Step one: your BMR with Mifflin-St Jeor
Mifflin-St Jeor estimates your basal metabolic rate, which is what you burn keeping the lights on: heart, brain, kidneys, body temperature. M.D. Mifflin and S.T. St Jeor published it in 1990 in the American Journal of Clinical Nutrition, and a 2005 review by Frankenfield and colleagues in the Journal of the American Dietetic Association found it came within 10 percent of measured resting metabolic rate more often than the older Harris-Benedict equation. That's the reason I used it rather than Harris-Benedict.
- Men: BMR = (10 x weight in kg) + (6.25 x height in cm) - (5 x age) + 5
- Women: BMR = (10 x weight in kg) + (6.25 x height in cm) - (5 x age) - 161
If you think in pounds and inches, divide pounds by 2.205 for kilograms and multiply inches by 2.54 for centimeters. The 200 lb man above is 90.7 kg, and 5'10" is 178 cm, so his BMR works out to 907 + 1,112.5 - 170 + 5 = 1,855 calories before he gets out of bed.
The equation never asks about body fat, so it drifts at the extremes, since muscle burns more at rest than fat does. Katch-McArdle uses lean body mass instead and fits better if you have a trustworthy body fat number, which most people don't.
Step two: the activity multiplier
BMR times an activity factor gives TDEE. The factors are coarse buckets, and the gaps between them are large.
| Activity level | Multiplier | What it usually means |
|---|---|---|
| Sedentary | 1.2 | Desk job, no structured training, under 5,000 steps |
| Lightly active | 1.375 | One to three training sessions a week |
| Moderately active | 1.55 | Three to five sessions a week |
| Very active | 1.725 | Six to seven sessions, or hard training most days |
| Extra active | 1.9 | Physical job on top of daily training |
Pick the lower one when you're torn. Moving that 1,855 BMR from 1.375 to 1.55 adds about 325 calories a day, which is a bagel with cream cheese, and most people overrate how active they are once they subtract the hours spent sitting. Three lifting sessions and a couple of walks is lightly active.
Step three: subtract a deficit
The default is 500 calories a day off your TDEE. That number traces back to Max Wishnofsky's 1958 calculation that a pound of body fat stores roughly 3,500 calories, so 500 a day gets you a pound a week on paper. In practice it's a straight line drawn through a curve. Kevin Hall's modeling work at the NIH shows the same deficit produces less loss as you get lighter, because a smaller body burns less, so a flat 500 slows down over months.
A more workable way to set it is percentage based: 15 to 25 percent below TDEE. For our example that's 1,910 to 2,170 calories, which brackets the 2,050 we landed on. Larger bodies can absorb a bigger absolute deficit than smaller ones.
There's a floor, and it matters. The 2013 AHA/ACC/TOS obesity guideline describes low-calorie diets as roughly 1,200 to 1,500 calories a day for women and 1,500 to 1,800 for men, and anything under that belongs in a clinician's hands. Very low intakes make protein and micronutrient targets hard to hit. For reference on what trial protocols use, the STEP 1 trial of semaglutide put participants on a diet with roughly a 500 calorie daily deficit plus 150 minutes a week of physical activity.
Step four: turning calories into grams
The conversion is fixed arithmetic. Protein 4 calories per gram, carbohydrate 4, fat 9, and alcohol 7 if you drink, which is why a couple of beers eats a chunk of the budget without filling you up.
Protein first. The range I use across this site is 1.2 to 1.6 g per kg of body weight, the band Leidy and colleagues describe for protein during weight loss in their 2015 review in the American Journal of Clinical Nutrition. If you train, the International Society of Sports Nutrition's position stand goes higher, 1.4 to 2.0 g/kg, and reviews from Stuart Phillips and colleagues argue for the upper end or above while in a deficit, since protein is what protects lean mass when calories are short. The man in the example lifts, so his 160 g against 90.7 kg is about 1.8 g/kg, inside the ISSN range. If you're carrying a lot of fat mass, use your goal weight so the number doesn't get absurd.
Fat second. The Institute of Medicine's Acceptable Macronutrient Distribution Range is 20 to 35 percent of calories from fat. Dropping far below that makes food joyless and can affect hormone production.
Carbs last, filling the remainder. Carbs are the lever you pull when you want more energy for training or more room at dinner.
| Macro | Grams | Calories per gram | Calories |
|---|---|---|---|
| Protein | 160 g | 4 | 640 |
| Fat | 60 g | 9 | 540 |
| Carbs | 218 g | 4 | 872 |
| Total | 2,052 |
That total misses 2,050 by two calories, which is noise. Round to something you can remember: 160 protein, 215 carbs, 60 fat.
Where these numbers are wrong, and what to do about it
Mifflin-St Jeor being within 10 percent is good for an equation and vague for a person. Ten percent of 1,855 is 185 calories, and that error rides all the way through the multiplier into your final target. The activity factor is a bigger source of slop than the BMR is.
Then there's logging. The classic demonstration is Lichtman and colleagues in the New England Journal of Medicine in 1992, who used doubly labeled water on people convinced they couldn't lose weight on low intakes and found their self-reported intake was far under what they were eating. Oil in the pan, the handful from the bag, the pour of cream closer to 60 mL than 30. Weighing food for two weeks recalibrates your eye more than any calculator will.
So treat the target as version one. Log consistently for two to three weeks, weigh yourself at the same time under the same conditions, and read the seven-day average rather than any single morning, since water, sodium, carbs, and menstrual cycle move the scale by pounds in a day. If the trend hasn't moved after two or three weeks of honest logging, take another 100 to 150 calories off or add activity, then hold it and watch again. Don't react to three days.
If you are taking a GLP-1
Follow your prescriber's guidance. That comes first for a reason: appetite suppression changes the problem. Plenty of people on semaglutide or tirzepatide land well under their calculated target without trying, and the risk shifts from eating too much to under-eating protein and micronutrients while losing weight quickly. A calorie target is still worth having, but on these medications it often works as a floor to reach rather than a ceiling to stay under, and that judgment belongs to the clinician who prescribed it.
What a tracker can do is give your prescriber real data at the next appointment: what you ate, how much protein you hit, and what the weight trend looked like alongside your dose dates. Dose amounts, timing, titration, and injection technique come from your prescriber and the manufacturer's instructions.