Key takeaways
- Resting-metabolism equations such as Mifflin-St Jeor predict within 10% for most adults, but not for everyone.
- Your activity level adds more uncertainty than the equation. One level up or down changes TDEE by 10 to 15%.
- Two calculators can differ by 200 to 300 kcal for the same person because they use different equations and factors.
- Two to four weeks of food logging and weigh-ins give you a measured TDEE that beats any formula.
On this page
- How accurate TDEE calculators are, in short
- Where the error comes from
- How accurate are the BMR equations?
- The activity level is the bigger problem
- Why two calculators give different answers
- Smartwatch versus calculator
- How to find your real number in 14 days
- When to doubt a measured result
- Questions about TDEE calculator accuracy
How accurate TDEE calculators are, in short
A TDEE calculator is a good starting point and a poor final answer. It estimates your resting metabolism from your sex, age, height and weight, then multiplies it by an activity factor you choose. Both steps are built on averages from groups of people, and you are not an average.
An error of 10% sounds small, but for someone with a TDEE of 2,500 kcal it means the truth could be anywhere from about 2,250 to 2,750. Over a month, a 250 kcal daily error adds up to roughly a kilo (2 lb) of weight you expected to lose and didn’t, or didn’t expect to gain and did. That is why every estimate needs checking against real life.
Where the error comes from
| Source | Typical size | Why it happens |
|---|---|---|
| BMR equation | About ±10% for most people; more for some | Equations describe averages. Muscle, genetics and health vary. |
| Activity level | 10 to 20% | You pick it yourself, and the factors are broad bands, not measurements |
| Your inputs | Small to large | Out-of-date weight, rounded-up height, guessed body fat |
| Day-to-day variation | 100 to 300 kcal a day or more | Busy days and quiet days burn different amounts |
| Food logging, when you check | Often 10 to 30% under, sometimes more | Missed oils, drinks, snacks and guessed portions |
The last row isn’t an error in the calculator itself, but it is the main reason people decide their calculator “must be wrong”. If you eat 2,000 kcal by your log but really eat 2,400, any TDEE will look too high.
How accurate are the BMR equations?
The best evidence comes from studies that measure resting metabolism with indirect calorimetry and compare it with what the equations predict. A systematic review for the American Dietetic Association in 2005 compared the common equations in healthy adults with and without obesity. Mifflin-St Jeor predicted resting metabolism within 10% of the measured value in more people than any other equation, and its errors were narrower. That is why most calculators, including ours, use it by default.
“Within 10% for most people” also means some people fall outside that band, occasionally by 15 to 20%. The groups most likely to be missed:
- Very lean, muscular people. Weight-based equations assume average body composition and tend to underestimate them. A lean-mass equation such as Katch-McArdle or Cunningham often fits better; a study in recreational athletes found the Cunningham equation predicted their resting energy well.
- Older adults with little muscle, who may burn less than predicted.
- People from populations the equations weren’t built on. Most well-known equations came from people of European descent. Studies in some Asian and tropical populations found that standard equations can overestimate resting energy.
- People with health conditions or medicines that change metabolism, such as thyroid disease.
The activity level is the bigger problem
Even a perfect BMR estimate gets multiplied by an activity factor, and that is where most of the uncertainty lives. Each step on the usual scale (1.2, 1.375, 1.55, 1.725, 1.9) changes your TDEE by about 10 to 15%. For someone with a BMR of 1,700 kcal, “lightly active” gives 2,338 kcal and “moderately active” gives 2,635: a 300 kcal gap from one click.
People also tend to rate themselves as more active than they are. Three gym sessions a week feels active, but if the rest of the week is spent at a desk and on the sofa, the day as a whole is often closer to lightly active. The factors themselves are rough, too: measured data in the FAO/WHO/UNU energy report put real-life sedentary lifestyles at about 1.4 to 1.7 times BMR, higher than the classic 1.2. The classic set is cautious, which suits weight loss but can underestimate people who move a lot without “exercising”.
If you aren’t sure, the activity level calculator scores your job, steps and workouts for you.
Why two calculators give different answers
Put the same details into five calculators and you may get five numbers. Common reasons:
- They use different BMR equations. The revised Harris-Benedict equation usually gives 50 to 100 kcal more BMR than Mifflin-St Jeor.
- They use different activity factors or describe the levels differently.
- Some add exercise separately instead of using an activity level.
- Some use body fat when you enter it; others ignore it.
Here is one 30-year-old man, 175 cm (5′9″), 80 kg (176 lb), 18% body fat, moderately active, run through four equations:
| Equation | BMR | TDEE (×1.55) |
|---|---|---|
| Mifflin-St Jeor | 1,749 | 2,711 |
| Katch-McArdle | 1,787 | 2,770 |
| Harris-Benedict (1984) | 1,830 | 2,836 |
| Cunningham | 1,943 | 3,012 |
A 300 kcal spread before anyone argues about his activity level. None of these is “the” right answer; they are four reasonable estimates. You can compare all four for yourself in the formula comparison.
