight line. A night that starts elevated and settles late often explains a rough morning recovery score.
• Use the typical-day card for honesty about "busy." Feeling frazzled but tracking below your typical Saturday is a useful correction in both directions.
• An elevated stretch is a prompt, not a verdict. If you can, take the short break. If a reading ever worries you medically, that's a clinician conversation, not an app feature.

The bottom line

RepTrack's Stress monitor makes one narrow, defensible claim: how far your heart rate is running above your own calm baseline when movement doesn't explain it, in 5-minute bins, corroborated by HRV when available, continuing through sleep, and displayed on a timeline where a gap is always a gap and never fake calm. It won't read your mind, it won't score a day it couldn't see, and it's one tap to hide. That narrowness is the point: a small claim kept honestly beats a grand one that was never true.

Sources

• Kim HG, et al. "Stress and Heart Rate Variability: A Meta-Analysis and Review of the Literature." Psychiatry Investigation, 2018, HRV suppression under psychological stress.
• Taelman J, et al. "Influence of mental stress on heart rate and heart rate variability." IFMBE Proceedings, 2009, heart-rate elevation under cognitive load.
• Shaffer F, Ginsberg JP. "An Overview of Heart Rate Variability Metrics and Norms." Frontiers in Public Health, 2017.
• Firstbeat Technologies white papers, heart-rate and HRV-based stress detection as used by consumer wearables (Garmin).
• WHOOP and Bevel product documentation, continuous stress timelines with sleep-period tracking and personal baselines.
• Plews DJ, et al. "Training adaptation and heart rate variability in elite endurance athletes." Sports Medicine, 2013, why baselines and trends beat spot readings.