Basal Metabolic Rate Chart: The Right Way to Read One

The number one mistake with a basal metabolic rate chart: treating it as your target. Those tidy tables that show “male, 30-39, 1,700 kcal” are population averages, not personal prescriptions. If you build a diet on the chart’s midpoint, you’re eating for a statistical composite of people who don’t look like you.

The fix isn’t to abandon charts. It’s to generate one that actually fits you – then use it as a rolling reference, not a fixed target.

The takeaway first

Use the Mifflin-St Jeor equation, not a lookup chart. Recalculate every time your weight moves by more than 2-3 kg. And know its accuracy ceiling: it’s within 10% of your real BMR for roughly 4 out of 5 non-obese people – and fewer than that for obese individuals or women specifically. More on those numbers below.

Why every BMR chart is a lie (kind of)

The classic DuBois-style chart – the one reprinted across a hundred health sites – buckets people by age and gender and hands you an average. According to Cleveland Clinic’s BMR reference, the average male sits around 1,696 calories per day and the average female around 1,410. Those figures are useful for population health reports. For you specifically, they’re a coin flip.

Your BMR is shaped by lean muscle mass, hormones, medications, past dieting, illness, and even ambient temperature. BMR accounts for roughly 60-70% of the calories you burn each day (per Garnet Health), so an error here cascades through every diet plan you build on top of it.

Charts do get one thing right: BMR runs 5-10% lower for women than men across most age groups. Beyond that, the age bracket is doing very little work.

Method A vs Method B: which formula wins?

Two equations dominate. Here’s the head-to-head.

The 2005 Frankenfield systematic review tested the four most commonly used clinical equations – Harris-Benedict, Mifflin-St Jeor, Owen, and WHO/FAO/UNU – and Mifflin came out ahead: lowest error rate, best coverage at the 10% accuracy threshold. That paper is still the reference nutrition software cites today.

There’s a third contender – Katch-McArdle – that uses lean body mass instead of total weight. It can be more accurate for lean, muscular people. It also inherits the error of whatever tool measured your body fat, which can throw your TDEE target off by more than 100 kcal/day. More on that in the edge cases.

The formula and a worked example

Mifflin-St Jeor, per the NutriAdmin reference:

Men: BMR = (10 × weight_kg) + (6.25 × height_cm) - (5 × age) + 5
Women: BMR = (10 × weight_kg) + (6.25 × height_cm) - (5 × age) - 161

The only difference between the two lines is the final constant. Everything else is identical.

Worked example

A 35-year-old woman, 72.7 kg, 167.6 cm tall:

BMR = (10 × 72.7) + (6.25 × 167.6) - (5 × 35) - 161
 = 727 + 1,047.5 - 175 - 161
 = 1,438.5 kcal/day

That’s her baseline – the calories she’d burn lying still for 24 hours. To get total daily energy expenditure (TDEE), multiply by an activity factor: sedentary 1.2, lightly active 1.375, moderately active 1.55, very active 1.725, or extremely active 1.9.

Turn it into a chart with ChatGPT

Static chart → living one. Paste this into ChatGPT or Claude (as of mid-2025, both handle this well):

Using the Mifflin-St Jeor equation, generate a BMR table for me at
weights from 65 kg to 80 kg in 1 kg increments. I'm a 35-year-old
woman, 167.6 cm tall. Also show TDEE at activity factor 1.375.
Return as a markdown table.

Now you have a chart that maps your BMR to your weight trajectory. Drop a row, gain a row – no re-doing arithmetic.

Pro tip: Ask the AI to add TDEE at 500 kcal deficit and 500 kcal surplus in the same table. Three columns of usable numbers from one prompt, and it kills the excuse to “eyeball” your intake later.

Edge cases nobody warns you about

The accuracy ceiling – and two different ways to hit it

“Mifflin-St Jeor is the most accurate” gets repeated everywhere and stopped before the data actually starts. Here’s what the research shows. The 2005 Frankenfield review put the within-10% accuracy at 82% for non-obese subjects – meaning roughly 1 in 5 people using the formula get a materially wrong number without knowing it. A separate 2020 study focused specifically on women found the within-10% rate dropped further to 71% – closer to 1 in 3 women getting a bad estimate. Two different populations, two different failure rates. The formula is the best available. It’s still a probability.

Katch-McArdle’s hidden failure mode

The math looks clean: BMR = 370 + (21.6 × lean body mass in kg). The problem is the input. Lean body mass requires knowing your body fat percentage accurately. Testing by CalcTypes found an 8 kg error in lean body mass produces roughly a 173 kcal/day BMR error – not small for calorie planning. If you’re getting body fat from a bathroom BIA scale, that number moves with your hydration and last meal. You’re building a precise-looking formula on top of a noisy input.

Aggressive dieting invalidates the chart

Here’s the one that catches serious dieters. Starvation can reduce BMR by as much as 30% (per calculator.net’s BMR reference, citing established metabolic research). Months into a hard cut, your real BMR may be running well below what any formula predicts. The chart hasn’t changed. Your body has.

The validation gap

The Frankenfield review flagged something calculator sites rarely mention: older adults and ethnic minorities were underrepresented in both the development and validation studies for Mifflin-St Jeor, so accuracy may be lower for these groups. If you’re 70+ or from an under-studied population, treat the output as a rougher starting point and calibrate from real-world feedback – weight trends tracked over 3-4 weeks.

What to actually do with your number

Track your weight daily for 2-3 weeks while eating what your chart predicts is maintenance. Weight trends flat? The estimate is in the right ballpark. Drifting up or down? Adjust by 100 kcal/day and re-observe for another two weeks.

BMR is a hypothesis you test, not a fact you obey.

FAQ

Is BMR the same as RMR?

No – RMR runs about 10-20% higher because it’s measured under looser conditions (no overnight fast, no lab). Most online calculators say “BMR” but technically output something closer to RMR. Fine for planning purposes; just don’t compare your calculator output directly to clinical BMR measurements.

My smart scale gives me a BMR number. Should I trust it over the formula?

Probably not. Smart scales estimate BMR from bioelectrical impedance readings that shift with hydration, recent meals, and even foot moisture. Test it yourself: weigh before and after a glass of water and watch the “BMR” reading move. The Mifflin-St Jeor formula uses inputs that don’t change hour to hour – height, age, sex, stable body weight – which makes it more consistent even if it isn’t more precise on any given day.

How often should I recalculate?

Every 2-3 kg of weight change. Age alone isn’t worth an annual update – the formula shifts by only 5 kcal/day per year of aging.

Your next step

Open ChatGPT, paste the prompt above with your actual numbers, and generate a table spanning 5 kg above and below your current weight. Save it. That’s your personal BMR chart – the one that actually applies to you.