Calories, deficits, and why the maths is the easy part
How a target is built from expenditure, what a realistic rate of loss looks like, and the reasons a correct calculation still fails in practice.
Last reviewed · 1,325 words
In short
- A calorie target is TDEE minus a deficit, not BMR minus a deficit. Taking it from BMR produces a target roughly a third too low.
- A 500-calorie daily deficit loses about half a kilogram a week. Faster than that costs muscle and rarely lasts.
- The 7,700 calories per kilogram rule holds early and overstates later, because expenditure falls as you get lighter.
- Protein at 1.6 to 2.2 g per kg preserves muscle in a deficit and is the single most useful thing to get right.
- Under-reporting intake by 30% to 50% is the norm rather than the exception, which is why weighing food for two weeks changes more than any formula.
A calorie is a unit of energy: the amount required to raise one kilogram of water by one degree Celsius. Food energy, body fat and physical work are all measured in the same unit, which is what makes the arithmetic possible at all.
Body weight follows energy balance. Eat more than you spend and the surplus is stored; eat less and stored energy is used. This is thermodynamics rather than a diet theory, and no eating pattern circumvents it — what patterns change is how easy the deficit is to maintain.
Building the target
Three steps, in order.
Basal metabolic rate, from Mifflin-St Jeor or Katch-McArdle if you know your body fat.
Total daily energy expenditure, BMR multiplied by an activity factor between 1.2 and 1.9.
BMR = 10 × weight(kg) + 6.25 × height(cm) − 5 × age + 5 for men, − 161 for women.
TDEE = BMR × activity factor.
daily target = TDEE + (weekly change(kg) × 7700) ÷ 7 — a deficit when the weekly change is negative.
The target, TDEE adjusted for your goal.
| Goal | Adjustment | Expected weekly change |
|---|---|---|
| Aggressive fat loss | −25% | 0.7–1.0 kg |
| Standard fat loss | −20% | 0.5–0.7 kg |
| Gentle fat loss | −10% | 0.25 kg |
| Maintenance | 0% | none |
| Lean gain | +10% | 0.25 kg |
| Bulk | +20% | 0.5 kg |
Take the deficit from TDEE, never from BMR. This is the most common structural error. Someone with a BMR of 1,500 and a TDEE of 2,300 who subtracts 20% from BMR arrives at 1,200 calories — a 1,100-calorie deficit, not a 460-calorie one. It works briefly, then produces muscle loss, exhaustion and a rebound.
What a kilogram of fat costs
Body fat stores about 7,700 calories per kilogram, so a 1,100-calorie weekly deficit is roughly 0.14 kg and a 3,850-calorie one is half a kilogram.
The rule holds well for the first weeks and drifts afterwards, for three reasons that all point the same way.
Expenditure falls with weight. A lighter body costs less to run and moves at a lower energy cost.
Adaptive thermogenesis. Sustained restriction reduces expenditure beyond what weight loss alone explains, by roughly 10% to 15% in aggressive dieting.
Adherence decays. Weeks one and two are followed accurately. Week nine usually is not.
Expect the arithmetic to be optimistic over months, and set the plan against a realistic rate rather than the fastest one that appears in a spreadsheet.
Protein is the part that matters most
Calories determine whether weight changes. Protein largely determines whether the change is fat or muscle.
| Situation | Protein per kg of body weight |
|---|---|
| Sedentary minimum | 0.8 g |
| Generally active | 1.2 g |
| Building muscle | 1.8 g |
| Losing fat, preserving muscle | 2.2 g |
In a deficit the body will take energy from wherever it can, and muscle is available. Adequate protein plus resistance training is what directs the loss towards fat instead. Someone at 75 kg cutting should be eating roughly 165 g of protein a day.
Protein also has the highest thermic effect of the three macronutrients — 20% to 30% of its calories are spent digesting it — and it is the most satiating per calorie, which makes a deficit easier to hold. For Indian vegetarian diets this usually requires deliberate planning: dal, paneer, curd, soya and whey are the practical levers, since a typical roti-sabzi-dal pattern lands well short.
