Apple Watch vs a chest strap: when heart rate accuracy matters

How accurate is Apple Watch heart rate for running?

Good by the standard of wrist sensors, and best on steady effort. In lab studies against an ECG, Apple Watch has repeatedly been the most accurate wrist device tested. Its error grows with intensity and with fast changes in effort, which is exactly when interval runners stare hardest at the number.

Gillinov and colleagues tested five optical wrist and forearm devices and a chest strap against an ECG in 50 adults on a treadmill, bike and elliptical. Agreement is a concordance coefficient; 1.0 is perfect.

Activity mattered as much as device. On the treadmill, all but one optical device agreed with the ECG at 0.88 to 0.93; on the elliptical, with its moving arm levers, none was accurate. The authors’ advice: when accurate heart rate is imperative, use a chest monitor with electrodes. Other studies found the same order. Wang and colleagues measured the strap at 0.99 and Apple Watch at 0.91 on a treadmill, wrist accuracy best at rest and worse with exercise. Shcherbina and colleagues put Apple Watch’s median heart-rate error at 2.0 percent, the lowest of seven wrist devices, though no device estimated energy expenditure to within 20 percent.

Why is a chest strap more accurate than a watch?

Because it reads a cleaner signal. A strap’s electrodes pick up the electrical spike of each heartbeat, which is hard to mistake. A wrist sensor shines light into the skin and reads the small pulsing change in blood volume, a softer signal that motion, cold and a loose fit can all blur.

The optical method, photoplethysmography, works well when the pulse is strong and the arm swings steadily. It struggles when skin blood flow drops, the band shifts, or your arm swing sets up a rhythm the sensor mistakes for your pulse. Montalvo and colleagues concluded that a chest strap could stand in for a clinical ECG in practical use, while Apple Watch Series 6 and 7 lost some accuracy as running speed rose.

When is Apple Watch heart rate inaccurate?

Mostly at transitions, at high intensity and in poor conditions. Steady running on a snug watch is the easy case; the first minute of an interval, cold hands, a loose band, some tattoos and unrhythmic movement are the hard ones.

• Interval starts and finishes. Horton and colleagues found a wrist optical sensor lagged the ECG whenever intensity changed, until a steady state returned, tending to read low as effort rose and high as it fell. Støve and colleagues found that delay explained 15 percent of one wrist device’s error and 2 percent of Apple Watch’s.
• Higher intensity. Thomson and colleagues found Apple Watch error of 2.4 to 5.1 percent across intensities, with agreement falling as effort rose.
• Fit. Apple support says to wear the watch snugly on top of the wrist and to tighten the band for workouts.
• Cold and tattoos. By Apple’s own account, wrist blood flow can fall too low for a reading in the cold, and some tattoos can affect the sensor.
• Irregular motion. Running’s rhythm reads better than irregular movement; gripping or swinging arms can confuse any wrist sensor.

For runners, a meta-analysis of 44 studies is reassuring: on treadmill walking and running, wrist sensors were off by only about half a beat per minute on average. The larger errors were in cycling and resistance training.

Does skin tone affect wrist heart rate accuracy?

The evidence is mixed. Light-based sensors could in principle be affected by skin pigment, and one 2017 study found higher error with darker skin. A larger 2020 study designed to test exactly this found no statistically significant difference across skin tones.

The 2017 study was Shcherbina and colleagues. Bent and colleagues tested 53 people across the full Fitzpatrick scale and saw no significant difference by skin tone, but error during activity ran about 30 percent higher than at rest. Apple Watch 4 had the lowest average error of the consumer devices: 4.4 beats per minute at rest, 4.6 during activity. Their results fit the idea that motion can pull a wrist sensor toward the arm swing’s rhythm, though they did not test that directly.

Do you need a chest strap for running?

Only when the exact number at a given second changes what you do. For an easy run, a long run or a steady tempo on a well-fitted watch, the wrist is usually close enough. For short hard intervals, cold days, heart-rate-led treadmill sessions and beat-to-beat analysis, a strap is the better tool.

• Short, hard intervals. Optical lag can make the first part of a 1 to 4 minute rep read lower than it is.
• Running by heart rate alone. When heart rate is all that steers you, as on a treadmill, its accuracy matters more.
• Heart rate variability and DFA a1. Methods that read the gap between beats need clean beat-to-beat data. DFA a1, a way of estimating the first threshold, was developed from lab ECG recordings, and Rogers and colleagues warn its accuracy may weaken as artifacts rise.
• No watch at all. Running with your phone only, a strap is your only source of heart rate.

How to test your own watch against a strap

Studies average many wrists; yours may do better or worse. Wear a strap, recording to a separate app or device, and the watch together for two runs: one steady easy run and one session of short hard reps. Then lay the traces over each other.

• Lines together on the easy run: the watch is fine for your easy and long running.
• Watch trails by a few beats at each rep’s start: that is lag. Judge the first part of a rep by effort or pace.
• Watch jumps to near your step rate and stays: it has likely locked onto your arm swing. Tighten the band, or wear the strap.
• Drops out or wanders only on cold mornings: low skin blood flow. A strap avoids it.

Even measured perfectly, heart rate lags at the start of a rep and creeps up over a long, hot run. That is physiology, not a sensor fault: steer by pace or effort and use heart rate as a check. More in Easy, Steady, Hard: RPE for runners and zone 2 running. This is training information, not medical advice: see our health disclaimer.

Can you connect a chest strap to Apple Watch or iPhone?

Yes, if it speaks Bluetooth. Apple Watch can connect to external Bluetooth heart-rate monitors such as chest straps, and iPhone can read any strap on the standard Bluetooth heart-rate profile. Straps that only broadcast ANT+ cannot talk to an iPhone or an Apple Watch.

Check the box for Bluetooth, not only ANT+. Wet the electrodes, wear the strap snug, and let one app at a time connect. If the first minutes read oddly, suspect dry pads.

How does Gears use heart rate?

In Gears, pace leads and heart rate guards. Each gear is a pace range you steer by, and heart rate can only make a gear easier, never harder: when your heart says the pace is too much on an Easy or Steady stretch, the coach eases the pace. Indoors, heart rate leads.

On iPhone alone, Gears reads any Bluetooth chest strap on the standard heart-rate profile and joins the nearest one, not another runner’s; a Garmin with Broadcast Heart Rate turned on works too. With Apple Watch, pair the strap in the watch’s Bluetooth settings and the watch uses it. With no heart-rate source, the iPhone coaches by pace and never pretends to know your heart rate. See Gears as a chest strap running app and as an Apple Watch running app.

What it doesn’t do: Gears doesn’t make an optical sensor more accurate, doesn’t use ANT+, and doesn’t turn a short run of heart-rate data into a verdict on your health.