You’re probably eating more than you think

Obese lady: I’ve been stalled for months, I just can’t lose weight! I’m already hardly eating anything. Is my metabolism broken?

Me: Obese lady, what does your diet look like? Are you tracking your food?

Obese lady: I eat a low-carb diet and don’t track. I can assure you that I’m well aware of what I’m eating though. For instance, all I had for lunch yesterday was an avocado salad.

People suck at knowing what they actually eat

Our obese lady isn’t alone in this dilemma. When researchers take obese people who claim to be eating less than 1,200 calories per day and not be losing weight, and assess how much they must actually be eating based on changes in body composition and measured daily energy expenditure (via doubly-labelled water), energy intake is underestimated by an average of 1,050 calories per day, or 47% of actual intake.1

When researchers fed these obese people a test meal and measured how much they ate, the obese people underestimated how much they ate by 20% the next day when asked about it.1

When the researchers compared these obese people to other obese people that didn’t claim to have problems with weight loss, they were more likely to attribute their obesity to genetic and metabolic factors rather than overeating.1 Of course, their total energy expenditure, resting metabolic rate, thermic effect of food, and thermic response to exercise were comparable.

Importantly, the obese people who didn’t victimize themselves still under-reported their food intake by an average of 20%, but that’s a far cry from the 47% seen in the obese people who think their metabolism is broke.

This problem isn’t exclusive to obese people, either, but it is more pronounced. In normal-weight folk, the extent of under-reporting food intake amounts to about 18% of actual intake.2 Only 8% of people over-estimated their intake, and only 11% were within 100 calories of actual intake.2 Another analysis suggested that normal-weight people underestimate their intake by an average of 10%.3

People underestimate their calorie intake by an average of 10–20%. This can be as high as 50% in people who think their metabolism is broken.

Under-reporting is food specific

Obviously there is a lot of variability in how accurately people report their food intake, and that probably stems a lot from what they are actually eating.

As illustrated below, only 17 of 41 food groups are estimated by people with relative accuracy (within 10% of actual intake).4 Overall, these people ate 3.6 kg (7.9 lbs) of food and recalled only 3 kg (6.6 lbs) of it — people forgot about an entire pound of food they ate!

It isn’t difficult to see why most people can’t lose weight. The foods they tend to forget about are some of the most calorie-dense, including flours, grains, and starches, sauces, and oils.

Consider for a moment that someone who thinks they are using just a tablespoon (15 mL) of oil will actually be using, on average, nearly twice as much (25 mL). That’s a 90 calorie difference, and we both know that people are using more than just a tablespoon.

Summing up

If you are struggling to lose weight and you don’t track what you eat, Occam’s razor suggests that you are likely eating more than you think.

Why you are overeating is another question entirely.


  1. 1.
    Lichtman S, Pisarska K, Berman E, et al. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med. 1992;327(27):1893-1898.
  2. 2.
    Mertz W, Tsui J, Judd J, et al. What are people really eating? The relation between energy intake derived from estimated diet records and intake determined to maintain body weight. Am J Clin Nutr. 1991;54(2):291-295.
  3. 3.
    de V, Zock P, Mensink R, Katan M. Underestimation of energy intake by 3-d records compared with energy intake to maintain body weight in 269 nonobese adults. Am J Clin Nutr. 1994;60(6):855-860.
  4. 4.
    Garden L, Clark H, Whybrow S, Stubbs R. Is misreporting of dietary intake by weighed food records or 24-hour recalls food specific? Eur J Clin Nutr. 2018;72(7):1026-1034.