Debug School

ashley
ashley

Posted on

Steps to Calories: MET Math, Stride Assumptions, and Error Budget

Turning a pedometer reading into calories is a classic "simple" calculation with a surprisingly large error budget. The MET-based method is the standard approach, but every term — step length, body mass, MET value, duration — carries assumptions worth making explicit in code.

MET formula

The ACSM equation for energy expenditure in kcal/min is:

kcal/min = MET * 3.5 * kg / 200
Enter fullscreen mode Exit fullscreen mode

where MET is the metabolic equivalent of the activity. Walking at 3 mph is about 3.0 METs; brisk walking at 4 mph is about 5.0 METs. The constant 3.5 is resting oxygen uptake in ml/kg/min (1 MET).

From steps to duration

Steps alone do not encode speed. A common estimate assumes a stride length of about 0.762 m (2.5 ft) for walking:

distance_m = steps * stride_length_m
time_min   = distance_m / speed_m_per_s / 60
Enter fullscreen mode Exit fullscreen mode

For 10,000 steps at stride 0.762 m and speed 1.34 m/s (3 mph):

distance = 7620 m = 7.62 km
time     = 7620 / 1.34 / 60 ~ 94.8 min
kcal/min = 3.0 * 3.5 * 70 / 200 = 3.675
calories = 94.8 * 3.675 ~ 348 kcal
Enter fullscreen mode Exit fullscreen mode

A 70 kg walker burns roughly 350 kcal over 10,000 steps at this pace.

Sensitivity: stride length and body mass

Stride length scales roughly with height — a common approximation is stride ≈ height * 0.414 for walking. A 1.80 m person has a stride near 0.745 m; a 1.55 m person near 0.642 m. At fixed step count, the shorter walker covers ~14% less distance, so a single global stride constant can bias the calorie estimate by that much.

Body mass enters linearly: the same walk burns 25% more calories for an 87.5 kg person than a 70 kg person.

Boundary conditions

  • Running changes both stride length and MET non-linearly; do not reuse a walking stride for a run.
  • MET tables are population averages. Age, fitness, terrain, and backpack load shift actual expenditure by 10–30%.
  • The ACSM formula overestimates for some populations; accelerometry-based equations (e.g. Freedson) are alternatives but require cadence data.
  • Round only the final kcal output. Rounding time or MET first compounds error.
  • Never present the result as exact; show it as an estimate with assumptions (stride, pace, mass) visible.

Verification and tool reference

For a step-to-calorie estimate with adjustable stride, pace, and mass, see the steps to calories calculator at Miryori.


Disclosure: The author leads content at Miryori, the free online calculator suite linked above. Educational arithmetic only, not medical advice.

Top comments (0)