TDEE, BMR, and Macros Explained — The Math Behind Calculators

A friend texted me last week: "MyFitnessPal says 1,847 calories. Some Reddit calculator says 2,310. Which one is right?" Both, actually. They're using the same equation; one assumed she was sedentary and the other assumed she walked her dog. That 463-calorie gap isn't a mystery — it's an activity multiplier. Open the hood of any TDEE calculator and you'll find the same three pieces wired together: a basal rate, an activity factor, and a macro split. Once you can read those numbers, the "which app should I trust" question mostly answers itself.

This post walks through the math your fitness tracker is hiding — what each variable means, where the numbers come from, and what to do when calculators disagree.

The Worked Example That Anchors Everything

Let me start with a concrete person and use her numbers throughout: Sara, 32 years old, 165 cm, 64 kg, female, runs three times a week. Goal: lose fat slowly.

By the end of this post you'll see exactly how a calculator turns those five inputs into:

BMR: 1,387 kcal (her metabolism if she lay in bed all day)
TDEE: 2,150 kcal (what she actually burns living her life)
Cut target: 1,720 kcal (TDEE minus a 20% deficit)
Macros: 128 g protein, 172 g carbs, 57 g fat

If you want to follow along with your own numbers, the TDEE Calculator does the arithmetic, but the math is short enough to do on a phone.

BMR: The Floor of Your Calorie Budget

Basal metabolic rate is the energy your body burns just being alive — heart pumping, kidneys filtering, brain running, cells turning over. No movement, no digestion, no shivering. It's roughly 60-70% of total daily burn for most people, which is why it's the number every calculator computes first.

The current best-in-class estimator is the Mifflin-St Jeor equation, published in 1990 and validated against thousands of measured values. It looks like this:

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

Plugging Sara in:

BMR = 10(64) + 6.25(165) − 5(32) − 161
    = 640 + 1031.25 − 160 − 161
    = 1,350 kcal

(Most calculators round to whole numbers and may add a half-kilogram of body composition adjustment, which is why you'll see 1,350-1,400 depending on the implementation.)

Why Mifflin and not the older Harris-Benedict? Mifflin's data set was modern and more diverse, and head-to-head validation studies show it's about 5% closer to indirect calorimetry on average. Harris-Benedict (1919, recalibrated in 1984) is still floating around the internet — that's why some old calculators give numbers ~100 kcal higher than newer ones for the same person.

A few things BMR is not: it's not the same as resting metabolic rate (RMR is measured 3-4 hours after eating and is ~10% higher), and it doesn't account for body composition. Two 70 kg people, one with 25% body fat and one with 12%, have noticeably different real BMRs because lean mass burns more at rest than fat mass. If you have body fat data, the Katch-McArdle formula (BMR = 370 + 21.6 × lean_mass_kg) is more accurate. Use the Body Fat % Calculator (Navy Method) to estimate it without a DEXA scan.

TDEE: BMR Plus Everything You Actually Do

TDEE — Total Daily Energy Expenditure — is BMR plus the calories you burn moving, digesting, and adapting to temperature. It breaks into four buckets:

TDEE = BMR + EAT + NEAT + TEF

EAT (Exercise Activity Thermogenesis): structured workouts. Running, lifting, cycling. Usually 5-15% of total burn.
NEAT (Non-Exercise Activity Thermogenesis): everything else — fidgeting, walking to the kitchen, standing while you talk on the phone. Wildly variable. Office worker NEAT might be 200 kcal/day; a server on her feet for 8 hours might be 800.
TEF (Thermic Effect of Food): the calories burned digesting what you eat. ~10% of intake. Higher for protein, lower for fat.

You don't measure these directly. Instead, calculators bundle them into an activity multiplier applied to BMR:

Sedentary (desk job, no exercise):     BMR × 1.2
Lightly active (1-3 workouts/week):    BMR × 1.375
Moderately active (3-5 workouts/week): BMR × 1.55
Very active (6-7 workouts/week):       BMR × 1.725
Extra active (physical job + lifting): BMR × 1.9

Sara runs 3x/week, so "moderately active" fits:

TDEE = 1,350 × 1.55 = 2,093 kcal

This is where calculators most often disagree, because honest activity classification is hard. People underestimate sedentary time and overestimate workouts. A 30-minute jog three times a week is not "moderately active" if you're at a desk for the other 23.5 hours of each day — that's "lightly active." The honest version of Sara's number might be:

TDEE = 1,350 × 1.4 = 1,890 kcal

A 200 kcal swing is the difference between losing weight and not. When in doubt, pick the lower multiplier and adjust upward after two weeks of tracking actual results.

