Quick Answer: Your maintenance calories are the intake at which your weight holds steady, and you find yours with two weeks of your own data rather than a calculator. Eat normally, log everything honestly including weekends, weigh daily under identical conditions, and compare weekly averages. If your weight held steady, your average intake is your maintenance. If it moved, adjust by roughly 500 calories per pound of weekly change.

Every body-composition strategy — cutting, bulking, recomposition — references one number. Maintenance calories: the intake where your weight holds steady.

And most people guess it from an online calculator that can miss by several hundred calories in either direction, then build months of effort on top of that guess.

The real number comes from two weeks of your own data. Here’s the protocol, and why it’s worth the fortnight.

Why Do TDEE Calculators Miss?

They average populations. You are not a population.

Metabolic rate varies between individuals of the same height, weight, age, and sex — sometimes by 200–300 calories a day, for reasons including body composition, hormonal factors, and genetics.

Activity multipliers are crude. The gap between “lightly active” and “moderately active” in most calculators is several hundred calories, and almost nobody knows which bucket they actually belong in.

NEAT is genuinely significant and unmeasured. Non-exercise activity thermogenesis — fidgeting, standing, pacing, gesturing — varies enormously between people and can account for hundreds of calories of daily difference. Two people with identical workouts can have meaningfully different totals because one of them can’t sit still.

So treat the calculator’s output as a starting hypothesis, not an answer. Your scale-and-log data is the experiment that corrects it.

The Two-Week Protocol

Four rules. All four matter.

Eat normally. This is the hardest one. The instinct is to eat “well” during the measurement period, which measures a version of you that doesn’t exist the other fifty weeks of the year. You want your actual baseline.

Log honestly, all fourteen days. Weekends confess too. A protocol that covers only weekdays produces a number that’s wrong by exactly the amount weekends differ, which is usually a lot. Weigh food where you reasonably can — eyeballed portions run low by 20–30% in most studies of self-reporting.

Weigh daily, identical conditions. Morning, after using the bathroom, before eating or drinking, same amount of clothing. Consistency of conditions matters more than the specific conditions.

Average each week separately. Week one average, week two average. Compare those two numbers, never individual days.

Then read the result:

Weight stable between the two weekly averages — your average daily intake is your maintenance. Done.

Weight trending down — you were in a deficit. Add roughly 500 calories per pound lost per week to your average intake to estimate maintenance.

Weight trending up — same math, subtract. A pound of weekly gain suggests you were roughly 500 above maintenance.

The Daily-Weight Truce

Daily weighing only works if you agree in advance not to react to any single reading.

Single readings swing wildly on water retention, sodium intake, carbohydrate intake, digestive timing, and — for those who menstruate — cycle phase, which can shift scale weight by several pounds independent of anything else.

The two-pound overnight “gain” is never fat. Two pounds of fat is roughly 7,000 calories. You did not eat 7,000 surplus calories yesterday. It’s water and food volume, and it will be gone in a few days.

The rule: watch weekly averages, let the daily numbers be weather. Any individual day is noise. Seven days averaged is signal.

This is precisely why the protocol runs two weeks rather than one. A single week’s average can still be skewed by where you are in a cycle or a particularly salty weekend. Two weeks gives you something to compare against itself.

One note worth stating plainly: if daily weighing makes you anxious or you have any history of disordered eating, skip it. Weigh weekly, or track by how clothes fit and how you’re performing in the gym. And anyone with a medical condition affecting weight or metabolism should work this out with a doctor or registered dietitian rather than a spreadsheet.

What the Number Is For

Once you know maintenance, every strategy becomes a simple offset from it.

Fat loss: 300–500 calories below maintenance. That’s roughly 0.5–1 pound per week, which is sustainable and preserves muscle far better than aggressive deficits. Keep protein high — around 0.7–1g per pound of body weight — and keep lifting.

Recomposition: eating at or slightly under maintenance with high protein and consistent resistance training. Slow, but it works particularly well for people newer to lifting or returning after time off. The scale barely moves while body composition changes underneath it, which is exactly why you need to stop reading the scale as the only signal.

Gaining: 200–300 above maintenance. More than that adds fat faster than muscle for most people.

And the underused option: just eating at maintenance while strength climbs. This is a completely legitimate long-term strategy that the diet industry never advertises because there’s nothing to sell. You get stronger, your body composition improves gradually, your energy is good, and nothing about it is miserable.

What Nobody Tells You

Your maintenance changes. It shifts with body weight, activity level, season, and age. Re-run the protocol every six to twelve months, or whenever your circumstances change meaningfully.

Metabolic adaptation is real but modest. Extended dieting does reduce your maintenance somewhat — partly from lower body weight, partly from reduced NEAT. It’s not the metabolic damage the internet describes, but it does mean the number you found in January isn’t the number in June if you’ve been dieting throughout.

Logging accuracy is the biggest error source, not the math. Studies consistently show people underreport intake substantially. If your data says you’re eating 1,800 and not losing weight at a calculated deficit, the intake number is the thing to check first.

Two weeks is the minimum, not the ideal. Three or four weeks gives a cleaner signal if you can sustain the logging.

Mini FAQ

How do I calculate my maintenance calories accurately?

Log everything you eat for two weeks while eating normally, weigh yourself daily under identical conditions, and compare the two weekly averages. If weight held steady, your average daily intake is your maintenance. Calculators give you a starting estimate; your own data gives you the actual number.

How long does it take to find your maintenance calories?

Two weeks minimum, three to four for a cleaner result. One week isn’t enough because normal water-weight fluctuation can mask or fake a trend over that timeframe.

Can you build muscle eating at maintenance calories?

Yes, particularly if you’re new to resistance training, returning after a break, or carrying higher body fat. Progress is slower than in a surplus, but it comes without the accompanying fat gain — and it’s sustainable indefinitely, which a bulk isn’t.

TL;DR:

• Calculators average populations and can miss by hundreds. Treat their output as a hypothesis.
• Two weeks: eat normally, log honestly including weekends, weigh daily under identical conditions.
• Compare weekly averages, never single days. A two-pound overnight jump is water, not fat.
• Steady weight = that intake is your maintenance. Moving weight = adjust ~500 cal per pound per week.
• Fat loss sits 300–500 under; recomp sits at or just under with high protein; eating at maintenance while getting stronger is a legitimate long-term strategy.

Two weeks of logs, daily weights averaged weekly, calculator corrected by data. Maintenance was never a mystery number — it’s an experiment you run once and reference for years.