Recovery & sleep · 3 min read

The cold that showed up in your metrics before your throat

By the time your throat hurts, your immune system has been at work for days — and the work shows up in resting heart rate, HRV and overnight temperature. Here is what the early drift looks like, and what it cannot prove.

Your body knew on Tuesday

The sore throat arrives on Friday. You call that day one. Your body disagrees. By the time a virus produces symptoms, your immune system has been working for days, and the work is metabolically expensive. Core temperature edges up. The heart works harder at rest. Heart-rate variability drops. Nothing hurts yet. Everything is already measurable.

The wearable studies made this concrete. Li and colleagues (2017) at Stanford watched heart rate and skin temperature deviate from baseline before participants reported feeling ill. Mishra and colleagues (2020) found a heart-rate signal at or before symptom onset in 63% of the COVID-19 cases they tracked by smartwatch. The signal is real. It is just quiet, and it only makes sense against your own baseline.

What the drift looks like

The pattern is boring, which is why it works. Resting heart rate sits 3–5 bpm above your 30-day average for two or three mornings. HRV drops 10–20% in the same window. Overnight temperature runs a few tenths of a degree warm. None of these is dramatic alone. Moving together, in the same direction, they are a pattern.

The unit that matters is deviation, not the absolute number. A resting heart rate of 62 is unremarkable for one person and a fever-grade anomaly for someone who lives at 49. A population chart cannot see the difference. Your own rolling average can. That is the entire trick, and it needs nothing more exotic than a consistent morning measurement.

Morning resting HR (bpm) 7-day trend
Three weeks of morning resting heart rate. The drift starts on day 14. The sore throat arrives on day 18.
+3–5 bpmtypical resting-HR drift before symptoms
−10–20%common HRV drop, same window
63%COVID cases flagged at or before symptoms

What the numbers cannot tell you

A deviation cannot name its cause. A late night, three drinks, a hard interval session, jet lag, a hot bedroom and an incoming cold all print roughly the same signature: resting heart rate up, HRV down, temperature up. The data says your body is under load. It does not say from what.

So one odd morning is weather. Check yesterday before you check your throat — the training log and the wine usually explain it. Two or three consecutive mornings of the same drift, with no lifestyle explanation, is a different message. That is the window where an easy decision is still available.

Worth keeping: “The sore throat is the press release. Resting heart rate was the leak.”

The cheap bet

You do not need certainty to act, because the costs are asymmetric. Downgrade one session for a false alarm and you lose almost nothing. Train hard into a real infection and you can buy a deeper illness and a week off. An easy day is cheap insurance against an expensive week. One caveat: if a fever arrives or symptoms sit below the neck, rest outright — do not downgrade, skip.

Heala's morning recovery score is built on this asymmetry. It reads sleep, HRV, resting heart rate and strain — with or without a watch — and flags the drift for you, so the easy-day decision is waiting with your coffee. The app launches 1 October 2026, and the waitlist is open now.

When the drift shows up

1
Check the cluster

One signal alone is noise. Look for resting HR up, HRV down and temperature up together.

2
Compare to your baseline

Judge deviation from your own rolling average, not a population chart. Your normal is the reference.

3
Downgrade, don't cancel

Swap the hard session for an easy one. Keep the habit, drop the load.

4
Give it 48 hours

Two drifting mornings with no lifestyle explanation is a pattern. One is just a Tuesday.

Where these numbers come from
Li et al., PLOS Biology, 2017 — wearable sensors detect physiological deviations before self-reported illness (Stanford)
Mishra et al., Nature Biomedical Engineering, 2020 — pre-symptomatic detection of COVID-19 from smartwatch data (Stanford)

Quick answers

Can a watch actually tell me I'm getting sick?

It can flag that something is off — resting heart rate up, HRV down, temperature raised against your baseline. It cannot say why. The same signature follows alcohol, hard training and poor sleep. Treat it as a prompt to look closer, not a diagnosis.

How big does the change need to be?

Judge against your own baseline, not an absolute number. Roughly: resting heart rate 3–5 bpm above your 30-day average, HRV down 10–20%, overnight temperature a few tenths of a degree warm — for two or more consecutive mornings, across more than one signal.

Should I train if my metrics are down but I feel fine?

Downgrade rather than cancel. Keep the session, cut the intensity. If it was an incoming infection, you may have spared yourself a harder, longer illness. If it was just a rough night, one easy day costs you nothing you will notice in a month.

Heala opens 1 October 2026

Free forever — Pro is optional.

Photo food logging, adaptive training, a morning recovery score and a 3D body twin, in one app. Join the list and your spot — and your founding discount — are saved.

  • 15% off for life · first 1,000 members
  • Reserve your @username
  • One email at launch · no newsletter
Save my spot → Free · takes ten seconds
Keep reading