Tracking Calories but Not Losing Weight? Where the Energy Is
If you are logging every day and the scale has not moved in a month, the energy is almost always in one of three places: food that was eaten but not recorded, a deficit that was smaller than you thought because your maintenance figure came from a formula, or a real loss hidden by water and too short a measuring window. None of those mean your body has stopped responding to arithmetic.
Work through them in that order, because that is roughly the order of likelihood. And be warned that the first category is uncomfortable — not because people lie, but because the things that go unlogged genuinely do not feel like meals.
Could my portions have drifted without me noticing?
This is the most common one, and it happens to careful people specifically because they were careful once.
You weighed your porridge oats for three weeks in January. By March you know what 50 g looks like, so you pour it. Except the poured amount is now 70 g, because the scoop got generous and nothing corrected it. That is about 80 extra kcal, every morning, in an entry that still says 50 g.
The same drift shows up in rice (75 g dry is about 260 kcal, 110 g is about 385), in peanut butter (a level tablespoon is about 95 kcal, a heaped one closer to 150), in chicken breast (a "breast" ranges from roughly 130 g to 250 g), and in olive oil, which is the worst offender because it is invisible in the pan and a tablespoon is about 120 kcal.
If four or five items have each drifted by a third, you have a few hundred kilocalories a day that your log does not know about.
What about the things that aren't really meals?
Make a list of what went into your mouth yesterday that you would not describe as eating:
Milk in tea or coffee. A generous splash is perhaps 20 kcal; four cups is closer to 80.
The oil or butter the vegetables were roasted in, when the entry is just "roast vegetables".
Tasting while cooking. Two spoons of bolognese, a corner of cheese, the pasta you checked for doneness.
Children's leftovers. Half a fish finger and three chips is 100 kcal.
Drinks. A pint of lager is around 180–200 kcal, a 175 ml glass of wine around 150. Two of either on a Friday is a meal's worth.
Chewing gum, cordial, a couple of squares of chocolate from the shared desk drawer.
Individually these are trivial, which is exactly why they are omitted. Collectively they are often the entire gap.
Why do restaurant and weekend days matter more than the rest?
Because those are the days you estimate by eye, and they are also the days furthest from your routine — so the estimate has the most room to be wrong, in the direction of too low.
A plate of carbonara made in a restaurant kitchen might be 900 kcal or 1,400 depending on the cream, the portion of pasta and how much oil finished it. Logging it as "pasta, 700" is not dishonest; it is a guess made without information. The honest version of that guess is usually higher than the instinctive one.
Describing the plate in a full sentence — what was in it, roughly how much, how it was cooked — gets you closer than picking the nearest database entry called "carbonara". This is the case Day So Far was built for: you write "spaghetti carbonara, restaurant portion, pancetta and plenty of cheese" and get an estimate with the richness accounted for rather than averaged away.
Does the recipe serving match what I actually served?
A tray bake says "serves four". Four people ate it, so you logged a quarter. But two of them were children and you took the corner piece. Your quarter was a third.
The fix is boring and effective: weigh the whole finished dish, weigh your plate, log the fraction. It takes twenty seconds and removes a systematic error that repeats every time you batch-cook.
What if my deficit was never as big as I thought?
Suppose your log is genuinely accurate. The other half of the sum can still be wrong.
Maintenance figures from online calculators are population averages with a wide spread around them. Two people of the same height, weight, age and sex can differ by several hundred kilocalories a day in what they actually burn, largely through how much they move when not exercising. If your calculator said 2,400 and your real figure is 2,050, then eating 2,000 is not the deficit you planned — it is about 50 kcal a day, which is roughly 0.1 kg a week. Real, but invisible for a month.
Then there is the average. Five weekdays at 1,800 and two weekend days at 3,000 is 15,000 kcal a week, or about 2,140 a day. You feel like a person eating 1,800. Your body is doing sums on 2,140. Look at your weekly average, never the weekday you remember best.
Could I be losing fat and not seeing it?
Yes, and this is why the window matters. Body weight moves with glycogen, sodium, fibre in transit and hormonal fluid shifts. A single salty meal or a high-carbohydrate day can put a kilogram on the scale overnight, and it has nothing to do with fat. For people who menstruate, cyclical water retention can mask several weeks of genuine progress.
If you are losing about 0.25 kg a week and your daily weight scatters by ±1 kg, a fortnight of data cannot distinguish the signal from the noise. Three to four weeks of weekly averages can.
A diagnostic you can run yourself
Give this a fortnight, and change nothing about what you eat while you do it. The point is to measure, not to cut.
Weigh every solid food you eat at home, in grams, for fourteen days. No eyeballing, even for things you "know".
Log the non-meals as well: oils, milk, drinks, tastes, leftovers off other people's plates.
Apply the same care to weekends and eating out as to Tuesdays. These are the days that decide your average.
Weigh yourself daily under the same conditions and compare week one's average to week two's, not Monday to Monday.
At the end, do the sum backwards. Take your average daily intake, and the weight change across the two weeks. Using the ordinary published figure of roughly 7,700 kcal per kilogram of body fat: if you averaged 2,200 kcal and lost 0.3 kg, that is about 2,300 kcal of deficit over fourteen days — around 165 a day — so your maintenance is somewhere near 2,365, not whatever the calculator said.
That backwards sum is the only maintenance figure that is about you. It is also the approach behind Day So Far's daily target, which compares what you logged against what the scale did over a couple of weeks rather than trusting an equation. Either way, the arithmetic is the same and you can do it on paper.
What not to do
The instinctive response to a stalled month is to cut another 300 kcal. Resist it for one more fortnight. If the real problem was unlogged oil and generous weekends, cutting the logged figure changes the number on the screen and not the number in your body — and you will be hungrier for no result.
Measure first. Adjust second, and modestly. Often the adjustment that works is not eating less but moving more on the days you barely move, or shifting the weekend average down rather than the weekday floor.
All of this is estimation, including the calculators, including any app. Treat every figure here as approximate and the direction of travel over several weeks as the thing that matters.