Mifflin-St Jeor vs Katch-McArdle: Which TDEE Formula Wins?

Mifflin-St Jeor wins for the general population, predicting resting metabolic rate within 10 percent of measured values in more adults than any other simple equation; Katch-McArdle wins for trained athletes and lean individuals with a recent body composition scan. The two equations ask different questions. Mifflin needs only height, weight, age and sex. Katch-McArdle needs lean body mass. As of August 2026, the Academy of Nutrition and Dietetics Evidence Analysis Library still recommends Mifflin-St Jeor with actual body weight as the default when indirect calorimetry is not available.

Calorie tracking apps quietly choose one of these formulas for you and call the output your TDEE. Most pick Mifflin-St Jeor because it is the most validated. A handful, especially those aimed at lifters, default to Katch-McArdle. The choice matters: the same person can see TDEE estimates 150-200 kcal apart depending on which equation runs in the background.

This guide compares both formulas, shows the math step-by-step in Australian (kg, cm, kJ) and US (lb, in, kcal) units, summarises what research suggests about accuracy, and explains which to use for your body. We will not skip the algebra most apps hide.

What is the difference between Mifflin-St Jeor and Katch-McArdle?

Mifflin-St Jeor estimates basal metabolic rate (BMR) from total body weight, height, age and sex. Katch-McArdle estimates it from lean body mass alone. Mifflin treats every kilogram the same. Katch-McArdle assumes fat tissue is almost metabolically silent and only credits your fat-free mass.

The two equations come from different research programmes. Mifflin-St Jeor was published in The American Journal of Clinical Nutrition in 1990 by Mifflin, St Jeor, and colleagues, who measured resting energy expenditure by indirect calorimetry in 498 healthy adults aged 19 to 78 and built a regression model with R² = 0.71. Katch-McArdle traces back to lean-mass research by Frank Katch and William McArdle and is mathematically identical to the older Cunningham formula.

Here are the two formulas side by side in both metric and US units.

The big practical difference: Mifflin-St Jeor does not need a body fat reading. Katch-McArdle does. If your only body fat estimate comes from a smart scale or a guess, Katch-McArdle inherits that error directly into your BMR.

How accurate is each TDEE formula in 2026?

A 2005 systematic review in the Journal of the American Dietetic Association found Mifflin-St Jeor predicted resting metabolic rate within 10 percent of measured values more often than any competing equation, and with the narrowest error range. A 2023 athlete meta-analysis found lean-mass-based equations like Katch-McArdle and Cunningham performed better in low-body-fat populations.

The Frankenfield 2005 review evaluated Mifflin-St Jeor against Harris-Benedict, Owen, and WHO/FAO/UNU equations across multiple studies of nonobese and obese adults. Mifflin was the most reliable in both groups. The Academy of Nutrition and Dietetics adopted this as their official recommendation for clinical practice and reaffirmed it in their adult weight management guideline.

That picture flips in athletes. A 2023 systematic review and meta-analysis in Sports Medicine by O'Neill, Corish and Horner pooled 29 studies, 1,430 athletes (822 women, 608 men) and 100 different prediction equations. Mifflin-St Jeor missed in both directions here and was listed among the equations that consistently underperform in athletes, while Cunningham (mathematically identical to Katch-McArdle) produced predictions that did not differ significantly from measured values. On precision, Cunningham landed within ±10 percent of measured RMR for 54.1 percent of athletes and Mifflin-St Jeor for 52.2 percent, while the Ten-Haaf equation reached 80.2 percent, the most precise of the 100 equations reviewed.

Accuracy also drops in specific populations. A 2022 validation study in older adults with obesity (mean age 73, BMI ≥30) found Mifflin-St Jeor under-predicted measured RMR by 548 kcal on average. That study favoured WHO/FAO/UNU and Harris-Benedict for that population, though research suggests no simple equation handles older obese adults well.

