Should you eat less or more to break a plateau?

Neither direction on the plate opens a deficit by itself#

If your weight has genuinely been flat for three or four weeks, the choice is usually presented as a fork: cut harder, or eat more to "restart" something. The arithmetic settles half of it immediately. Raising your intake cannot widen the distance between intake and expenditure — no plausible metabolic response to eating more is anywhere near large enough to spend the extra food and then some. So "eat more to lose weight" is never a claim about energy balance. It is a claim about one of two other things: that eating more will make you move more, or that eating more will make you stick with the plan.

Cutting harder is a claim about energy balance, but usually not the one people think. The deficit that stopped working almost never stopped because your body defeated it; it stopped because the two numbers you were subtracting drifted, and the one that drifts most is the one you wrote down. That makes "eat less" primarily a measurement instruction rather than a food instruction — and it has a measured price tag. This article sizes both levers so you can pick, rather than alternating between them until something moves. The staging of every lever, cheapest first, belongs to how to break a genuine plateau; this page is about the fork itself.

The "eat less" lever is mostly a measurement lever#

Before you take food away, it is worth knowing how much food is already missing from the page — and the cleanest demonstration used people who log for a living.

Ten female registered dietitians and ten comparable non-dietitian women kept 7-day weighed food records while their true energy expenditure was measured with doubly labeled water. Expenditure was similar in both groups (2,154 ± 105 vs 2,315 ± 90 kcal/day). The dietitians underreported their intake by 223 ± 116 kcal/day, a gap not statistically distinguishable from their expenditure. The controls underreported by 429 ± 142 kcal/day¹.

Ten per group is a very small study, and dietitians differ from other people in more ways than record-keeping skill. But read the two numbers as a gap and something useful falls out: the difference between an expert logger and an ordinary one was about 206 kcal a day — my subtraction of their two figures, not a comparison the paper draws. That is roughly the size of the calorie cut most people reach for at a plateau. You can plausibly find as much deficit by logging more carefully as by eating less, and finding it costs you no food at all. That is why a weighed, no-estimates week comes before any change to the target.

It also explains why the second-most-common mistake is worse than the first. Cutting a target you cannot measure means the real intake was never where you thought, so you have moved an unknown number by a known amount and learned nothing.

The "eat more" lever has one real mechanism, and it arrives bundled with movement#

Now the interesting half, because there is a genuine physiological idea here and it is routinely misquoted. It is called energy flux: not how big your deficit is, but how large both sides of the ledger are. Eating 1,800 and burning 2,000 is the same 200-calorie deficit as eating 3,000 and burning 3,200 — but the two states may not regulate appetite equally well.

A review by Melby and colleagues assembles the case, defining high flux by metabolic scope — total daily expenditure divided by resting expenditure — with 1.7–1.8 as the target and around 1.3 as the sedentary low-flux state. Its two anchor experiments are short randomized crossovers reported within the review: weight-reduced participants held at energy balance reported significantly lower hunger and greater fullness in a four-day high-flux condition than in a low-flux one, and a 16-person study found better acute appetite control at a metabolic scope of 1.8 — lower ghrelin, higher GLP-1 — while the low-flux condition produced a 17.5% positive energy imbalance when people ate freely².

The prospective data point the same way. Two independent cohorts — 154 adolescents followed 3 years and 75 college-aged women followed 2 — had intake and expenditure measured objectively by doubly labeled water at baseline. Low energy flux, not energy surfeit, predicted future increases in body fat in both samples, and high flux appeared to protect partly by way of a higher resting metabolic rate³.

Three things belong on that before you spend it. The Hume paper drew published objections in its own journal, including one arguing the study had misdefined flux⁴ — a dispute about the exposure variable, not a footnote. The review's own hedge is unusually blunt: "most of these concepts have not been rigorously tested," and its crossover evidence runs to days rather than months. And you should know who is making the argument: the review declares no external funding, yet its senior author discloses stock options in one weight-management company, equity in a second, and a patent on the "Energy Gap" — the very concept the recommendation is built around.

