⚕️All tools are for informational and educational purposes only — not medical advice.Full disclaimer

Open analysis · Dataset 001

How far apart are common calorie formulas?

Across 72 predefined adult profiles, the three equations differed by an average of 81.6 kcal/day. In 20 profiles, the gap was at least 100 kcal/day.

Published 31 July 2026Reproducible computational analysisCC BY 4.0 data

72

profiles compared

81.6 kcal

average formula spread

258.3 kcal

largest formula spread

4.7%

average spread vs Mifflin

Download and reproduce the results

The CSV contains all inputs, all three predictions, the absolute spread, and the spread as a percentage of the Mifflin–St Jeor estimate. You may reuse it with attribution.

Download CSV (72 rows)

What we found

Formula choice changed the estimated resting requirement by 81.6 kcal/day on average in this grid. The largest gap was 258.3 kcal/day. These are differences between equations, not errors measured against indirect calorimetry.

The largest disagreements appeared in heavier, younger male profiles. That pattern follows from the different weight, height, and age coefficients in each equation; it does not establish which estimate is correct for an individual.

If the same activity multiplier were applied to every estimate, the maintenance-calorie gap would grow proportionally. At a multiplier of 1.55, an 100 kcal resting-energy gap becomes approximately 155 kcal/day. Activity multipliers introduce another independent source of uncertainty.

Largest differences in the grid

ProfileMifflinRevised H–BOriginal H–BSpread
Male, 20 y, 160 cm, 115 kg20552283.32313.3258.3 kcal
Male, 20 y, 175 cm, 115 kg2148.82355.32388.4239.6 kcal
Male, 40 y, 160 cm, 115 kg19552169.82178.2223.2 kcal
Male, 20 y, 190 cm, 115 kg2242.52427.32463.4220.9 kcal
Male, 40 y, 175 cm, 115 kg2048.82241.82253.3204.5 kcal
Male, 60 y, 160 cm, 115 kg18552056.22043.1201.2 kcal
Male, 40 y, 190 cm, 115 kg2142.52313.72328.3185.8 kcal
Male, 20 y, 160 cm, 95 kg18552015.42038.3183.3 kcal

Method

We generated a complete grid from two equation sex constants, ages 20/40/60 years, heights 160/175/190 cm, and weights 55/75/95/115 kg. That produced 72 profiles. No personal records, patient data, or tracking data were used.

For each profile, we calculated predicted kcal/day using all three equations. “Spread” is the highest prediction minus the lowest. Values were calculated at full precision and rounded to one decimal place for publication. The implementation that generates this page also generates the CSV, preventing the displayed analysis and download from drifting apart.

Mifflin–St Jeor

Published 1990

Male constant

10W + 6.25H − 5A + 5

Female constant

10W + 6.25H − 5A − 161

Revised Harris–Benedict

Published 1984

Male constant

88.362 + 13.397W + 4.799H − 5.677A

Female constant

447.593 + 9.247W + 3.098H − 4.330A

Original Harris–Benedict

Published 1918/1919

Male constant

66.473 + 13.752W + 5.003H − 6.755A

Female constant

655.096 + 9.563W + 1.850H − 4.676A

W = weight in kg, H = height in cm, A = age in years.

Limitations

  • This is a formula comparison using synthetic profiles, not a clinical validation study.
  • It does not compare predictions with measured resting energy expenditure.
  • The grid is deliberately regular and is not representative of any national population.
  • The equations use sex-specific constants and do not model body composition, pregnancy, illness, medication, or athletic status.
  • Individual energy needs can differ materially from every estimate shown here.

Primary sources

  1. Mifflin MD et al. A new predictive equation for resting energy expenditure in healthy individuals.PubMed 2305711.
  2. Roza AM, Shizgal HM. The Harris Benedict equation reevaluated: resting energy requirements and the body cell mass.PubMed 6741850.
  3. Harris JA, Benedict FG. A biometric study of human basal metabolism.PubMed 16576330.