Why you're not losing weight in a deficit
On this page
A real calorie deficit always takes weight off. When the scale won’t move for three or four weeks, your deficit is smaller than you think. The usual gaps are food you didn’t log, workouts eaten back, weekends, a target set for a heavier body, and water hiding the loss.
What we’d do
Don’t cut harder yet. For two weeks, log every bite and every drink, weekends included, and weigh the oils, nuts and sauces once so you know what a spoonful is. Weigh yourself most mornings. Stop eating back your watch’s exercise calories. Re-work your target for the weight you are now. Then judge the trend over four weeks.
If it’s still flat after four honest weeks, take off 100 to 200 calories or add 3,000 steps a day. If it’s flat after that and you also feel cold, tired or unwell, ask your doctor about your thyroid and your medicines.
If you remember three things
- A deficit works. The question is whether you’re in one. In a famous 1992 study, people who said they couldn’t lose on 1,200 calories were eating about twice what they logged.
- Each gap is 100 to 500 calories. That’s about the size of a whole deficit.
- Judge it over four weeks. Water can hide a week or two of real loss.
Can you eat less and still not lose?
It’s one of the most searched questions about weight. You log 1,500 calories a day, you walk, you skip dessert, and a month later the scale is where it started. It feels like your body is broken, and that’s when people start looking up starvation mode and damaged metabolisms.
It’s like driving with a GPS that thinks you’re two miles from where you are. Every turn it gives you is right for the wrong place, so you keep missing the exit. Your plan is the GPS. If the numbers going into it are off by a few hundred calories, every decision after that is off too.
In 1992, researchers in New York took ten people who said they couldn’t lose weight on under 1,200 calories a day and measured everything for two weeks. Their burn was within 5% of normal for their size. They had under-reported what they ate by 47% and over-reported their exercise by 51%, while describing their eating as fairly normal. Food you don’t notice is food you don’t log.

Every gap, sized in calories
| The gap | Typical size | How to check |
|---|---|---|
| Food you didn’t log | People report 12 to 20% less than they eat, so 250 to 450 a day on an 1,800 calorie log | Log every bite for two weeks, and weigh the oils and nuts once |
| Restaurant and packaged food above its label | Lower-calorie restaurant meals averaged 18% more than stated, about 125 calories on a 700 calorie meal | Log restaurant meals higher than the menu says |
| Workouts eaten back | 100 to 300 a workout | Don’t eat back the watch’s number |
| Weekends | Two big days can cancel more than half the week’s gap | Log Saturday and Sunday as carefully as Tuesday |
| A target set for a heavier body | About 10 a day per pound you’ve lost, so 200 after 20 pounds | Re-work your target every 10 to 15 pounds |
| Your body adapting | About 40 to 90 a day in a well-run trial | Hold your weight for a few weeks and it shrinks |
| A calculator that was off for you | 10% or more for about 1 in 5 people, often 200 or more | Measure your real number over four weeks |
| Water hiding fat loss | No calories, but it can hide a week or two of loss | Judge four weeks, not one |
Food you didn’t log
Strong Under-reporting has been measured against doubly labeled water, which counts the calories you really burn, in hundreds of people.
The 1992 study was a small group who’d struggled for years. A larger one, the OPEN study of 484 adults aged 40 to 69 in Maryland, found that people under-reported their calories by 12 to 14% (men) and 16 to 20% (women) when asked what they ate yesterday, and by 31 to 38% on food questionnaires. Look first at the cream in the coffee, the oil in the pan, the bites while cooking and the kids’ leftovers.
Labels miss too. A 2010 study measured 29 reduced-calorie restaurant meals and 10 frozen meals in a lab. The restaurant food averaged 18% more calories than stated and the frozen meals 8% more. Some restaurant items held twice what they claimed, and the free side dishes raised the meal to 245% of the stated entree on average.
Workouts eaten back
Many food apps add your exercise calories to your budget for the day, so a 400 calorie workout on the watch becomes 400 calories more to eat. Two things go wrong. Watches guess calories badly: in a Stanford test of seven devices, median errors ran from 27% to 93%. And your body makes up part of any workout: in a 2021 study of 1,754 adults, only about 72% of the extra calories burned in activity showed up as extra calories burned that day. Eat back a 400 calorie workout and the 28% alone leaves you about 110 calories over plan, before any watch error. More in should you eat back exercise calories?
Weekends
In a 2008 Washington University study, 48 adults aged 50 to 60 weighed in on runs of seven straight mornings through a year-long diet or exercise program. Before the diet started they gained weight on weekends and not on weekdays, because they ate more on Saturdays and moved less on Sundays. On the diet, the calorie-cutting group stopped losing on weekends, and the exercise group gained.
Here’s how that adds up. Say your maintenance is 2,500 and you eat 2,000 Monday to Friday, a gap of 2,500 for the week. Saturday is brunch, drinks and pizza at 3,500, and Sunday a barbecue at 3,000. Your week’s gap falls from 2,500 to 1,000, about 140 a day, small enough to vanish in the scale’s daily swings.