Newer validation work keeps landing in the same place for the general population. A 2025 study in Archives of Public Health measured RMR by indirect calorimetry in 148 healthy young women and tested nine prediction equations against it, including a new population-specific one built from the same data. Mifflin-St Jeor came out the most accurate of the nine, with a group-level bias of 2.9 percent and 60.1 percent of individuals predicted within ±10 percent of measured RMR, rising to 83.3 percent in the participants with obesity. The purpose-built local equation matched it but did not beat it, which is the pattern most challengers to Mifflin have followed for three decades.

Here is a rough accuracy snapshot:

How do you calculate TDEE step-by-step in AU and US units?

You calculate BMR first, then multiply by an activity factor between 1.2 (sedentary) and 1.9 (very heavy training). Convert the kcal result to kilojoules for AU labels by multiplying by 4.184. Most calorie apps hide both the activity factor and the BMR step. Doing the math by hand once is the best way to spot where your app is wrong.

The standard activity multipliers, popularised after the original Katch & McArdle textbook and now used across the calorie-tracking industry, are:

1. Sedentary (1.2): desk job, little or no exercise.
2. Lightly active (1.375): light exercise or sport 1-3 days a week.
3. Moderately active (1.55): moderate exercise 3-5 days a week.
4. Very active (1.725): hard exercise 6-7 days a week.
5. Extra active (1.9): physical job plus hard daily training, or two-a-day sessions.

Most apps overestimate this factor. If you commute by car, sit at a desk, and lift three times a week, you are probably between 1.375 and 1.55, not 1.725. The math compounds: misreading by one tier inflates daily calories by 250-400 kcal, which is enough to stall a calorie deficit entirely.

Worked example: AU units (kilojoules and centimetres)

Imagine you are a 35-year-old woman, 165 cm, 70 kg, moderately active (PAL ≈ 1.55), with a DEXA body fat reading of 28 percent.

Mifflin-St Jeor:

• BMR = (10 × 70) + (6.25 × 165) − (5 × 35) − 161
• BMR = 700 + 1,031.25 − 175 − 161 = 1,395 kcal/day
• TDEE = 1,395 × 1.55 ≈ 2,162 kcal/day
• In kilojoules: 2,162 × 4.184 ≈ 9,046 kJ/day

Katch-McArdle:

• Lean body mass = 70 × (1 − 0.28) = 50.4 kg
• BMR = 370 + (21.6 × 50.4) = 370 + 1,088.6 = 1,458 kcal/day
• TDEE = 1,458 × 1.55 ≈ 2,260 kcal/day
• In kilojoules: 2,260 × 4.184 ≈ 9,456 kJ/day

The two formulas land within 100 kcal of each other here, well inside normal day-to-day variation. For a typical Australian adult of average body composition, both will guide reasonable food intake.

Worked example: US units (kcal, pounds, inches)

Now consider a 28-year-old man, 5 ft 10 in (178 cm, 70 in), 175 lb (79.4 kg), training hard six days a week (PAL ≈ 1.725), with a body fat reading of 10 percent from a hydrostatic scan.

Mifflin-St Jeor (metric form):

• BMR = (10 × 79.4) + (6.25 × 178) − (5 × 28) + 5
• BMR = 794 + 1,112.5 − 140 + 5 = 1,772 kcal/day
• TDEE = 1,772 × 1.725 ≈ 3,057 kcal/day

Katch-McArdle:

• Lean body mass = 79.4 × (1 − 0.10) = 71.5 kg
• BMR = 370 + (21.6 × 71.5) = 370 + 1,544.4 = 1,914 kcal/day
• TDEE = 1,914 × 1.725 ≈ 3,302 kcal/day

The gap here is 245 kcal/day. Over a year that is roughly 12 kg of food intake difference. For this athletic profile, Katch-McArdle is the more defensible estimate because Mifflin penalises every kilogram of his bodyweight equally, including muscle.

If you do not have a DEXA, hydrostatic, or BodPod reading, smart-scale body fat is too noisy to feed into Katch-McArdle reliably. Skinfold calipers used by a trained practitioner are an acceptable middle ground.

