Many food logs add the calories from a workout to what you have left for the day. Whether to eat them depends on two estimates stacked on top of each other: the budget you started with, and the exercise figure added to it. Both are useful, and neither is as precise as the round number on the screen suggests.

How a calorie budget is built

Most budgets start with resting energy, the energy your body uses over a day at rest. Few people have it measured, so apps estimate it from weight, height, age and sex. A common choice is the Mifflin-St Jeor equation, derived in 1990 from measurements of 498 healthy adults aged 19 to 78 (Mifflin et al.):

  • Women: 10 × weight in kg + 6.25 × height in cm − 5 × age in years − 161
  • Men: the same, with + 5 in place of − 161

A 2005 systematic review of four widely used equations found Mifflin-St Jeor the most reliable: it came within 10% of measured resting energy in more people than the others. The review also warned of noteworthy errors for individuals, and noted that older adults and ethnic minority groups in the US were underrepresented in the research (Frankenfield et al.).

Resting energy is then multiplied by an activity factor to cover the rest of your day. The European Food Safety Authority defines this physical activity level as the energy you use over 24 hours relative to the energy you need at rest. It bases its example adult requirements on a moderately active level of 1.6 and calls these levels rough indications. They take in your work, daily exercise and household tasks (EFSA), so a higher multiplier already includes some exercise. A goal then shifts the total: a deficit to lose weight, a surplus to gain, nothing to maintain.

Imoonk's estimated setup multiplies Mifflin-St Jeor resting energy by 1.2, then subtracts 250 kcal, adds nothing, or adds 250 kcal for lose, maintain or gain. Those are product defaults to start from. They don't measure what you burn, and you can set a manual budget instead. A factor of 1.2 is low next to EFSA's 1.6, and exercise is added on top separately.

If your budget's multiplier already assumes you train most days and you also add every workout, some of that exercise is counted twice, so it's worth knowing which kind of budget you have.

Why the exercise number is shakier than it looks

Exercise estimates usually come from a table of METs or from a wearable. A MET is a multiple of resting energy. The Compendium of Physical Activities, a reference widely used in research, standardises 1 MET as 3.5 ml of oxygen per kg of body weight per minute, roughly 1 kcal per kg per hour (Herrmann et al.). It lists brisk walking at 3.5 to 3.9 mph (5.6 to 6.3 km/h) at 4.8 METs (Compendium: walking).

That 4.8 includes the 1 MET you'd have used anyway sitting on the sofa. The whole figure is gross energy, and the part above rest is net. The Compendium converts METs to kcal per minute as METs × 3.5 × weight in kg ÷ 200 (Compendium: unit conversions). For a 70 kg person walking briskly for 30 minutes, that gives about 176 kcal gross and 140 kcal net. Your budget already covers the resting part, so adding the gross figure counts about 37 kcal twice. That's small for one walk, and it repeats with every workout you log.

The MET baseline is also a convention. The 3.5 ml figure was first taken from one man, aged 40 and weighing 70 kg. When researchers measured a large, mixed group of 769 adults at rest, their average was 2.6 ml per kg per minute, and the standard value overestimated their resting oxygen use by about 35% (Byrne et al.). The 2024 Compendium says its values can be a starting point but don't reflect precise individual energy expenditure (Herrmann et al.).

Wearables estimate energy from motion sensors, often together with heart rate. A meta-analysis of 60 validation studies found that adding heart rate or heat sensing reduced the error in most types of activity. Pooled estimates came out slightly low, some devices overestimated in some activities, and the authors wrote that they would be hesitant to consider any device sufficiently accurate. Even the most accurate one overall, a research-grade armband, ranged from 21% under to 15% over across studies (O'Driscoll et al.).

A 2017 laboratory study put seven wrist-worn devices on 60 people while measuring their actual energy use as they sat, walked, ran and cycled. Most devices measured heart rate well, yet none estimated energy with an error below 20%, and median errors ranged from 27% to 93% depending on the device (Shcherbina et al.). Devices have changed since then. A 2024 review of 24 systematic reviews found that wearables tended to underestimate energy expenditure, and said that differences in how studies are run make a conclusive verdict on accuracy hard to reach (Doherty et al.). Taken together, it's fair to treat any single workout figure as possibly a fifth or more out, in either direction.

For activities it estimates, Imoonk counts only the energy above rest: (MET − 1) × 3.5 × weight in kg ÷ 200 × minutes, and nothing at 1 MET or below. Workouts imported from Apple Health (optional and read-only) use their recorded active calories when available. Remaining is your base budget plus exercise additions minus the food you've recorded, with the exercise additions shown separately.

Three ways to handle it

Each of the three has a cost.

  • Eat none back. Exercise becomes a buffer against errors elsewhere, such as an optimistic watch or a food log that runs low, as most studies comparing reported intake with measured energy use find logs do (Burrows et al.). The cost is that the gap between what you eat and what you use grows with every session, and after long or frequent workouts it can get bigger than you meant.
  • Eat part back, for example half. This hedges against an overestimate without ignoring the work you did. The fraction is arbitrary, so it's only as good as the check you run on it later.
  • Eat all of it back. This takes the numbers at face value. Wearables lean slightly low on average in pooled studies, but any single day's figure can be well off, so this approach depends most on watching the trend.

Let the trend decide

The check that works for all of these estimates is your weight trend over several weeks. From one day to the next, the scale moves for reasons that have little to do with body fat. In one study of adults living normally, about 84% of the weight change over two weeks was fat-free mass, which includes water (Bhutani et al.). In a European study of 1,421 people in a weight-maintenance programme who used smart scales, weight also followed a weekly rhythm of about 0.35%, highest after the weekend and lowest on Friday (Turicchi et al.).

So compare like with like: the average of a week's weigh-ins, taken at the same time of day, against the same average a few weeks later. If the trend is moving the way you intend, your approach is working for you, whichever it is. If it isn't, the share of exercise calories you eat back is a simple thing to adjust, and the effect is easier to read if you change one thing at a time.

Regular weighing isn't for everyone. If you have a medical condition, are pregnant, or have a history of disordered eating, a GP or registered dietitian is a better guide to your energy needs than any equation or watch.

Sources

  1. A new predictive equation for resting energy expenditure in healthy individuals — The American Journal of Clinical Nutrition, 1990.
  2. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review — Journal of the American Dietetic Association, 2005.
  3. EFSA sets average requirements for energy intake — European Food Safety Authority, 2013.
  4. 2024 Adult Compendium of Physical Activities: A third update of the energy costs of human activities — Journal of Sport and Health Science, 2024.
  5. Walking — Compendium of Physical Activities, 2024.
  6. Unit conversions — Compendium of Physical Activities, 2024.
  7. Metabolic equivalent: one size does not fit all — Journal of Applied Physiology, 2005.
  8. How well do activity monitors estimate energy expenditure? A systematic review and meta-analysis — British Journal of Sports Medicine, 2020.
  9. Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse Cohort — Journal of Personalized Medicine, 2017.
  10. Keeping Pace with Wearables: A Living Umbrella Review of Systematic Reviews Evaluating the Accuracy of Consumer Wearable Technologies in Health Measurement — Sports Medicine, 2024.
  11. Validity of Dietary Assessment Methods When Compared to the Method of Doubly Labeled Water: A Systematic Review in Adults — Frontiers in Endocrinology, 2019.
  12. Composition of two-week change in body weight under unrestricted free-living conditions — Physiological Reports, 2017.
  13. Weekly, seasonal and holiday body weight fluctuation patterns among individuals engaged in a European multi-centre behavioural weight loss maintenance intervention — PLOS ONE, 2020.