Smartwatch versus calculator
A watch seems more personal, but its calorie numbers are often less accurate than people assume. In a 2017 Stanford study, 60 volunteers wore seven popular wrist devices. Heart rate was measured well, but energy burned was not: the best device had a median error of about 27% and the worst about 93%. A 2020 meta-analysis of 60 studies reached a similar conclusion: accuracy varied a lot by device and activity.
Watches are excellent for steps and trends: whether you moved more this week than last. For calories, use them as a rough guide and let your weight trend have the final word. More in our guide to smartwatch calorie accuracy.
How to find your real number in 14 days
- Pick your estimate from the TDEE calculator. If unsure of your activity level, take the lower one.
- Eat close to a fixed amount every day for at least 14 days. It doesn’t have to be your TDEE; it just has to be steady and fully logged.
- Weigh every morning under the same conditions.
- Do the maths:
Real TDEE = average calories eaten − (weight change in kg × 7,700 ÷ days)
Example: you ate an average of 2,100 kcal for 21 days and your weekly average weight fell by 0.6 kg. That is 0.6 × 7,700 ÷ 21 = 220 kcal a day from storage, so your real TDEE is about 2,320 kcal.
The adaptive TDEE calculator does this for you, and its daily log fits a trend line so one odd weigh-in doesn’t throw the result. Three to four weeks gives a steadier answer than two, and avoid starting the window in the first week of a new diet, when water losses exaggerate fat loss.
When to doubt a measured result
- Much lower than the estimate: check your logging first, especially oils, drinks and weekends. If your log is complete, the next suspects are fewer daily steps than you think and, less often, a health problem such as an underactive thyroid.
- Much higher than the estimate: often real. Muscular people and people with active jobs or lots of walking really do burn more. Just make sure the window didn’t include a big water drop.
- Jumping around between checks: your windows are too short, or your eating changed a lot within them. Use four weeks.
Questions about TDEE calculator accuracy
Can a TDEE calculator be off by 500 calories?
For some people, yes, especially if the activity level is wrong or body composition is unusual. For most people the error is smaller, around 150 to 300 kcal. Two to four weeks of tracking will show you which group you are in.
Which TDEE formula is the most accurate?
For most adults, Mifflin-St Jeor. If you are lean or very muscular and know your body fat, Katch-McArdle or Cunningham may fit better. The activity level you choose matters more than the formula.
How long should I test my calculated TDEE?
At least two weeks, ideally three to four. Shorter tests are thrown off by water, salt and the menstrual cycle.
Is my smartwatch more accurate than a TDEE calculator?
Usually not for calories. Studies have found large errors in wrist devices’ energy estimates. Watches are good for steps and heart rate; your weight trend is the best check on calories.
Why is my real TDEE lower than the calculator says?
The most common reasons are an activity level set too high, food that isn’t making it into the log, and fewer daily steps than you think. After a long diet, your burn can also run a little below what your size predicts.
Why do two TDEE calculators give different numbers?
They use different BMR equations, different activity factors or count exercise differently. Differences of 200 to 300 kcal are common and don’t mean either one is broken.
Can any TDEE calculator be 100% accurate?
No. Even lab measurements have some error, and your burn changes from day to day. The aim is a good starting estimate, then a measured number from your own data.
Sources
- Frankenfield D, Roth-Yousey L, Compher C. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. J Am Diet Assoc. 2005;105(5):775-789. pubmed.ncbi.nlm.nih.gov/15883556/
- Mifflin MD, St Jeor ST, Hill LA, et al. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51(2):241-247. doi.org/10.1093/ajcn/51.2.241
- ten Haaf T, Weijs PJM. Resting energy expenditure prediction in recreational athletes of 18-35 years: confirmation of Cunningham equation and an improved weight-based alternative. PLoS One. 2014;9(10):e108460. doi.org/10.1371/journal.pone.0108460
- Henry CJK. Basal metabolic rate studies in humans: measurement and development of new equations. Public Health Nutr. 2005;8(7A):1133-1152. doi.org/10.1079/PHN2005801
- FAO, WHO, UNU. Human energy requirements: report of a joint FAO/WHO/UNU expert consultation. FAO Food and Nutrition Technical Report Series 1. Rome: FAO; 2004. www.fao.org/4/y5686e/y5686e00.htm
- Shcherbina A, Mattsson CM, Waggott D, et al. Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. J Pers Med. 2017;7(2):3. doi.org/10.3390/jpm7020003
- O’Driscoll R, Turicchi J, Beaulieu K, et al. How well do activity monitors estimate energy expenditure? A systematic review and meta-analysis of the validity of current technologies. Br J Sports Med. 2020;54(6):332-340. doi.org/10.1136/bjsports-2018-099643
- Lichtman SW, Pisarska K, Berman ER, et al. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med. 1992;327(27):1893-1898. doi.org/10.1056/NEJM199212313272701