Your maintenance calories and a target for your goal, calculated three ways so you can see how much the formulas disagree. Deficits are set as a percentage of your own expenditure rather than a fixed number — which is why a small person and a large athlete need different absolute deficits for the same rate of change.
Calorie Calculator
LiveMost people overestimate this. If you train hard four times a week but sit the rest of the time, "lightly active" is usually closer than "very active".
The formulas
Why a percentage beats a fixed deficit
Most calculators offer fixed options — "lose 1 lb a week" subtracts around 500 calories, "lose 2 lb" subtracts 1,000. That works for an average-sized person and fails badly at the edges:
| Person | Fixed −1,100/day | 20% deficit |
|---|---|---|
| Smaller woman, TDEE 1,650 | 550 kcal — unsafe | 1,320 kcal |
| Average adult, TDEE 2,400 | 1,300 kcal — steep | 1,920 kcal |
| Large athlete, TDEE 3,500 | 2,400 kcal — fine | 2,800 kcal |
The same absolute deficit is trivial for one person and dangerous for another. A percentage scales with the individual, which is why sports nutrition guidance is written that way and why this calculator uses it.
The calculator also applies a floor: 1,500 calories for men and 1,200 for women. Below those levels it becomes difficult to meet micronutrient requirements from food, and sustained intake beneath them belongs under medical supervision rather than a web page.
A realistic rate of change
| Rate per week | What it means | Trade-off |
|---|---|---|
| 0.25–0.5% of body weight | Gradual — 10% deficit | Easiest to sustain, preserves muscle and training performance best. Slow enough that progress is hard to see week to week |
| 0.5–1% of body weight | Standard — 20% deficit | The range most guidance recommends. Visible progress with manageable hunger |
| Above 1% | Faster — 25%+ deficit | More muscle loss, worse training quality, higher dropout. Appropriate short-term for people with substantial fat to lose, not as a default |
Note that the rate is proportional to body weight, not absolute. Someone at 110 kg can safely lose over 1 kg a week; someone at 55 kg cannot, and attempting it costs muscle rather than fat. The calculator reports your rate in both kilograms and percentage of body weight so you can see which band you are in.
Which BMR formula should you use?
| Formula | Best for | Notes |
|---|---|---|
| Mifflin-St Jeor | Almost everyone | Published 1990, and validation studies consistently find it the most accurate of the general equations. The sensible default |
| Harris-Benedict | Comparison only | Dates to 1919, revised 1984. Tends to overestimate by 5–10% against modern measurement, partly because the original population differed from today's |
| Katch-McArdle | Lean, trained people | Works from lean mass rather than total weight, so it does not penalise muscularity. Requires a body fat figure — and inherits that figure's error |
The calculator shows all three where it can. When they disagree by more than about 150 calories, that spread is itself the useful information — it tells you how much uncertainty sits in the estimate before you have gathered any personal data.
Where the activity multiplier goes wrong
This is the largest source of error in the whole calculation, and it is systematic rather than random: almost everyone picks a level too high.
The multipliers describe your entire day, not just your training. Someone who lifts hard four times a week and then sits at a desk and on a sofa for the remaining twenty-two hours is not "very active" — they are closer to lightly active with some training on top. Occupation dominates: a nurse or a builder who never trains often has a higher TDEE than a desk worker who trains daily.
Pick one level lower than your instinct says. Eat at that figure for two to three weeks, weighing yourself several times a week and taking the weekly average. If the average holds steady, that was your true maintenance — and it is worth far more than any formula. If it moves, adjust by 100–150 calories and repeat. Two weeks of your own data beats every equation on this page.
What the research actually says
Every default in this calculator traces to published evidence rather than convention. The specifics, and where they come from:
Mifflin-St Jeor as the default equation
A systematic review in the Journal of the American Dietetic Association compared the four equations most used clinically and found Mifflin-St Jeor predicted resting metabolic rate within 10% of measured values more often than any other, with the narrowest error range. A later validation in Clinical Nutrition measured 337 adults by indirect calorimetry and found it the only unbiased equation tested — every other one systematically overestimated, except Livingston which underestimated.
Two caveats that matter. Accuracy falls in obesity: the same study found roughly 87% of non-obese estimates landed within 10%, against 75% in obese participants. And a 2024 study of people living with overweight found accuracy around 73% even for the best-performing equations. The 10% uncertainty this calculator quotes is not conservatism — it is what the measurement literature shows.
Why Harris-Benedict runs high
A 2024 comparison against indirect calorimetry in overweight and obese adults measured mean BMR at 1,581 kcal, against 1,690 predicted by Mifflin-St Jeor and 1,788 by Harris-Benedict. Only 36.8% of Harris-Benedict estimates fell within 10% of measured, against 50.4% for Mifflin-St Jeor. The calculator shows Harris-Benedict for comparison, not as a recommendation.
The 0.5–1% weekly rate
This is not a round number picked for convenience. Garthe et al. (2011) randomised elite athletes to lose either 0.7% or 1.4% of body weight per week with identical resistance training. The slower group gained 2.1% lean body mass and improved on every strength test; the faster group's lean mass did not change and strength gains largely failed to appear. Both lost similar total weight — but the slow group's loss was 31% fat against 21% in the fast group.