A smaller body, and an adapting one
Each pound you lose lowers your maintenance by about 10 calories a day, by Kevin Hall’s model of body weight. Twenty pounds down, a target that started 500 under maintenance is only 300 under.
Your body also turns its burn down a little more than your new size explains. In a 2020 trial, 71 adults with obesity ate 1,000 calories a day for 8 weeks and lost about 31 pounds (14 kg). Their resting burn ended about 92 calories a day below what their smaller bodies predicted. After four weeks of holding their weight steady, the gap was about 38, and in those followed for a year it was too small to measure. The Biggest Loser contestants were the extreme: about 275 calories at the end of the show and about 500 six years later, after very fast loss on an intensive diet and exercise program.
The case that your body is the problem
Adaptation is real. Burn falls more than size explains during a diet, and in extreme cases it lasts.
Hunger climbs. A 2016 analysis estimated appetite rises by about 100 calories a day for every kilogram lost, more than three times the drop in burn.
Some conditions and medicines matter. Nearly 5 in 100 Americans aged 12 and older have an underactive thyroid, and the NIDDK lists some medicines for epilepsy, depression, diabetes, high blood pressure and allergies, and steroids, among those that can make you hungrier or burn less.
The case that it’s the math
Burn is normal in people who can’t lose. In the 1992 study it was within 5% of predicted. Intake was the gap.
Adaptation is small. About 40 to 90 calories a day in a well-run trial, and it halves once your weight holds.
A real deficit never stops working. In the Minnesota starvation experiment of 1944 to 1945, men on a diet modeled on the food available to war victims kept losing for six months, most of them about 24% of their weight.
Both are true, and the math is bigger. Of the body’s pushback, hunger is the larger part, more than three times the drop in burn in the 2016 analysis.
Water hides fat loss
Your weight can swing from day to day with no change in fat. Glycogen, the body’s carb store, is kept with three to four times its weight in water, so a high-carb weekend refills it and adds the water back. A week of real fat loss can look flat, then show up all at once. Four weeks of morning weigh-ins show the trend through it.
How to find your gap in four weeks
- Log everything for two weeks. Drinks, cooking oil, sauces, bites and weekends. Weigh the calorie-dense things once.
- Weigh in most mornings. After the bathroom, before breakfast, same scale.
- Leave the watch’s exercise calories out. Your usual workouts belong in your maintenance, not on top.
- Work out your real maintenance. After four weeks, your average intake and your weight trend give it. Measure your real TDEE shows the arithmetic.
- Then change one thing by a little. 100 to 200 calories, or 3,000 more steps a day.
In Gyroscope you can photograph each meal with Food XRAY to estimate calories, macros and micronutrients, and it connects your nutrition with your health data and AI coach. Weigh in most mornings and your weight trend gives a calorie-burn estimate. More on tracking it.
15 sources
- Lichtman et al., Discrepancy between self-reported and actual caloric intake and exercise in obese subjects, New England Journal of Medicine, 1992pubmed.ncbi.nlm.nih.gov
- Subar et al., Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study, American Journal of Epidemiology, 2003pubmed.ncbi.nlm.nih.gov
- Urban et al., The accuracy of stated energy contents of reduced-energy, commercially prepared foods, Journal of the American Dietetic Association, 2010pmc.ncbi.nlm.nih.gov
- Shcherbina et al., Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort, Journal of Personalized Medicine, 2017pmc.ncbi.nlm.nih.gov
- Careau et al., Energy compensation and adiposity in humans, Current Biology, 2021pubmed.ncbi.nlm.nih.gov
- Racette et al., Influence of weekend lifestyle patterns on body weight, Obesity, 2008pmc.ncbi.nlm.nih.gov
- Hall et al., Quantification of the effect of energy imbalance on bodyweight, The Lancet, 2011pmc.ncbi.nlm.nih.gov
- Martins et al., Metabolic adaptation is an illusion, only present when participants are in negative energy balance, American Journal of Clinical Nutrition, 2020pmc.ncbi.nlm.nih.gov
- Fothergill et al., Persistent metabolic adaptation 6 years after The Biggest Loser competition, Obesity, 2016pmc.ncbi.nlm.nih.gov
- Polidori et al., How strongly does appetite counter weight loss? Quantification of the feedback control of human energy intake, Obesity, 2016pubmed.ncbi.nlm.nih.gov
- Kreitzman, Coxon and Szaz, Glycogen storage: illusions of easy weight loss, excessive weight regain, and distortions in estimates of body composition, American Journal of Clinical Nutrition, 1992pubmed.ncbi.nlm.nih.gov
- Frankenfield, Bias and accuracy of resting metabolic rate equations in non-obese and obese adults, Clinical Nutrition, 2013pubmed.ncbi.nlm.nih.gov
- Minnesota Historical Society, MNopedia, Starvation experiment, Dr. Ancel Keys, 1944-1945mnhs.org
- NIDDK, Hypothyroidism (underactive thyroid)niddk.nih.gov
- NIDDK, Factors affecting weight and healthniddk.nih.gov