Open any TDEE calculator on the internet and you will probably be looking at the Mifflin-St Jeor equation. It is the default for a reason: peer-reviewed accuracy, simple inputs, and decades of validation data. But "most accurate predictive equation" is not the same as "perfect" — and if you cut or bulk by the wrong 300 kcal you can stall progress for months. Here is what the research actually shows.
The Equation, Briefly
Mifflin-St Jeor predicts Basal Metabolic Rate (BMR) — the energy your body burns at complete rest:
- Men: BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age) + 5
- Women: BMR = (10 × weight in kg) + (6.25 × height in cm) − (5 × age) − 161
Total Daily Energy Expenditure (TDEE) is then BMR multiplied by an activity multiplier ranging from 1.2 (sedentary) to 1.725 (six training days per week).
How Accurate Is the Number?
The Academy of Nutrition and Dietetics' systematic review compared every major predictive equation against indirect calorimetry (the gold-standard "metabolic cart") in healthy adults. The headline numbers:
- Mifflin-St Jeor: ±10% accuracy in roughly 82% of subjects.
- Harris-Benedict (original): ±10% in ~38% of subjects — overshoots BMR by an average of 5%.
- Katch-McArdle: ±10% in ~74% — good if you know your actual lean mass, useless if your body fat estimate is off.
Translation: for a 2,500 kcal/day target, Mifflin-St Jeor will usually land within ±250 kcal of your true number.
Where the Equation Breaks Down
The 18% who fall outside the ±10% window tend to share predictable characteristics:
- Highly muscular athletes — Mifflin uses only weight, not body composition. A 90 kg bodybuilder and a 90 kg sedentary adult get the same predicted BMR.
- Obese subjects (BMI > 30) — Mifflin tends to overestimate BMR; consider using lean-mass-adjusted weight.
- Older adults (60+) — Age coefficient flattens out around 60; Mifflin may underestimate BMR in active seniors.
- Sub-clinical hypothyroidism — Predicted BMR can be 10–15% higher than measured.
Activity Multipliers Are the Biggest Source of Error
This is the dirty secret of every TDEE calculator: BMR prediction is fairly tight, but the activity multiplier you select can be off by a full level. Most people overestimate. A "moderate" lifter who trains 4×/week and walks 5,000 steps daily is closer to 1.4 than the textbook 1.55. Three-quarters of TDEE errors trace back to this single dropdown.
Pragmatic fix: start with one bracket lower than what feels right, then adjust based on 14-day bodyweight trend.
How to Calibrate Your Personal TDEE
- Run your numbers in our TDEE Calculator.
- Eat at the predicted maintenance for 14 days, weighing daily and averaging.
- Compare your 14-day average to baseline:
- ± 0.5 kg → your TDEE estimate is on target.
- Gained 1+ kg → real TDEE is ~250 kcal lower than predicted.
- Lost 1+ kg → real TDEE is ~250 kcal higher than predicted.
- Adjust intake by the offset, repeat.
Two weeks of careful tracking beats any equation in existence.
Bottom Line
Mifflin-St Jeor is the right place to start. It is the most accurate predictive equation available without a $1,200 metabolic cart. But treat the output as a hypothesis, not a verdict — and let two weeks of real-world weight data calibrate it for you.