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

A Quick History

1919. James Arthur Harris and Francis Gano Benedict publish their equation, derived from calorimetry measurements of a few hundred subjects in the 1910s. For decades, it's the standard. Every hospital, every dietitian, every textbook.

1984. Roza and Shizgal revise the Harris-Benedict coefficients using additional data. The revised version (shown above) improves the fit somewhat, but it's still anchored to a population profile from an earlier era.

1990. Mifflin, St Jeor, and colleagues publish a new equation built from indirect calorimetry on about 500 contemporary adults. The sample spans ages 19 to 78 and includes overweight subjects. This matters: the people being measured finally looked like the people using the formula.

That's the core of the whole debate. Harris-Benedict describes bodies from a century ago. Mifflin-St Jeor describes bodies like yours.

Why Mifflin-St Jeor Predicts Better Today

People in 1919 were, on average, leaner and far more physically active than people today. No desk jobs at scale, no cars for every errand. A formula fitted to those bodies tends to overestimate the resting burn of a modern adult, typically by around 5%.

Validation research since the 1990s has repeatedly compared common equations against measured BMR. Mifflin-St Jeor predicts within 10% of measured values for more people than Harris-Benedict does, especially in the general (including overweight) population. That track record is why the Academy of Nutrition and Dietetics recommends Mifflin-St Jeor as the preferred equation for healthy adults.

An overestimate sounds harmless. It isn't. Say your calculator claims you burn 2,100 calories but you really burn 2,000. You'll eat 100 calories over budget every day while believing you're on plan. Over months, that's the difference between losing and plateauing, and you'll blame yourself instead of the formula.

Worked Example: How Big Is the Gap?

Take a 40-year-old woman, 165 lb (75 kg), 5'6" (168 cm).

Mifflin-St Jeor: BMR = 10 × 75 + 6.25 × 168 − 5 × 40 − 161 BMR = 750 + 1,050 − 200 − 161 = 1,439 calories

Harris-Benedict (revised): BMR = 9.247 × 75 + 3.098 × 168 − 4.330 × 40 + 447.593 BMR = 693.5 + 520.5 − 173.2 + 447.6 = 1,488 calories

Gap at the BMR level: about 49 calories. Small. But watch what the activity multiplier does to it.

The multiplier scales the error along with everything else. And for some body profiles, the BMR gap itself runs well over 100 calories before any multiplier touches it.

That second step is the one most people skip: BMR is only the resting figure, and the activity factor is what turns it into the number you actually eat against. Our guide to how TDEE turns BMR into your daily burn covers each factor and where people misjudge their own.

When the Difference Actually Matters

Let's be fair to Harris-Benedict. A 50-80 calorie difference is inside the noise of everyday eating. One extra tablespoon of peanut butter erases it. Cases where the choice of formula is nearly irrelevant:

• You're maintaining weight and adjusting by feel anyway.
• You track loosely and round your portions.
• You're going to calibrate against real scale data within a few weeks (which you should).

Cases where it genuinely matters:

• Small deficits. If you're running a gentle 275-calorie deficit, an 80-calorie overestimate wipes out nearly a third of it.
• Smaller, older, or less active bodies. Lower TDEEs leave less margin. An overestimate that's trivial at 2,800 calories is meaningful at 1,700.
• Higher body-fat percentages. This is where Harris-Benedict's overestimate tends to be largest, and it's exactly the population most likely to be using a calorie calculator to lose weight.
• Long timelines. A daily overestimate compounds. At 75 calories a day, that's about 0.7 lb (0.3 kg) of expected-but-missing loss per month.

Since a pound of fat stores roughly 3,500 calories (~7,700 per kg), small daily errors turn into visible scale differences within a couple of months.

The Verdict

Use Mifflin-St Jeor. It's built on modern bodies, validated more broadly, recommended by the leading dietetics body, and simpler to compute. Harris-Benedict isn't wildly wrong, but when both formulas are free, there's no reason to pick the one with a known bias toward overestimating.

For the winning formula on its own terms, our full guide to the Mifflin-St Jeor equation walks through the coefficients, worked examples for both sexes, and the cases where it still misses.

Two closing caveats, whichever formula you use:

1. Every formula is an estimate, accurate to about ±10% at best. Neither one sees your muscle mass, genetics, or hormones.
2. Real-world data outranks both. Eat at your calculated target for 2-3 weeks, watch the scale, and adjust by 100-200 calories if results don't match. Your body is the final calculator.

Want the Mifflin-St Jeor math done for you, including the activity multiplier and a safe deficit target? Run your stats through the free calorie calculator.

FAQ

Which formula do most calorie calculators use?

Quality modern calculators mostly use Mifflin-St Jeor, following the Academy of Nutrition and Dietetics recommendation. Some older tools still run Harris-Benedict, which is why two calculators can give different answers for identical inputs.

Is the Harris-Benedict equation obsolete?

Not obsolete, just outperformed. It still lands within 10% of measured BMR for many people. But it was derived from a 1919 population and tends to overestimate modern adults by around 5%, so there's rarely a reason to prefer it.

Why do the two formulas give me different numbers?

They were fitted to different populations decades apart, so their coefficients differ. For most adults, Harris-Benedict returns a slightly higher number, typically 40-120 calories more at the BMR level, with the gap widest at higher body weights.

This is general information, not medical advice. If you have a medical condition, talk to your doctor or a registered dietitian before changing your diet.