Why Your Apple Watch Can't See Your Heart During a Deadlift

The set ends. You rack the bar, catch your breath, and glance at your Apple Watch. It says 132 bpm. You felt like you were at 155, maybe higher. You're not imagining it.

The Apple Watch is the most popular fitness wearable among lifters by a wide margin. It's also systematically wrong about your heart rate during the exact exercises you care about most: deadlifts, pull-ups, farmer's carries, heavy rows. Two peer-reviewed validation studies put numbers on the gap, and the gap isn't small.

This isn't a software bug Apple will patch in watchOS 11. It's physics. The optical heart rate sensor—photoplethysmography, or PPG—uses green light to measure blood volume changes under the skin. During a heavy compound lift, two things corrupt that signal: your wrist moves sharply and unpredictably, and your grip compresses the forearm muscles enough to reduce blood flow at the sensor site. The result is a lag of 5 to 15 seconds on the way up and another 5 to 15 on the way down. By the time the Watch shows you hit 140, you've already been at 150 for ten seconds and are now resting.

If you're using intra-workout heart rate to gauge effort, you're feeding yourself delayed, dampened data. The fix isn't a different watch band. It's knowing which signals from the Watch to trust and which to ignore.

The Two Studies That Proved the Gap

Bunn et al. published the first head-to-head validation of the Apple Watch Series 3 against a clinical ECG chest strap in 2018. Across multiple exercise modalities, the Watch's mean absolute error for resistance exercise was 7 to 10 beats per minute. That's the average miss—some reps were closer, some were worse. The exercises that produced the largest errors were deadlifts, pull-ups, and farmer's carries. Every exercise that required sustained tight grip and wrist involvement.

Falter et al. replicated the finding in 2022 with the Apple Watch Series 6, Polar Vantage V, and Fitbit Sense. The Series 6 had improved overall—coefficient of variation under 5% across five activity types—but still lagged the chest strap during sharp heart rate transitions. The kind of transitions that define every working set of a heavy compound lift.

The common thread: wrist-PPG works well for steady-state movement where the wrist moves rhythmically and grip pressure is low. Running, cycling, walking. It works poorly for the short, sharp, high-force movements that make up strength training.

What Actually Goes Wrong in the Wrist

The PPG sensor shines green light through the skin and measures the reflected variation as blood pulses through the microvasculature. Three factors corrupt the signal during lifting.

Motion artifact is the first. The accelerometer-based motion compensation that Apple uses works well for steady-state movement—it can predict the wrist's position during a run stride. It cannot predict the violent, non-repetitive motion of a deadlift lockout or a pull-up pull-through. The sensor loses lock on the pulse wave and interpolates until it reacquires.

Grip-induced occlusion is the second, and it's specific to lifting. When you grip a barbell or pull-up bar hard, the forearm muscles contract and compress the superficial blood vessels near the wrist. Less blood reaches the skin surface under the sensor. The PPG signal weakens. The Watch has to work harder to extract a pulse, and the result is noisier.

Skin tone and tattoo absorption are the third factor, though less common. The green light wavelength is absorbed by melanin and blocked entirely by tattoo ink. If the sensor sits over a dark tattoo, the Watch may show no reading at all.

None of these are fixable with a firmware update. They're inherent to the physics of optical heart rate sensing at the wrist.

The Signal Delay in Real Training

Imagine you're doing a set of five deadlifts at 85% of your 1RM. Your heart rate climbs quickly as you approach the bar, spikes during the set, and begins dropping the moment you set the bar down. The chest strap reads the rise in under two seconds and the drop almost as fast. The Apple Watch takes 5 to 15 seconds to register the peak and another 5 to 15 to reflect the recovery.

In practice, this means: by the time the Watch shows you've reached zone 5, you've already been there for the entire set. You start your rest period, and the Watch continues climbing for another ten seconds before it turns around. Your logged recovery heart rate appears higher and slower than it actually is.

This matters if you're using heart rate to decide when to start your next set. The typical prescription is to rest until heart rate drops to a certain threshold—say, 120 bpm. If the Watch is showing you 120 when your actual heart rate is already 110, you're cutting your rest short. Over a session, that compounds into higher fatigue and lower quality on later sets.

