Heart rate is a weak calorie signal at easy effort

When you are barely above resting — an easy walk — heart rate says almost nothing about how many calories you are burning, so it tends to under- or over-shoot. The mechanical cost of moving at a known pace and hill grade is steadier and closer to the truth in that range.

Why it works

Near rest the heart rate hardly changes with effort, and non-metabolic influences (posture, arousal, heat, hydration, cardiovascular drift) make up a large fraction of that small change — so converting heart rate to energy is imprecise. The energy cost of walking or running at a measured pace and grade is well characterized and far more stable at low intensity.

What it means in practice

In the workout calorie engine, when a distance workout's heart rate shows easy (Zone 1 / near-resting) intensity, use the pace-and-grade estimate for those minutes instead of the heart-rate estimate; keep heart rate for moderate-and-up and for non-distance workouts.

The evidence

• Leonard, W.R. (2003). Measuring human energy expenditure: What have we learned from the flex-heart rate method?. American Journal of Human Biology.

  The heart rate–energy relationship is weak and non-linear at low intensities, where heart rate barely tracks metabolic rate. The Flex-HR method handles this by substituting resting metabolic rate whenever heart rate falls below an individually-calibrated 'flex' point, applying the linear heart-rate regression only above it. Group-mean estimates are accurate, but individual error is larger and the flex point must be calibrated per person.

• Johansson, H.P., Rossander-Hulthén, L., Slinde, F. et al. (2006). Accelerometry combined with heart rate telemetry in the assessment of total energy expenditure. Br J Nutr.

  Using movement at low-to-medium intensity and heart rate only at higher intensity improved estimation over heart-rate-alone approaches and agreed well with doubly labeled water. The explicit crossover — mechanical signal when easy, heart rate when hard — is the core of the validated combined method.

• Achten, J., Jeukendrup, A.E. (2003). Heart rate monitoring: applications and limitations. Sports Medicine.

  Heart-rate-derived VO2/energy estimates can deviate up to ~20% from measured values. The heart-rate–VO2 relationship is distorted by cardiovascular drift, dehydration, and ambient temperature — non-metabolic influences that make up a larger share of the small heart-rate elevation at low intensity, where the relationship is least reliable.

Why we call confidence high

Two independent methods converge: the Flex-HR method (Leonard 2003) drops heart rate below an individually-set low-intensity flex point, and validated combined accelerometry+heart-rate models (Johansson 2006) use movement at low/medium intensity and heart rate only when hard. Achten & Jeukendrup 2003 explains why — non-metabolic factors dominate the small heart-rate rise near rest, giving up to ~20% error.

Where it applies

Healthy adults during low-intensity locomotion (walking, easy running/cycling with a pace signal)

Does not apply to: non-locomotor exercise with no pace/distance signal (strength training, yoga, HIIT) where heart rate stays the best available intensity signal; moderate-to-high intensity where the heart-rate–energy relationship is reliable and captures effort that pace misses.