Goal Adjustments: Cut, Bulk, Maintain

TDEE tells you maintenance. To change body composition, you shift intake up or down:

Cut (fat loss):    TDEE × 0.80   (20% deficit)
Maintain:          TDEE × 1.00
Lean bulk:         TDEE × 1.10   (10% surplus)
Aggressive bulk:   TDEE × 1.20   (20% surplus)

For Sara cutting:

Target = 2,093 × 0.80 = 1,674 kcal

A 419 kcal/day deficit predicts ~0.4 kg/week of fat loss (since 1 kg of fat ≈ 7,700 kcal). Faster cuts work but cost lean mass and adherence. Slower cuts are easier to sustain. Anything past a 25% deficit is in "crash diet" territory and your body will fight back with reduced NEAT and hunger hormones.

For weight prediction over time, the math is forgiving in the short term and lying in the long term — your TDEE drops as you lose weight (smaller body, lower BMR), so a fixed calorie target produces diminishing returns. Recalculate every 4-6 kg of weight change.

Macros: Splitting the Calories Into Protein, Carbs, Fat

Once you have a calorie target, you allocate it across three macronutrients. Each gram of macronutrient delivers a fixed amount of energy:

Protein:  4 kcal/g
Carbs:    4 kcal/g
Fat:      9 kcal/g
Alcohol:  7 kcal/g  (technically a fourth, often ignored)

The strategy most evidence-based coaches use (Examine.com has the receipts) is:

Set protein first based on body weight, not calorie percentage.
Set fat second based on a minimum for hormonal health.
Fill the rest with carbs.

Protein

The USDA Dietary Guidelines RDA of 0.8 g/kg is a floor for not getting deficient — it's not optimal for body composition. Active people aiming for fat loss or muscle gain do better at:

1.6 - 2.2 g/kg of bodyweight

For Sara at 64 kg, that's 102-141 g protein. Pick the upper end during a cut (more satiety, more lean mass retention): ~128 g protein → 512 kcal.

Fat

A practical floor is 0.6 g/kg — enough for hormones, vitamin absorption, satiety. Some people thrive at 1.0+ g/kg; others prefer lower fat with more carbs. Both work.

For Sara, going moderate at 0.9 g/kg: ~57 g fat → 513 kcal.

Carbs

Whatever's left.

Remaining kcal = 1,674 − 512 − 513 = 649 kcal
Carbs = 649 ÷ 4 = 162 g

Final split for Sara on a cut: 128 g protein, 162 g carbs, 57 g fat → 1,674 kcal. The Macro Calculator does this allocation in one step and lets you nudge the protein/fat ratios.

Why Two Calculators Give You Different Numbers

Now you can diagnose the 1,847 vs 2,310 disagreement my friend ran into:

Different BMR equation. Mifflin-St Jeor vs Harris-Benedict vs Katch-McArdle. ~5-10% spread.
Different activity multipliers. Some apps use 1.2 / 1.375 / 1.55, others use 1.3 / 1.5 / 1.7. ~10% spread.
Different default goal. Some default to maintenance, some to a 15% cut, some to a 25% cut. ~25% spread on its own.
Body fat input. Calculators that ask for body fat use Katch-McArdle and skip the activity assumption error.
Rounding and unit conversion. Lbs vs kg, cm vs inches — small but real.

The fix isn't to trust the calculator with the highest accuracy claim. It's to track for two weeks, then adjust. Start with any reasonable estimate. Weigh yourself the same time each morning. If trend weight isn't moving in the direction you want after 14 days, adjust intake by ±150 kcal and repeat.

What Calorie Math Doesn't Cover

A few things even good calculators don't capture:

Hydration. Underdrinking blunts performance and skews scale weight. The Water Intake Calculator gives a sane default based on bodyweight and climate.
Body composition. Two 64 kg women with identical TDEE can look completely different. Pair calorie targets with strength training and the Ideal Weight Calculator to anchor expectations to multiple medical formulas, not one number on a wall chart.
BMI alone is a blunt instrument. A muscular athlete and a sedentary person can have the same BMI. Use it as a sanity check, not a verdict.
Cardio intensity matters. A 45-minute Zone 2 jog and a 45-minute interval session burn different calories and produce different adaptations. The Heart Rate Zone Calculator (Karvonen) maps target HR to your reserve, which is more accurate than the "220 minus age" shortcut on most treadmill consoles.