What the Apple Watch IS Good At

The Watch does three things well that matter more than intra-workout heart rate: resting heart rate, heart rate variability (HRV), and sleep tracking. All three are measured at rest, with no motion artifact and no grip occlusion. The PPG sensor works cleanly when you're still.

Resting heart rate trends downward over weeks of consistent training and upward during illness or accumulated fatigue. HRV, measured during a morning orthostatic test or overnight, is a reliable marker of recovery status. Sleep tracking gives you duration and consistency data that directly inform training readiness.

These are the signals worth building a training decision framework around. They're less exciting than the live heart rate display during a set, but they're more actionable. We covered this in detail in Apple Watch numbers that change training. The short version: use the Watch for what it's good at, ignore it for what it's not.

What to Use Instead for Intra-Workout Heart Rate

If you need accurate heart rate during heavy compound lifts, the answer is a chest strap. Polar H10, Garmin HRM-Pro, Wahoo TICKR—any of them will give you ECG-accurate readings with sub-second lag. The chest strap measures electrical activity directly from the heart, not optical blood flow at the wrist. Motion and grip don't affect it.

The trade-off is convenience. A chest strap is one more thing to put on, one more thing to wash, one more battery to charge. For most lifters, the convenience of the Watch outweighs the accuracy loss for daily training. The key is knowing when the loss matters.

It matters during interval-style training where rest periods are short and precise. It matters during RPE-based programming where you're using heart rate as a cross-check on perceived effort. It matters when you're trying to track recovery between sets of heavy compounds.

For steady-state cardio, general daily activity, and resting metrics, the Watch is fine. For the deadlift, pull-up, and farmer's carry, it's not.

A Better Framework: Use HRV, Not Intra-Workout HR

If you're not willing to wear a chest strap, the smarter move is to stop relying on intra-workout heart rate entirely and shift your attention to the resting metrics the Watch handles well.

Morning HRV, measured consistently, gives you a recovery score that correlates with your body's readiness to train. Low HRV suggests accumulated fatigue, poor sleep, or impending illness. High HRV relative to your baseline suggests you're recovered and ready for a hard session. We wrote a full low HRV training guide that walks through how to adjust your session based on that single number.

The advantage of HRV over intra-workout heart rate is that it's measured under controlled conditions—same time, same position, no motion. The PPG sensor works as designed. The data is clean and actionable.

Resting heart rate is the second signal. A rising trend over several weeks is a red flag that warrants a deload or a check-in on sleep and stress. A sudden spike is often the first sign of illness. Both are more useful for long-term training decisions than the noisy, delayed heart rate reading from your last set of deadlifts.

The Recovery Score Problem

Many lifters use Apple Watch's native Training Load or a third-party app that weights heart rate during training into a recovery score. If that heart rate data is systematically wrong for your main lifts, the recovery score is built on a foundation of noise.

Dorsi's approach avoids this problem by not relying on intra-workout heart rate for adaptive decisions. The adaptation engine uses your logged weights, reps, RPE, and resting recovery metrics—not the wrist-PPG heart rate from your working sets. The result is a training plan that adjusts to what you actually did and how you actually recovered, without the artifact of a sensor that can't see your pulse during a deadlift.

The Short Version

Your Apple Watch is not bad. Wrist-PPG can't do this one thing. The deadlift, pull-up, and farmer's carry produce motion artifact and grip occlusion that corrupt the optical heart rate signal. Two validation studies confirm the error: 7 to 10 bpm average miss, 5 to 15 seconds of lag.

Use the Watch for resting heart rate, HRV, and sleep. Use a chest strap if you need accurate intra-workout heart rate. Or stop chasing intra-workout heart rate entirely and let your recovery metrics guide your training.

The number on your wrist during a set is the least reliable number your Watch gives you. The numbers it records when you're still are the ones that actually change training.

Sources

Bunn et al. published the first head-to-head Apple Watch validation in the Journal of Sports Medicine and Physical Fitness in 2018, showing that resistance exercise—especially deadlifts and pull-ups—produced the largest heart rate errors compared to a clinical ECG chest strap. Falter et al. replicated and extended the finding in the European Journal of Sport Science in 2022, demonstrating that even the Series 6, with improved optical sensors, still lagged behind chest-strap readings during sharp heart rate transitions. The broader PPG motion-artifact literature, summarized in reviews by the American College of Sports Medicine, confirms that wrist-based optical sensors are inherently limited by motion and tissue compression—physics, not software.