How your numbers are worked out
Every figure below is read from the code that decides it, not written down beside it. Change how a target works and this page changes in the same commit.
Your numbers
We start with your body: weight, height, age and sex go into the Mifflin-St Jeor equation, which estimates what you burn at rest. If your body fat has been properly measured (a DEXA scan or calipers), we use Katch-McArdle instead.
Then your day. Two concrete questions, how you spend most of it and how often you train, place you on one of 5 bands, from 1.4 to 2.05 times that resting figure. Two, rather than one about how active you feel: a single label invites everyone to pick the flattering one.
Then your goal. Losing sets a deficit held to 25% of what you burn; gaining sets a surplus capped at 600 calories. Protein lands between 1.6 and 2.4 g per kg of bodyweight, fat never drops below 0.6 g/kg, and on a cut carbs hold at 2 g/kg. When your pace cannot fit inside those floors we stretch the timeline rather than cut the food, and no target drops under 1,200 calories for a woman, 1,500 for a man.
Why a band, not a number
Your target is a range because a range is what it is. A figure from a questionnaire lands within 10% of what you actually burn for only a minority of people, so that is the band we show. Once we have learned from your weigh-ins it tightens to 7%, then 5%.
Edges round to the nearest 50 calories on purpose: a range printed to the last calorie reads like a measurement. Above a certain weight it carries one more line: Formulas run a little high at higher weights, so the band allows for it.
What a weigh-in does
One weigh-in never moves your target. Bodyweight swings by a kilo on water and salt alone, so we read the trend, not the reading.
3 weigh-ins on separate days inside 21 days is the rhythm we ask for.
What Pro learns
On Pro we compare what you logged against what your weight actually did over the last 21 days, and move the estimate towards your body rather than the equation.
It wants 10 genuinely logged days before it says anything, steps at most once every 7 days, and never moves your target by more than 100 calories at once, so a strange week bends the number rather than breaking it.
Fasting days
On a day you fast, protein is held at its full amount. Everything else bends around your window: about 350 calories per hour of eating, so an ordinary window changes nothing and only a short one lowers it, never below 1,200.
We plan three sittings, not two: protein stops being useful past about 0.55 g/kg in one meal, and a third keeps each under it. Where most of a week is a fast we hold any deficit to 15%.
We will not tell you that fasting lowers what you burn. The measured evidence says expenditure is unchanged; what falls is how much people move. So a lighter fasting day is yours to choose, never ours to assume.
Micronutrients
Six, each quoted against the Daily Value printed on a packet:
- Iron 18mg
- Vitamin D 20µg
- B12 2.4µg
- Zinc 11mg
- Magnesium 420mg
- Omega-3 1.6g
Where a food states its own figures, from a label, a barcode or a food database, we show them as declared. Where it states nothing we say so, and if we can estimate from a matching generic food we name it. You are always told which of the two you are reading.
Halal on a scan
Scan a barcode and we read the whole ingredient list, sub-ingredients included, name every ingredient worth checking and why, and say whether the pack is certified.
We never answer “halal” or “haram”. That is a ruling we are not qualified to give, and the additives where it matters most are the ones the packet does not source. We hand you the line and the reason; the ruling is yours.
Sources
The constant in our code, and the source recorded beside it.
ACTIVITY_MULTIPLIERS- FAO/WHO/UNU (2004). Human Energy Requirements, Table 5.3
DEFICIT_CAP_FRACTION- Murphy & Koehler (2022). Nutrients 14(21):4481
BULK_SURPLUS_CEILING_KCAL- Iraki et al. (2019). J Int Soc Sports Nutr 16:24
MIN_FAT_G_PER_KG- Aragon et al. (2017) J Int Soc Sports Nutr 14:16
MIN_DAILY_CALORIES- Ours: the long-standing clinical convention for the lowest intake
BAND_HALF_WIDTH_PCT- a published algorithm-accuracy analysis of 748 people, read 2026-09-05
BMI_BAND_NOTE_THRESHOLD- Frankenfield, JAND 2005
RHYTHM_WEIGH_INS- Ours: weigh-ins must SPAN at least ten days
MIN_LOGGED_DAYS- Ours: an honesty gate, not a statistical power calculation
FASTING_PERIOD_DEFICIT_CAP_FRACTION- Lessan et al. (2018). Am J Clin Nutr 107(1):54-61
PER_MEAL_PROTEIN_CEILING_G_PER_KG- Schoenfeld & Aragon (2018). J Int Soc Sports Nutr 15:10
COMFORT_KCAL_PER_HOUR- Ours: No literature sets a kcal-per-hour of eating window
MICRO_DAILY_VALUES- FDA Daily Values, 21 CFR 101.9(c)(8)(iv)
Something look wrong? We would rather change a number than defend one.
Tell us what we got wrong →