Every version of "eat more" with evidence behind it is really "burn more, and eat to match." Nobody has shown the food half working alone.

That is the sentence to carry away. The flux case never says raise intake at a fixed activity level; it says raise expenditure and let intake follow, so that energy balance is achieved at a higher turnover. Practically, that means the lever is daily movement and the extra food is the permission slip, not the intervention. Whether expenditure keeps scaling indefinitely with activity is a separate and contested question — the review takes an explicit side against the constrained-expenditure model, which is not a settled matter.

What eating more cannot do#

Two claims travel with this advice and neither survives contact with the numbers.

The first is that under-eating has switched your body into a hoarding state, so eating more will release the brake. Metabolic adaptation is real and it is measured in tens to low hundreds of calories a day — enough to slow a diet, never enough to stop one, and never enough to make a genuine deficit stop emptying fat stores. The full dismantling is the starvation-mode myth, and if that were the mechanism, the covertly-dieted patients whose weight loss stalled without any diet to lapse from would have needed a different explanation.

The second is reverse dieting: that stepping calories up slowly, at a fixed activity level, raises your metabolic rate enough to let you eat more at the same weight. Notice what the most sympathetic serious treatment of post-diet eating — the flux review above — actually rests on. Its entire case is that people who have lost weight need to raise physical activity to re-establish balance at a higher intake. It does not argue that intake alone moves the expenditure side, because the response to overfeeding is far too small to matter. Raising calories after a diet is a sound thing to do carefully, for a reason that has nothing to do with metabolism: it is how you find your new maintenance without overshooting it.

Picking the lever from the symptom#

The fork resolves once you stop asking "more or less" and start asking what your last four weeks actually looked like.

Read the first row as the default, because it is the cheapest, the most likely, and the only one that costs nothing if you are wrong. And note what is missing from the table entirely: cutting harder. Steepening a deficit charges you in muscle and in hunger, and it charges most where the stall was never metabolic to begin with — which, on the diagnostics above, is nearly everywhere. It belongs at the end of the queue, and most people never reach it.

FAQ#

Can eating more calories actually make you lose weight?#

Not directly — no metabolic response to extra food is large enough to spend the food and open a deficit as well. It can help indirectly in two ways. Raising physical activity and eating to match puts you in a higher-turnover state that appears to regulate appetite better: low energy flux, not energy surplus, predicted future body-fat gain across two cohorts followed with doubly labeled water³. And a planned period at maintenance can make a long diet survivable. Neither is the food doing the work by itself.

Is reverse dieting the answer to a stall?#

No, and it is aimed at a different problem. Gradually raising intake is a sensible way to arrive at your post-diet maintenance without overshooting it, but the claim that it lifts your metabolic rate enough to let you eat more at the same weight has no trial behind it, and the size of the adaptation it targets is tens to low hundreds of calories a day. Even the review that argues hardest for eating more after weight loss builds its case on raising activity, not on raising food.

How much is tighter logging actually worth?#

Plausibly as much as the calorie cut you were considering. In a controlled comparison against doubly labeled water, registered dietitians underreported their own intake by 223 ± 116 kcal/day and matched non-dietitians by 429 ± 142¹. Ten women per group is a small study, but the gap between skilled and unskilled logging was around 200 calories a day — found without removing any food, and available to anyone who weighs everything for a week.

Sources#

1. Champagne CM, Bray GA, Kurtz AA, et al. Energy intake and energy expenditure: a controlled study comparing dietitians and non-dietitians. J Am Diet Assoc. 2002;102(10):1428-1432.
2. Melby CL, Paris HL, Sayer RD, Bell C, Hill JO. Increasing Energy Flux to Maintain Diet-Induced Weight Loss. Nutrients. 2019;11(10):2533.
3. Hume DJ, Yokum S, Stice E. Low energy intake plus low energy expenditure (low energy flux), not energy surfeit, predicts future body fat gain. Am J Clin Nutr. 2016;103(6):1389-1396.
4. Schoeller DA. Misdefined energy flux and increased fatness. Am J Clin Nutr. 2016;104(5):1485-1486.