Why weight loss stalls: the plateau, measured
A stall on the scale is not always a broken metabolism. Trials that measured calorie burn directly find a real slowdown, but modeling suggests a quieter drift in what people eat is a major contributor to the plateau.

A stall on the scale is not always a broken metabolism. A review of thirty-three studies found a real drop in calorie burn in most of them.1 But modeling built on real trial data found that small, uncounted lapses in eating can produce the common early plateau.2 Both are true, for different reasons.
What the trials actually measured
Researchers can measure how many calories a body burns at rest, and over a full day, before a diet starts and again after weight has come off. If the number after loss is lower than expected for the new, smaller body, that gap is called metabolic adaptation. It is the body spending less than its size alone would predict.
The systematic review pooled thirty-three studies that measured this directly, rather than estimating it. Adaptive thermogenesis turned up in twenty-seven of them, most often in the resting number.1 But the review's own caveat matters: well-designed studies reported smaller or non-significant values than the average, and the effect faded once weight had been held steady for a stretch rather than actively falling.1
How much the metabolism actually slows
The clearest measurements come from a 2015 rerun of the Minnesota Starvation Experiment with modern tools. Thirty-two men without obesity were overfed, then restricted to half their energy needs for three weeks, then refed. Body weight changed by 1.8 kg up, 6.0 kg down, and 3.5 kg back up across the three phases.4 Resting calorie burn fell by 266 kilocalories a day during restriction, alongside drops in heart rate, blood pressure and several hormones.4
Only part of that drop is adaptation in the strict sense. Some of it is simply a smaller body needing less fuel. In this trial, adaptation came out to 108 kilocalories a day, which was 48% of the total fall in resting burn.4 Changes in the makeup of the lean tissue explained 36 of those, leaving 72 for adaptation proper.4 The rest of the fall tracked the tissue that had been lost.
The other plateau: drift in what's eaten
The 2014 modeling study asked a different question: why do so many people plateau around six months, when the physics of a steady calorie deficit predicts a slower, later stall? The authors built two models from real trial data, one letting metabolism adapt and one letting reported eating drift from the plan. The metabolic model pushed final weight higher but did not explain the timing. The adherence model reproduced the early, six-month stall.2 Their conclusion was blunt: an intermittent lack of diet adherence, not metabolic adaptation, is a major contributor to that early plateau.2
That does not mean willpower failed. It means small departures from a plan, an extra portion here, a skipped log there, add up gradually and are hard to track, while a change in metabolism is a single number that is easy to blame.
Diet breaks and the MATADOR test
The best direct trial on breaking up a diet is MATADOR, in men with obesity. One group dieted continuously for sixteen weeks. The other alternated two weeks of dieting with two weeks back at maintenance calories, for the same sixteen weeks of active restriction spread over thirty weeks total. The intermittent group lost more weight and more fat: 14.1 kg versus 9.1 kg.3 During the maintenance blocks, weight barely moved, changing by an average of 0.0 kg.3
The absolute drop in resting calorie burn was similar between groups, but once differences in body composition were accounted for, the intermittent group's adaptation was smaller.3 A pause built into the plan, rather than one forced by a stall, appears to blunt some of the slowdown that a long, unbroken diet produces.
What happens after the plateau
Adaptation does not disappear once weight loss stops; it is one of several forces pulling weight back up afterward. A 2023 review lists resting calorie burn and adaptive thermogenesis alongside hormonal changes in hunger and reward, and shifts in fat tissue itself, as physiological factors behind weight regain.5 The review lists these side by side rather than naming one mechanism as the cause.
That framing matters for anyone reading a plateau as a verdict on effort. A stall can be the metabolism settling into a lower-burn state, a change in hunger and reward signals, or nothing more dramatic than a birthday, a stressful week, or a habit sliding a little off plan. The trials above cannot sort one person's plateau into its causes; they can only say which forces are real, and roughly how large each has been measured to be in the people who were studied.
What remains unknown
The systematic review's own studies disagreed with each other by a wide margin, and the authors flagged that better methods tend to shrink the effect, not grow it.1 That is a warning against treating any single adaptation number as fixed.
The Minnesota rerun and MATADOR both come from men. The rerun cut intake by half for only three weeks, well past what typical dieting looks like. Neither says how a smaller, everyday deficit behaves in a mixed group over a year.
And the two explanations for a plateau, a slower metabolism and a quiet drift in intake, were tested in different studies with different methods. No trial in this set measured both in the same people at the same time, so how much of any one person's stall belongs to each is still a guess.
What the evidence does support is more modest than either extreme. Metabolic adaptation is real and measurable, but it is smaller in careful studies than in the average of them, and it fades once weight is no longer actively falling.1 A plateau at six months is common, and a lapse-driven model reproduces it without any change in metabolism.2 Reading it as one or the other, purely metabolic or purely behavioral, is the mistake this set of trials keeps correcting.
QUESTIONS THIS POST ANSWERS
- Is a weight loss plateau caused by a slower metabolism?
- Partly, but the slowdown is smaller and less universal than it sounds. A systematic review of thirty-three studies covering 2,528 people found a measurable drop in calorie burn in most, but well-designed studies reported smaller or non-significant values, and the effect faded once weight was held steady for a while. Separately, mathematical modeling of real weight-loss data found that small lapses in following the diet, not metabolic adaptation, are a major contributor to the common six-month plateau.
- Does a diet break help get past a plateau?
- The best direct test found that it does, at least for the slowdown in calorie burn. In the MATADOR study, men who alternated two weeks of dieting with two weeks at maintenance calories lost more weight and fat than men who dieted continuously for the same sixteen weeks of dieting, and their drop in calorie burn was smaller once body composition was accounted for. The trial ran sixteen weeks of active dieting in men with obesity, so it does not say what a year of that pattern would do.
REFERENCES
Each reference was re-checked against PubMed before publication. How sources are checked
- 1Nunes CL, et al. Does adaptive thermogenesis occur after weight loss in adults? A systematic review. The British journal of nutrition. 2022. Source
- 2Thomas DM, et al. Effect of dietary adherence on the body weight plateau: a mathematical model incorporating intermittent compliance with energy intake prescription. The American journal of clinical nutrition. 2014. Source
- 3Byrne NM, et al. Intermittent energy restriction improves weight loss efficiency in obese men: the MATADOR study. International journal of obesity. 2018. Source
- 4Müller MJ, et al. Metabolic adaptation to caloric restriction and subsequent refeeding: the Minnesota Starvation Experiment revisited. The American journal of clinical nutrition. 2015. Source
- 5van Baak MA, et al. Obesity-induced and weight-loss-induced physiological factors affecting weight regain. Nature reviews. Endocrinology. 2023. Source