Refeed Days and Diet Breaks: A Simple, Calculator-Backed Guide
Every source is linked inline, so you can check the claims yourself.
Six years after the finale of The Biggest Loser, the contestants who'd kept the most weight off were also burning the fewest calories at rest, even accounting for their smaller bodies. Kevin Hall and colleagues found their resting metabolic rate sat about 500 calories a day below what body composition alone would predict, a gap that hadn't closed since the show ended. That's the physiology refeed days and diet breaks are trying to answer for: what happens to your metabolism during a long stretch of eating less, and whether there's anything useful to do about it along the way.
This is basically the question behind a refeed day calculator or diet break calculator: not "how many calories should I eat" but "when should I eat more of them, and why." MacroMentor's calculator already flags this for you automatically. Here's what the research actually supports, and where it runs thin.
What a deficit does to your hormones
Restrict calories long enough and leptin, the hormone that signals "you have enough fat stored," falls. A CALERIE trial found six months of restriction cut 24-hour leptin levels by roughly 44 percent, and the size of that drop predicted extra metabolic slowdown beyond what the weight loss itself explained. Michael Rosenbaum and Rudolph Leibel showed the reverse works too. Give weight-reduced people enough leptin to get back to pre-diet levels, and their energy expenditure, muscle efficiency, and thyroid output drift back toward normal. That's real mechanism, backed by a controlled trial.
One caveat worth sitting with: Hall's own 2022 reanalysis of the Biggest Loser data pushed back on the doom version of this story. The contestants who kept exercising the most also showed the biggest apparent metabolic adaptation, which fits a model where the body reallocates energy spending to hold total output in a narrow range, rather than one where dieting simply breaks your metabolism for good.
The size of that slowdown isn't uniform from person to person either, and it's worth holding the exact figure loosely. Whatever number your calculator hands you at the start of a cut drifts as you go, so treat the target as a starting line, not a fixed setting.
Refeed day vs. diet break: not the same thing
The two terms get used interchangeably online, and they shouldn't be. A refeed day is short, usually a single day, sometimes two or three, with calories pushed up toward or past maintenance and the extra coming mostly from carbs. A diet break is longer: one to two weeks at full maintenance before dropping back into a deficit. That distinction matters for what the evidence can and can't tell you.
Diet breaks: decent evidence, modest payoff
The best-known trial here is MATADOR, which put 51 obese men through either 16 straight weeks of restriction or the same restriction split into two-week blocks separated by two-week maintenance breaks. The intermittent group lost more fat (12.3 kg vs. 8.0 kg) and kept more of it off six months later. Promising stuff.
It hasn't held up everywhere, though. The ICECAP trial, run in resistance-trained adults, found no difference between continuous and broken-up dieting on body composition or resting metabolism. A 2023 trial in trained women found the same null result for fat loss and RMR, though the diet-break group did report less "disinhibited" eating, a real behavioral win even without a metabolic one. A 2024 meta-analysis pooling 12 trials and 881 people settled the question about as well as it can be settled right now: diet breaks don't produce more fat loss than dieting straight through, but they do blunt the drop in resting metabolic rate a little, somewhere around 40 to 50 calories a day. Small and real, though it gets oversold online as a bigger fat-loss lever than the data supports.
Refeed days: plausible, but thinly tested
Here's the honest part. Nobody has run a trial isolating a single refeed day and measuring whether it changes fat-loss outcomes. A 2019 review of intermittent dieting in athletes put it bluntly: "no published research is currently available" on the approach in athletic populations, and called the proposed benefits "still speculative." The closest data point is a 2000 study where three days of carb overfeeding in lean, weight-stable women raised leptin 28 percent and daily energy expenditure by 7 percent. That's a different setup entirely from a refeed dropped into an active cut, so treat it as a hint, not proof.
What is solid: a day of higher carb intake restocks muscle glycogen, and that shows up directly in how much you can push the next hard session. That's the one benefit of a refeed day that doesn't need an asterisk next to it.
How often should I refeed?
No trial nails down an exact number, so this is a starting point rather than settled science: a refeed day every one to two weeks during a moderate cut, and a full one to two week diet break every six to eight weeks if the deficit is steep or the fatigue is piling up. MacroMentor's calculator flags this automatically once your target drops more than 20 percent below your maintenance calories, since that's the zone where a deficit starts asking a lot of your training, your hormones, and your patience. The 20 percent line is a practical guardrail from MacroMentor's own product logic rather than a figure issued by a health agency, though it broadly tracks the public-health caution around very-low-calorie diets that go well past that point without supervision.
None of this replaces the basic math of the deficit itself. Run your numbers at /calculator and it'll tell you your targets, along with a nudge for when it's time to back off for a day.
References
- 01Kevin Hall and colleagues found (pubmed.ncbi.nlm.nih.gov)
- 02CALERIE trial (pubmed.ncbi.nlm.nih.gov)
- 03Michael Rosenbaum and Rudolph Leibel showed (pubmed.ncbi.nlm.nih.gov)
- 04Hall's own 2022 reanalysis (pubmed.ncbi.nlm.nih.gov)
- 05MATADOR (pubmed.ncbi.nlm.nih.gov)
- 06ICECAP trial (pubmed.ncbi.nlm.nih.gov)
- 072023 trial in trained women (pubmed.ncbi.nlm.nih.gov)
- 082024 meta-analysis (pubmed.ncbi.nlm.nih.gov)
- 092019 review of intermittent dieting in athletes (pubmed.ncbi.nlm.nih.gov)
- 102000 study (pubmed.ncbi.nlm.nih.gov)
Your turn
See your own numbers.
MacroMentor runs the whole calculation in your browser, picks the formula your data supports, and shows every step. No account, nothing stored.
Start the calculator