A 2021 meta-analysis by Murphy and Koehler found lean mass losses rose as daily deficits grew, with deficits under roughly 500 kcal allowing some participants to gain a little muscle while losing fat. Above that, lean mass loss began and scaled with deficit size. A 2021 narrative review in Nutrients on fat loss in resistance-trained athletes reaches the same conclusion: set intake for 0.5–1.0% body weight loss weekly to maximise fat-free mass retention.
Protein at 1.8 g/kg
A 2024 systematic review and meta-analysis of 47 studies (n = 3,218) found increased protein significantly prevents muscle mass decline during weight loss, and that intake below 1.0 g/kg carries higher risk of muscle loss while above 1.3 g/kg is associated with muscle gain. The International Society of Sports Nutrition recommends 1.2–2.0 g/kg for active individuals; reviews of physique athletes in a deficit recommend 1.8–2.7 g/kg.
The calculator's 1.8 g/kg sits at the lower end of the athletic range and comfortably above the general threshold — a defensible default for someone who trains. If you are dieting hard, are already lean, or are older, higher is reasonable. If you are sedentary and have substantial fat to lose, 1.2–1.6 g/kg is likely sufficient.
A finding worth knowing
A 2026 meta-analysis in Diabetes, Obesity and Metabolism examined lean mass loss across weight-loss methods. Lean tissue accounted for 25–39% of total weight lost on GLP-1 medications, and a statistically indistinguishable 26.2% through lifestyle intervention alone. Adding resistance training brought that down to 17.5% — the single most favourable profile of any approach studied.
The implication is direct: how you lose weight determines what you lose. Calorie targets set the rate; protein and resistance training determine the composition of what comes off. A calculator can only give you the first.
Mifflin et al., Am J Clin Nutr 1990 · Frankenfield et al., J Am Diet Assoc 2005 · Frankenfield, Clin Nutr 2013 · Garthe et al., Int J Sport Nutr Exerc Metab 2011 · Murphy & Koehler meta-analysis 2021 · Ruiz-Castellano et al., Nutrients 2021 · Clinical Nutrition ESPEN systematic review 2024 · Diabetes Obes Metab 2026. Cited for the specific figures above; consult the originals for full methodology.
Common mistakes
Frequently asked questions
How accurate is this?
Mifflin-St Jeor predicts measured resting metabolic rate within about 10% for most people, which is the best any general equation manages. That still means a few hundred calories of uncertainty for an individual. The activity multiplier adds more, since it is a coarse five-point scale describing something continuous. Treat the output as a starting point and refine it against two or three weeks of your own weight data.
Why is my target higher than other calculators give?
Because this uses a percentage deficit with a safety floor rather than a fixed subtraction. A 1,000-calorie deficit sounds decisive and is genuinely unsafe for a smaller person — it can leave them under 800 calories a day. Percentage deficits scale to the individual, and the floor prevents a target that makes adequate nutrition impossible. If a faster rate is appropriate for you, the 25% option is there.
How fast should I lose weight?
Around 0.5 to 1% of body weight a week suits most people. For someone at 80 kg that is 0.4 to 0.8 kg weekly. Faster is possible with more starting body fat, but the cost rises: more muscle lost alongside the fat, worse training quality, and a substantially higher chance of abandoning it. The rate that gets sustained beats the rate that looks best on paper.
Should I eat back the calories my watch says I burned?
No — the activity multiplier already includes your regular training. Adding a tracker's estimate counts it twice. Fitness trackers also overestimate energy expenditure considerably; validation studies find errors from 20% to 90% depending on activity and device. If your training load changes substantially, move up an activity level instead.
What macros should I aim for?
Protein matters most, particularly in a deficit — 1.6 to 2.2 g per kilogram of body weight preserves muscle while losing fat, and the calculator uses 1.8 g/kg as a sensible middle. Fat should stay above roughly 0.6 g/kg for hormonal function. Carbohydrate fills the remainder and fuels training. Beyond those floors the split is largely preference, and adherence matters more than optimisation.
Why has my weight loss stalled?
Three usual causes, in order of likelihood. Your TDEE has fallen because you are lighter — recalculate every 4–5 kg. Portions have drifted upward, which happens gradually and invisibly without weighing food. Or it is water retention masking real fat loss, which is common and resolves on its own. Wait three weeks before concluding a genuine plateau; two weeks of flat weight is normal variation.
Which formula does this use by default?
Mifflin-St Jeor, which validation research consistently finds the most accurate general equation. All three are shown so you can see the spread — when they disagree by more than 150 calories, that gap tells you how much uncertainty sits in the estimate. If you know your body fat percentage and carry above-average muscle, Katch-McArdle will likely suit you better.
Is my data stored?
No. Everything runs in your own browser — nothing is transmitted, logged or saved, and closing the tab discards it. Weight, age and body composition are among the more personal things anyone enters into a website, and none of it leaves your device.
Related calculators
See the full list of Health calculators, or try:
- TDEE Calculator — maintenance expenditure on its own
- BMR Calculator — resting rate before activity
- Macro Calculator — a detailed protein, carb and fat split
- Protein Intake Calculator — target by body weight and training
- Body Fat Calculator — for the Katch-McArdle equation
- FFMI Calculator — muscle relative to height