Which TDEE formula should you use for your body?

Use Mifflin-St Jeor if you do not have a recent and accurate body fat percentage, which covers most people. Use Katch-McArdle if your body fat is below roughly 20 percent and you have a DEXA, hydrostatic, BodPod, or trained-skinfold reading from the last six months.

Five practical decision rules:

1. No body fat reading? Use Mifflin-St Jeor. Garbage in, garbage out. Mifflin-St Jeor avoids the lean body mass guess entirely.
2. Athletic build, body fat under 15-20%, recent DEXA scan? Use Katch-McArdle. This is where Mifflin-St Jeor is least reliable, missing in both directions, while Cunningham and Katch-McArdle track measured values closely.
3. Average build, body fat 20-30%? Either works. The two formulas typically agree within 100 kcal/day in this range.
4. Higher body fat (over 35%)? Use Mifflin-St Jeor with actual (not adjusted) weight. The Academy of Nutrition and Dietetics evidence review specifically recommends this for overweight and obese adults.
5. Older adult over 70? Treat any formula as a rough estimate. Validation in this group is thin and predictions can be off by 500+ kcal. Indirect calorimetry is the only accurate option if available.

These are the same rules registered dietitians apply, with one twist: most apps do not ask which formula you want. kcalm defaults to Mifflin-St Jeor and lets you adjust based on tracked outcomes rather than swapping equations.

Why do calorie apps disagree about your TDEE?

Two apps using the same Mifflin-St Jeor equation can produce TDEE estimates 200-400 kcal apart for the same person because they pick different activity multipliers, round inputs differently, or apply hidden adjustments for goal weight. The formula is the easy part. The activity factor is where most of the error lives.

Some apps cap activity at 1.55 even if you select "very active". Others bake a 200 kcal deficit into "maintenance" by default. A few use older Harris-Benedict revisions instead of Mifflin and never tell you. If your app shows you a single number with no working, you cannot tell whether the disagreement is the equation, the activity factor, or a hidden goal adjustment.

The fix is simple: calculate your BMR by hand once, decide your activity multiplier honestly, and treat the app's number as a starting estimate. Track for two weeks at the resulting calorie target and compare actual weight change against the predicted change. If your weight is flat at a supposed 500 kcal deficit, your real maintenance is probably 500 kcal lower than the app says. This calibration matters more than the formula choice, especially during a sustained cut, where metabolic adaptation can pull TDEE down further.

For the deeper science behind the Mifflin equation specifically, see our Mifflin-St Jeor explainer. For practical macro splits once your TDEE is set, see the body recomposition guide.

Frequently Asked Questions

Is Katch-McArdle more accurate than Mifflin-St Jeor?

Not for most people. Research suggests Katch-McArdle wins in trained athletes and lean individuals because Mifflin penalises every kilogram of bodyweight equally, including muscle. In the general population, Mifflin-St Jeor predicts resting metabolic rate within 10 percent of measured values more often than Katch-McArdle, and the Academy of Nutrition and Dietetics adopted Mifflin as the default recommendation in 2014. Katch-McArdle only outperforms when your body fat percentage is measured accurately, usually via DEXA, hydrostatic weighing, or BodPod.

Do Australian dietitians use Mifflin-St Jeor or NHMRC reference values?

Most use both. The NHMRC Estimated Energy Requirements published in the Nutrient Reference Values for Australia and New Zealand are population-level guidance derived from FAO/WHO methodology and reported in kilojoules. For individual clients, Australian Accredited Practising Dietitians commonly use Mifflin-St Jeor for BMR and then apply a physical activity level (typically 1.4 to 1.75) consistent with NHMRC guidance, before converting the final number to kilojoules.

What is the difference between BMR, RMR, and TDEE?

