Mifflin-St Jeor vs Harris-Benedict: Which BMR Formula Wins

Updated on April 25, 2026

Two Formulas, One Question: How Many Calories Do You Actually Burn at Rest?

If you've ever plugged your stats into a calorie calculator online, there's a decent chance it used one of two equations — Mifflin-St Jeor or Harris-Benedict — without telling you which one or why it matters. The difference between them isn't just academic. On a 185-pound man, the two formulas can spit out BMR estimates that differ by 70 to 130 calories per day. Over a week, that's almost a pound of body weight in caloric miscalculation if you're not careful.

So which one should you trust? The honest answer is that it depends on who you are — but one formula holds a measurable edge for most people.

A Quick Primer: What BMR Actually Means

Basal Metabolic Rate is the number of calories your body needs to keep the lights on — breathing, circulation, cell repair, temperature regulation — while you're completely at rest. It doesn't include the energy cost of digesting food, moving around, or exercising. It's the floor of your caloric needs.

BMR is not the same as your Total Daily Energy Expenditure (TDEE). To get TDEE, you multiply BMR by an activity factor (typically ranging from 1.2 for sedentary to 1.9 for very active). But since both formulas feed into that multiplier, errors in BMR estimation compound downstream. Getting BMR right matters.

Harris-Benedict: The Old Guard

The original Harris-Benedict equations were published in 1919 by James Arthur Harris and Francis Gano Benedict after studying 239 subjects at the Carnegie Institution in Washington. For decades, it was the standard in clinical nutrition.

The original equations were revised in 1984 by Roza and Shizgal to correct for some data errors, and these revised versions are what most people mean when they say "Harris-Benedict" today. The formulas are:

  • Men: 88.362 + (13.397 × weight in kg) + (4.799 × height in cm) − (5.677 × age in years)
  • Women: 447.593 + (9.247 × weight in kg) + (3.098 × height in cm) − (4.330 × age in years)

The formula was developed on a relatively small, homogeneous sample of mostly lean, young adults in the early 20th century — a population that looks quite different from most people using online calculators today.

Mifflin-St Jeor: The Challenger

In 1990, M.D. Mifflin and S.T. St Jeor published a new set of equations derived from a study of 498 subjects — a larger, more diverse group that included both obese and non-obese individuals. The formulas:

  • Men: (10 × weight in kg) + (6.25 × height in cm) − (5 × age in years) + 5
  • Women: (10 × weight in kg) + (6.25 × height in cm) − (5 × age in years) − 161

Notice that the equations share the same structure — the difference lies in the constants. Mifflin-St Jeor uses a simpler multiplier for weight and assigns a flat offset (+5 for men, −161 for women) instead of Harris-Benedict's more elaborate intercept terms.

Head-to-Head: Running the Numbers

Let's put both formulas to work on three different profiles to see where they diverge.

Example 1 — 35-year-old woman, 5'5" (165 cm), 145 lbs (65.8 kg):

  • Harris-Benedict: 1,476 kcal/day
  • Mifflin-St Jeor: 1,421 kcal/day
  • Difference: 55 calories

Example 2 — 45-year-old man, 6'0" (183 cm), 210 lbs (95.3 kg):

  • Harris-Benedict: 2,033 kcal/day
  • Mifflin-St Jeor: 1,960 kcal/day
  • Difference: 73 calories

Example 3 — 52-year-old woman, 5'3" (160 cm), 175 lbs (79.4 kg):

  • Harris-Benedict: 1,569 kcal/day
  • Mifflin-St Jeor: 1,477 kcal/day
  • Difference: 92 calories

A pattern emerges quickly: Harris-Benedict consistently comes in higher. The gap grows with age and heavier body weights, which is precisely where the older formula's training data limitations become most apparent.

What the Research Actually Says About Accuracy

This isn't a debate that has to rely on theory. Multiple validation studies have compared both formulas against indirect calorimetry — the gold-standard method for measuring actual resting energy expenditure, which measures oxygen consumption and CO₂ production in a metabolic chamber or using a ventilated hood.

A widely cited 2005 study published in the Journal of the American Dietetic Association by Frankenfield et al. compared five predictive equations against measured REE in 200 adults. Mifflin-St Jeor predicted within 10% of measured values in 82% of participants. Harris-Benedict hit that target in only 67% of subjects and systematically overestimated intake needs — particularly in overweight and obese individuals.

A 2003 analysis in Obesity Research reached similar conclusions, finding that Harris-Benedict overestimated BMR by an average of 5% in obese subjects. In absolute terms, that 5% error translates to 100–150 calories per day for many people — enough to explain why someone following a "deficit" based on Harris-Benedict might see slower-than-expected fat loss.

The American Dietetic Association (now the Academy of Nutrition and Dietetics) updated its clinical guidelines to recommend Mifflin-St Jeor as the preferred equation for both obese and non-obese adults precisely because of this body of evidence.

Where Harris-Benedict Might Still Hold Its Own

That said, Harris-Benedict isn't worthless. A few situations where it may perform comparably or have practical advantages:

  • Lean, athletic individuals: Both formulas struggle somewhat with highly muscular bodies (neither accounts for body composition), but some practitioners report Harris-Benedict produces more reasonable estimates for very lean, muscular people where fat mass is a smaller proportion of total weight.
  • Historical consistency: If someone has been tracking their intake against Harris-Benedict estimates for years and has established what their actual maintenance looks like empirically, switching formulas mid-process adds noise without necessarily improving accuracy.
  • Legacy clinical settings: Some older clinical protocols were built around Harris-Benedict and include reference ranges calibrated to it. In those contexts, switching equations mid-protocol could introduce inconsistencies.

The Elephant in the Room: Neither Formula is Highly Accurate

Here's something that should temper expectations for both: a ±10% error margin is considered acceptable performance for a predictive BMR equation. That means for someone with a true BMR of 1,800 calories, "within 10%" means the formula might output anywhere between 1,620 and 1,980 and still be considered accurate. That's a 360-calorie window.

Both formulas also completely ignore body composition. Two men who are both 5'10", 180 pounds, and 40 years old will receive identical BMR estimates — even if one is 15% body fat and the other is 35% body fat. Muscle burns more calories at rest than fat, so the leaner individual genuinely has a higher BMR. The Katch-McArdle formula addresses this by using fat-free mass as the input variable, but it requires a reasonably accurate body fat percentage measurement to be useful.

Additionally, neither formula captures:

  • Hormonal status (thyroid function significantly affects BMR)
  • Ambient temperature adaptations
  • The thermal effect of specific medical conditions
  • Genetic variability in metabolic rate (which research suggests can vary by as much as 25% between individuals with similar demographics)

Which One Should You Use?

For most people — particularly those who are overweight, over 40, or who don't have precise body composition data — Mifflin-St Jeor is the better starting point. It was developed on a more representative sample, validated in more recent studies, and endorsed by the major clinical nutrition organizations.

If you're quite lean and muscular, the Katch-McArdle formula (which uses lean body mass) may outperform both, provided you have a reliable body composition measurement.

Regardless of which equation you start with, treat the number as a hypothesis, not a prescription. Spend 3–4 weeks eating at a calculated intake, log your weight daily and average it weekly, and compare the trend to what the math predicts. That feedback loop is how you calibrate the formula to your actual biology — because ultimately, your body's response is the most accurate measurement tool available.

The formulas give you a reasonable starting line. What you do at the starting line is up to you.