The Practical Takeaway

If you only remember three things:

BMR is just the floor. It tells you what your body costs to keep idling. The interesting number is TDEE.
The activity multiplier is the biggest source of error. Be honest, lean low, recalibrate after two weeks of real data.
Set protein first, fat second, carbs last. Aim for 1.6-2.2 g/kg protein on a cut. Everything else is preference.

Calorie math is approximate by nature — bodies aren't bomb calorimeters and "average daily activity" hides enormous variability. Use the calculator to get a starting target, then let the scale and the mirror tell you whether to adjust. The math is the map, not the territory.

FAQ

Why do MyFitnessPal and Cronometer give me different TDEE numbers?

Almost always different activity multipliers and BMR equations. MyFitnessPal historically used Mifflin-St Jeor with conservative multipliers; Cronometer offers multiple equations and tighter activity tiers. The 200-500 kcal disagreement between apps for the same person is normal. Pick one, track for two weeks, then adjust based on actual scale trends rather than which app you trust more.

Should I use Mifflin-St Jeor or Katch-McArdle?

Mifflin-St Jeor if you don't know your body fat percentage; Katch-McArdle if you do. Katch-McArdle is more accurate because it uses lean mass directly (BMR = 370 + 21.6 × lean_mass_kg), which avoids the assumption that two 70 kg people have the same metabolism. Without body fat data (DEXA, BodPod, or Navy method), Mifflin-St Jeor is the modern default and beats the 1919-era Harris-Benedict by ~5%.

Why does my weight loss stall after a few weeks?

Because your TDEE drops as you lose weight — a smaller body has lower BMR, and people unconsciously reduce NEAT (fidgeting, walking) on a cut. A 419 kcal deficit on day one becomes 250 kcal four weeks in, even at the same intake. The fix is to recalculate every 4-6 kg of weight change and adjust either calories down or activity up. Plateaus are math, not metabolism damage.

Is 1.6-2.2 g/kg protein really necessary?

For body composition during a cut or muscle gain, yes — meta-analyses consistently find better lean mass retention at 1.6+ g/kg compared to the 0.8 g/kg RDA floor. The RDA prevents deficiency; it doesn't optimize. Sedentary maintenance can stay around 1.2 g/kg without issues. Above 2.5 g/kg shows diminishing returns; the upper end of the 1.6-2.2 range is where research consistently lands.

Should I count alcohol as carbs or fat?

Neither — alcohol is its own macro at 7 kcal/g, and it's not a great fuel source. Most calculators ignore it because it's hard to track accurately and rarely consumed in trackable amounts. If you drink regularly, allocate the calories from alcohol against your daily target (a glass of wine ≈ 125 kcal, a beer ≈ 150 kcal) and reduce carbs/fat to match. Don't add it on top of your existing macros.

What's the difference between BMR and RMR?

BMR (basal metabolic rate) is measured under strict laboratory conditions: post-absorptive, supine, awake, no recent activity. RMR (resting metabolic rate) is more practical — measured 3-4 hours after eating, sitting, but otherwise relaxed. RMR is typically 5-10% higher than BMR. Most "BMR calculators" actually estimate RMR; the practical difference is usually within calculator margin of error.

How do I handle workout-day vs rest-day calorie targets?

Two approaches: average everything (eat the same daily target whether you work out or not, since weekly totals matter) or cycle (higher carbs on training days, slightly lower on rest days). Averaging is simpler and works for most people. Cycling can improve workout performance but adds tracking complexity. Both produce identical body composition results when total weekly calories match.

Can I trust my smartwatch's calorie burn estimate?

Generally no — wrist-based optical sensors over-estimate calorie burn by 30-90% according to multiple validation studies. Heart rate strap data is more accurate but still has 15-25% error vs lab measurement. Treat watch-reported calories as a relative trend (today vs yesterday) rather than absolute truth. Don't add "calories burned" back to your daily target unless you've cross-validated with weight trends over 2+ weeks.