Is Your Calorie Deficit Too Small?

Prescribed 25 percent, delivered 12#

Your deficit is almost certainly smaller than the one you set. That is the normal case, not a personal failure, and on its own it is not what has stalled you. The trap is that a small deficit and a dead deficit look identical for weeks at a time — so a diet that is merely slow gets diagnosed as broken, and the reflex is to cut again from a number you never verified in the first place.

The cleanest measurement of that gap comes from CALERIE 2, which put 143 non-obese adults on a prescribed 25% calorie restriction for two years and checked what they actually ate with doubly labeled water instead of their own logs. Achieved restriction was 19.5% at month 6, 10.8% at month 12, 8.7% at month 18 and 8.0% at month 24 — a mean of 11.7% across the trial¹. These were supervised, motivated volunteers with dietitian contact and biomarker feedback, and they delivered less than half the deficit they were assigned. They also lost 9.9% of their bodyweight by month 6 and were still 10.4% down at two years. Both facts belong in the same sentence: the deficit was much smaller than planned, and it worked anyway.

So before you cut, work out which of the two questions you are actually asking. Was my deficit too small to do anything? is rarely yes. Is my deficit now too small to see, or gone entirely? is usually the real question, and it has an answer you can check this week.

The deficit leaks in three places, and none of them is your metabolism#

A plateau is a closed deficit, and this cluster's pillar on why plateaus happen covers the physiology. What is worth separating out is where the arithmetic goes, because the three leaks are different sizes and only one of them responds to eating less.

Leak	What it does	Rough size
Intake logged below intake eaten	Shifts the whole ledger down	12–14% (men) and 16–20% (women) on 24-hour recalls vs doubly labeled water (Subar et al., 2003)
Maintenance falling with bodyweight	Moves the target under a fixed intake	Grows with every kilogram lost
The plan measured against an old body	Makes the same restriction a weaker one	CALERIE's %CR is calculated against baseline energy requirements

That third row is the underrated one. In CALERIE, an 8% restriction at month 24 is 8% of the energy needs of a body that had already shed 7.5 kg — our reading of how the trial defines the number, not a finding the paper states. Applied to your own diet, it means a target set in January is not a smaller-and-smaller deficit because your metabolism broke; it is a smaller deficit because the person it was calculated for no longer exists. That is a scheduled event, and you can work out roughly when a quarter of it is gone.

The first row is the largest and the most commonly misread. Chronic under-recording is not lying and it is not a character flaw — the research on how it happens points at memory and portion drift more than motive. But it does mean your logged deficit is systematically wider than your real one, in a known direction. If you calculated a 400 kcal deficit and your log runs 15% low on a 2,200-calorie day, roughly 330 of those calories were never real.

A deficit narrower than the noise in your own log#

Here is the part almost nobody prices in. When nutrition researchers want to flag implausible food records, they need to know how much a normal person's intake bounces around by itself. The working figure is a within-subject variation in energy intake of 23%, alongside 8.5% variation between an estimated and a measured basal rate and 15% between-person variation in physical activity level³.

Set those next to the deficit you chose. A 15–20% cut — the band the lean-mass evidence supports — is smaller than the day-to-day scatter in what you eat. It does not disappear into that scatter, because scatter averages out and a deficit does not, but it means no single day and no single three-day snapshot can tell you whether you are in a deficit. If the days were statistically independent, seven of them would shrink 23% scatter to about 9% (our arithmetic; real days are correlated, so treat 9% as the optimistic floor). Two weeks of averaged intake, against a maintenance number carrying its own 10%-plus error, is the shortest honest look you get.

A calculated deficit is a hypothesis about two estimates. The weight trend is the only instrument that reads the answer directly.

This is also why the popular move — cut another 200 — is worse than it feels. You are subtracting a precise-looking number from a quantity you have not measured, which leaves you with a deficit you know even less about, plus the muscle and appetite costs that come with cutting harder.

Three checks that cost nothing, before you cut anything#

1. Run the plausibility ratio on your log. Divide your average reported intake by your estimated basal metabolic rate. Sustained values below about 1.35 mark a record that cannot represent habitual intake for anyone who is not losing weight rapidly — the classic cut-off derived by Goldberg et al., 1991 for screening under-recorders. Worked example: a 1,600-calorie log against an estimated 1,450 kcal basal rate gives 1.10. That number does not describe a slow metabolism; it describes a log that is missing food, because 1.10 × BMR is close to bed rest.

2. Back-calculate the deficit instead of predicting it. Two to four weeks of consistent logging plus your weekly weight averages give you a deficit denominated in your own tracking, which is the only version that is any use — the maintenance-calorie back-calculation is the same procedure run in reverse.

3. Check the multiplier, not just the calories. Most calculated deficits die at the activity step rather than the food step, because two adjacent dropdown options can differ by hundreds of calories a day. Where real adults actually land on activity level is a narrower band than the dropdowns suggest, and picking one adjective too high is the single easiest way to invent a deficit that never existed.

Only after those three does "eat less" become a rational move — and even then, moving more usually beats cutting further, because it widens the gap without shrinking the food.

Small is not the same as broken#

There is a version of the answer that runs the other way, and the evidence supports it. Population-scale modelling puts the energy imbalance behind ordinary adult weight gain at a few tens of calories a day, and estimates that closing a gap of about 100 kcal/day would prevent weight gain in most people⁵. One hundred calories a day is roughly a hundred grams of fat a week. On a bathroom scale, against day-to-day fluid swings, that is invisible for a month and obvious in a year.

So a genuinely small deficit is not a broken one. It is one you are reading with the wrong instrument on the wrong clock. CALERIE's participants spent the second year restricting by about 8% and holding a 10% loss, which is exactly what a small deficit does once it has been eaten into by a smaller body: it stops moving the scale and starts defending the position. If your measurements and your waistband are still changing while the scale is flat, what you have is not a stalled deficit at all.

The practical rule: judge a deficit by four weeks of weekly averages, not by a fortnight; audit the log before the target; and treat the calculated number as a starting bid that your own weight trend is entitled to overrule.

FAQ#

How small is too small for a calorie deficit?#

There is no floor below which a deficit stops working — there is a floor below which you cannot detect it. A deficit around 100 kcal/day is enough to change bodyweight over a year but produces roughly 100 g of fat loss a week, which disappears inside ordinary daily fluid variation. If you want visible monthly feedback, 10–20% of your maintenance is the practical range; below that, plan to judge it quarterly.

If I can't verify my deficit, should I just cut another 200 calories?#

Not first. Subtracting a precise number from an unmeasured one gives you a deficit you understand even less, at the cost of more hunger and more lean-tissue risk. Run the plausibility ratio, average two to four weeks of intake against your weight trend, and check that your activity multiplier isn't one adjective too generous. If all three are clean and four weeks of weekly averages are genuinely flat, then adjust — by 100–150 kcal, not 500.

Can a 200-calorie-a-day deficit actually do anything?#

Yes, on a longer clock than most people give it. Two hundred calories a day is on the order of 0.2 kg of fat a week, so a month of perfect adherence is under a kilogram — well inside the noise of a few weigh-ins and easily masked by a single high-carbohydrate weekend. It compounds: the same deficit held for a year is a different-looking person. The failure mode isn't the size, it's abandoning it at week five for lack of evidence.

Sources#

This article was researched and drafted with AI assistance and reviewed for accuracy by the BurnWeek team. It is general information, not medical advice. How we research and correct our articles →