Most People Underreport Their Calories by About a Third: 7 Calorie Myths, Checked
The most common reason a calorie deficit does not work is that people eat more than they think. In UK data comparing reported intake with measured energy use, adults reported on average 32% fewer calories than they actually ate. Here are seven calorie myths and what the research shows, from forgotten snacks to the weekend.
In short
- Measured studies find most adults underreport their intake by around a third.
- Underreporting happens at every body size, not only with obesity.
- Restaurant calories, watch numbers and “small” snacks are common blind spots.
- Logging does not have to be perfect: regular, honest logging was linked to clearly more weight loss.
Before the myths, one thing to know about the studies: the strongest ones use doubly labelled water, a lab method that measures how much energy someone really burns over one to two weeks. If weight stays the same, that is also how much they ate, so it can be compared with what people report.
The 7 myths
“I know roughly what I eat.”
Most people do not. Reported intake is typically about a third too low.
The UK Office for National Statistics compared reported intake with measured energy use and found adults reported on average 32% fewer calories than they ate (ONS, 2018 (opens in a new tab)). In a 2022 study with 221 adults, people reported about 1830 kcal a day while measurements showed about 2760 kcal (Waterworth et al., 2022 (opens in a new tab)). In a small, often quoted study, people who said they could not lose weight on few calories had underreported by 47% (Lichtman et al., 1992 (opens in a new tab)).
“Only people with obesity underreport.”
No. Underreporting shows up at every body size.
In the 2022 UK study, people without obesity underreported by 31%, about 820 kcal a day. People with obesity underreported by 42%, about 1290 kcal a day, but after accounting for their larger energy needs the difference disappeared (Waterworth et al., 2022 (opens in a new tab)). It is human, not a character flaw.
“Small snacks do not really count.”
They add up faster than any meal.
A handful of nuts (30 g, about 215 kcal), a latte and a few crisps are easy to forget because they do not feel like meals. Forgetting 200 kcal a day is 1400 kcal a week. If you log every main meal but no snacks, your diary shows a deficit that does not exist. A quick photo of every bite closes the gap.
“The calories on the menu are exact.”
Roughly, on average. Not dish by dish.
In a lab test of 269 restaurant dishes, 19% had at least 100 kcal more than stated, and dishes listed as lower calorie were more often understated (Urban et al., 2011 (opens in a new tab)). More in restaurant meals have more calories than the menu says.
“I worked out, so I can eat anything.”
A workout burns less than most people think, and watches misjudge it.
A 30-minute run at about 8 km/h burns around 280 kcal on top of resting for a 70 kg person, based on the Compendium of Physical Activities (opens in a new tab). That is about one butter croissant (60 g, 259 kcal). Watch numbers are not precise either: in a Stanford test of seven wrist devices, none measured energy burned with an error under 20% (Shcherbina et al., 2017 (opens in a new tab)). Check a workout with the calories burned calculator.
“Weekends do not count.”
They count twice: more food, less movement.
In a one-year trial with 48 adults, participants on a calorie-restricted diet stopped losing weight on weekends, and before the diet they gained weight on weekends, with more food on Saturdays and less activity on Sundays (Racette et al., 2008 (opens in a new tab)). How to plan for it: the weekend effect.
“Tracking only works if it is perfect.”
Regular beats perfect.
In a 16-week app program with 449 adults, people who logged two or more times a day lost 9.1% of their weight and people who logged about once a day lost 7.7%, a difference that was not statistically significant. People who mostly logged later lost 5.7%, and people who barely logged lost 2.1% (Brockmann et al., 2026 (opens in a new tab)). These studies show a link, not proof, but the message is clear: an estimated entry is worth more than a perfect one you never make. More in how often should you log meals.
The myths at a glance
| Myth | What the research shows |
|---|---|
| I know what I eat | Reported intake is about 32% too low on average (ONS) |
| Only people with obesity underreport | Without obesity: 31% or about 820 kcal a day |
| Snacks do not count | 200 forgotten kcal a day = 1400 kcal a week |
| Menu calories are exact | 19% of dishes were 100 kcal or more above the stated value |
| Exercise lets me eat anything | 30 min running ≈ 280 kcal for 70 kg; watches err by 20% or more |
| Weekends do not count | Weight loss stalled on weekends in a one-year trial |
| Tracking must be perfect | Logging once a day did about as well as twice a day |
What to do with this
- Log right after eating, not in the evening from memory.
- Log snacks and drinks too, including the latte and the bite from the kids’ plate.
- Weigh or check portions now and then, especially oils, nuts, cheese and pasta.
- Judge by the trend. If your weight does not move for 3 weeks, look at the log first, then at the target. See how many calories should I eat.
Frequently asked questions
Why am I not losing weight in a calorie deficit?
Most often because the real intake is higher than logged: forgotten snacks, drinks, cooking oil, larger portions and weekends. Log everything for two weeks, right after eating, before changing the target.
How much do people underestimate calories?
In UK studies with measured energy use, adults reported about a third less than they ate: 32% on average in ONS data, about 900 kcal a day in a 2022 study.
Is it a problem if my log is not perfect?
No. In an app program, logging about once a day was linked to almost as much weight loss as logging twice a day. Consistency matters more than precision.
Should I eat back the calories my watch shows?
Only part of them. Wrist devices misjudge energy burned by 20% or more. Count a share and check your weight trend after a few weeks.
What is doubly labelled water?
A lab method where people drink water with harmless stable isotopes. From how fast they leave the body, researchers calculate how much energy someone burned over one to two weeks.
Sources
- ONS Data Science Campus (2018): Evaluating calorie intake for population statistical estimates (ECLIPSE)
- FoodNavigator (2018): Britons underestimate calorie intake by one third
- Waterworth et al. (2022): Obese individuals do not underreport dietary intake to a greater extent than nonobese individuals when data are allometrically-scaled. Am J Hum Biol
- Lichtman et al. (1992): Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med
- Urban et al. (2011): Accuracy of stated energy contents of restaurant foods. JAMA
- Shcherbina et al. (2017): Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. J Pers Med
- Compendium of Physical Activities: Running
- Racette et al. (2008): Influence of weekend lifestyle patterns on body weight. Obesity
- Brockmann et al. (2026): Within-day patterns of self-monitoring dietary intake and weight loss in a behavioral weight management program. Obes Sci Pract
Published · Updated
This article is general information, not medical advice. If you are pregnant, have a medical condition or an eating disorder, talk to a doctor before changing how you eat.