Fat should not fall below about 0.6 g per kg, since hormone production depends on it. Carbohydrate takes whatever remains, and it is the macronutrient with the least evidence for a specific target.
Why a correct calculation still fails
Intake is under-reported. Validation studies consistently find 30% to 50% under-reporting, and the people doing it are not lying — they are estimating. Oil, ghee, drinks, tasting while cooking and the biscuit with tea are the reliably forgotten items. Two tablespoons of cooking oil is 240 calories, which is most of a modest deficit.
Exercise is over-credited. Fitness trackers overstate calorie burn by 30% to 90%. Eating back what a watch credits you with is the most efficient way to eliminate a deficit you correctly calculated.
Weight is not fat. Daily swings of one to two kilograms come from water, sodium, glycogen and digestive contents. A single reading tells you nothing; a seven-day average tells you what is happening. Starting resistance training often adds water to muscle, which masks real fat loss for two to three weeks and causes people to abandon a plan that is working.
Restriction is too aggressive to hold. A 25% deficit that lasts four weeks loses less than a 15% deficit held for four months. The best plan is the one that is still being followed in month six.
Adjusting
Set the target, hold it for two to three weeks, then compare the trend to what was predicted.
If weight is falling faster than about 1% of body weight a week, the deficit is too large — add 100 to 200 calories. If nothing has moved in three weeks with accurate tracking, the TDEE estimate was too high; reduce by 10%.
Recalculate every 5 kg regardless, because the target that created the loss will not sustain it.
Where calorie counting is the wrong tool
For people with a history of disordered eating, tracking every item can do more harm than the weight change is worth. Portion-based approaches, plate composition rules, or working with a dietitian achieve the same energy balance without the daily numbers.
The calculator produces an estimate of a physiological quantity. It cannot tell you whether tracking is a good idea for you, and that question comes first.
Measuring intake without misery
Weighing everything indefinitely is neither necessary nor sustainable. What is genuinely useful is a short, accurate period that calibrates your estimates.
Weigh food for two weeks. Not forever — long enough to learn what 30 g of almonds, a tablespoon of oil and a standard roti actually are. Most people discover their portions are 20% to 40% larger than assumed, and the correction persists long after the scale is put away.
Weigh raw, not cooked. Rice roughly triples in weight when cooked and dal absorbs a variable amount of water. Raw weights are consistent; cooked weights are not.
Count the invisible calories. Cooking oil and ghee are the largest and most commonly omitted item in Indian home cooking — two tablespoons is 240 calories. Then drinks with sugar, chai with milk and sugar several times a day, and food eaten while cooking.
Keep the routine simple. Repeating five or six known meals removes almost all the estimating. Precision matters far less than consistency in what you are estimating.
Why weight loss stalls
A plateau almost always has one of four causes, in descending order of frequency.
Intake has drifted upward. Portions grow, tracking becomes approximate, untracked items reappear. This is by far the most common and the least welcome explanation.
Expenditure has fallen. A lighter body burns less, and NEAT declines unconsciously during a diet.
Water is masking fat loss. New training, higher sodium, poor sleep or the menstrual cycle can hold one to three kilograms of water and hide several weeks of real progress.
Not enough time has passed. Three weeks of a flat seven-day average is a plateau. Ten days is not.
The response, in order: verify tracking accuracy for a week, add movement rather than cutting food further, and only then reduce the target by 100 to 150 calories. Cutting calories at every stall leads somewhere that cannot be maintained.
What this calculator assumes
- BMR from Mifflin-St Jeor, or Katch-McArdle where a body fat percentage is supplied.
- An activity multiplier applied to BMR to reach TDEE, with the deficit or surplus taken from TDEE.
- 7,700 calories per kilogram of body fat, which is accurate early in a diet and optimistic later.
- Protein recommendations are per kilogram of current body weight; at high body fat, lean mass is the better basis.
- Estimates carry roughly ±200 calories of individual variation and are not medical advice.