BMR is the energy your body burns at complete rest, measured under tightly controlled conditions including overnight fasting and a 24-hour rest. RMR is a slightly looser version measured after a few hours rest and a short fast; it is typically 5-10 percent higher than true BMR. TDEE adds the calories you burn from movement, exercise, and the thermic effect of food on top. In practice, calorie apps treat BMR and RMR as interchangeable, then multiply by an activity factor to estimate TDEE.

How accurate is a smart scale's body fat reading for Katch-McArdle?

Not accurate enough to rely on. Bioelectrical impedance smart scales typically show errors of 4 to 8 percentage points compared to DEXA, and the error increases at very low or very high body fat. Feeding a wrong body fat percentage into Katch-McArdle propagates that error directly into BMR. If your only reading comes from a household scale, Mifflin-St Jeor is the safer default.

How often should you recalculate your TDEE?

Recalculate any time your weight changes by more than 3-5 kg, or every 8-12 weeks during an active diet phase. Both Mifflin-St Jeor and Katch-McArdle use weight as a direct input, so a lighter bodyweight produces a lower predicted BMR. Recalculating also surfaces metabolic adaptation, where measured TDEE falls below the formula's prediction. For people coming off a long deficit, see our reverse dieting guide for how to raise calories without rapid weight regain.

Can kcalm switch between Mifflin-St Jeor and Katch-McArdle?

kcalm uses Mifflin-St Jeor by default to match what registered dietitians recommend for the general population. The app calibrates your maintenance calories from your actual logged intake and weight trend over the first 14 days, which corrects for activity-factor errors regardless of which formula generated the starting estimate. If you have a recent DEXA reading and want to start from Katch-McArdle instead, the TDEE calculation guide shows the math.

Can you run both formulas without doing the math yourself?

Yes. A BMR calculator that runs Mifflin-St Jeor and Katch-McArdle side by side needs only height, weight, age and sex, and unlocks the lean-mass equation once you enter a body fat percentage. Running both results through a TDEE calculator with every activity multiplier then shows how far apart the two daily targets sit before you commit to either number.

Does Katch-McArdle use a different activity multiplier than Mifflin-St Jeor?

No. Both use the same 1.2 to 1.9 scale, which is exactly why the multiplier matters more than the equation. In the athletic worked example above, the two formulas differ by 142 kcal at the BMR stage, but multiplying by 1.725 stretches that gap to 245 kcal per day. Choose your activity tier honestly first, then worry about the formula.

Sources

1. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51(2):241-247. https://pubmed.ncbi.nlm.nih.gov/2305711/
2. Frankenfield D, Roth-Yousey L, Compher C. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. J Am Diet Assoc. 2005;105(5):775-789. https://pubmed.ncbi.nlm.nih.gov/15883556/
3. O'Neill JER, Corish CA, Horner K. Accuracy of Resting Metabolic Rate Prediction Equations in Athletes: A Systematic Review with Meta-analysis. Sports Med. 2023;53(12):2373-2398. https://pmc.ncbi.nlm.nih.gov/articles/PMC10687135/
4. Porter J et al. Validation of resting energy expenditure equations in older adults with obesity. Clin Nutr ESPEN. 2022. https://pmc.ncbi.nlm.nih.gov/articles/PMC9761489/
5. National Health and Medical Research Council. Dietary energy — Nutrient Reference Values for Australia and New Zealand. eatforhealth.gov.au. https://www.eatforhealth.gov.au/nutrient-reference-values/nutrients/dietary-energy
6. Academy of Nutrition and Dietetics. Adult Weight Management Evidence Analysis Library: RMR equations. https://www.andeal.org/template.cfm?template=guide_summary&key=4341
7. Academy of Nutrition and Dietetics. Determination of Resting Metabolic Rate: Evidence Analysis Library recommendation. https://www.andeal.org/template.cfm?template=guide_summary&key=621
8. Validity of RMR equations in underweight, normal weight, overweight, and obese Emirati female young adults. Arch Public Health. 2025;83. https://link.springer.com/article/10.1186/s13690-025-01727-9