Should You Eat Back Your Exercise Calories?
By Mac Burke, qualified sports nutritionist
The short answer
Usually no. Fitness watch calorie estimates were off by 20% or more in lab testing, and people tend to eat more after exercise without noticing. Set one daily calorie target that already includes your normal training, then adjust it from your weekly average weight. Eat a little more only on unusually long or hard days.
For most people trying to lose fat, eating back exercise calories is a mistake. The numbers your watch or app shows are rough estimates, and adding them back to your food often wipes out the deficit you planned. A better system is one fixed daily target that already accounts for your usual training. Let your weekly weight trend tell you if it needs to change.
How accurate are fitness tracker calorie numbers?
Not very. In a Stanford study, Shcherbina and colleagues (2017) tested seven wrist-worn devices, including the Apple Watch and Fitbit Surge, on 60 adults while measuring their real energy use in the lab. Most devices measured heart rate well. None achieved an error in energy expenditure below 20%. Error was higher for walking, for men and for people with a higher BMI.
So if your watch says a session burned 500 kcal, the real number could easily be 100 kcal higher or lower, and some devices were worse than that. Eating back an estimate that large and that shaky is a gamble.
Why do people lose less weight from exercise than expected?
Because the body and the brain push back. Researchers call it compensation.
In the DREW trial, Church and colleagues (2009) gave 411 sedentary women different amounts of supervised exercise for six months. The two smaller doses produced about the weight loss predicted. The largest dose, about 194 minutes a week, produced only about half the predicted loss.
The E-MECHANIC trial by Martin and colleagues (2019) looked for the reason. Across 24 weeks of supervised exercise, people lost less weight than expected mainly because they ate more, driven by bigger appetites. Changes in their metabolism and their activity outside the sessions did not explain the gap.
Put those findings together and the risk is clear. If your appetite already nudges you to eat a bit more after training, deliberately adding back a generous watch estimate doubles the problem.
So what should you do instead?
Use one daily target that assumes your normal week of training. Here is a simple setup:
- Track your food and morning weight for two weeks while training as usual.
- If your average weight is stable, that intake is your maintenance, training included.
- Subtract enough to lose about 0.5-1% of body weight per week, the range Helms and colleagues (2014) recommend for keeping muscle.
- Eat to that number every day, training day or not.
- Review the weekly average weight every two weeks and adjust by 100-200 kcal if needed.
This approach treats your training as part of maintenance, so you never need to guess how much a session burned. The scale trend corrects any error for you.
What does this look like in a real week?
Take an 80 kg person whose tracking shows they hold their weight on about 2,500 kcal a day while training four times a week. They want to lose about 0.5% of body weight a week, which is 0.4 kg. Helms and colleagues note the rough rule that a 500 kcal daily deficit gives about 0.45 kg of fat loss a week, while stressing that the body adapts and the rule is only a starting estimate. So this person starts at about 2,050 kcal a day.
On leg day their watch claims 600 kcal burned. If they eat that back, they land at 2,650 kcal, which is above their maintenance, because the training was already counted. Do that four times a week and the planned deficit disappears. That is the trap. Their maintenance number already included those sessions, so the watch figure is counted twice.
Instead, they eat 2,050 kcal every day and check the weekly average weight after two weeks. If it is falling by about 0.4 kg a week, nothing changes. If it is flat, they drop 100-200 kcal. The scale does the correcting, not the watch.
When does it make sense to eat more on training days?
There are good reasons to eat a bit more on some days. The key is that it is planned, not driven by a watch number.
- A long endurance session, such as a two-hour ride or a long run, well beyond your usual week.
- A sport or event day with far more activity than normal.
- You are trying to build muscle and struggle to eat enough, so extra food after training helps you hit your surplus.
- Your performance is suffering and you want more carbohydrate around your hardest sessions.
If you like eating more on training days, shift food rather than add it. Move some calories from rest days to training days so your weekly total stays the same.
Do steps count as exercise calories too?
Treat steps the same way. Daily walking is part of your normal energy output, and NEAT tends to drop when you diet, according to a review by Levine (2002). Holding a steady step target helps keep your maintenance stable. It is not a bank of calories to spend at dinner.
What if you are very active or an athlete?
If your training load changes a lot from week to week, a fixed target can drift away from your needs. In that case, set two or three intake levels, such as rest, normal training and big training days, based on your own tracking. Keep protein the same and move carbohydrate up and down. If you train for a sport at a high level and struggle with energy, low mood or missed periods, see a GP or sports dietitian, because under-eating for long periods carries real health risks.
Planning food in advance also stops the after-training snack spiral. Meal plans in the MBM app, launching soon, weigh every ingredient in grams, so your intake stays on target on training days and rest days alike.
Common questions
What about apps that add exercise calories automatically?
Turn the feature off or ignore the added calories. Set your target manually based on your tracked maintenance so one number governs your intake.
Is it ever worth eating back half?
Some people use half as a compromise. It can work, but it still relies on a shaky estimate. A fixed target adjusted from your weight trend is simpler and more reliable.
Does lifting burn fewer calories than the app says?
Wrist trackers estimate weight training poorly because they lean on heart rate and arm movement. Treat any lifting calorie number as a rough guide at best.
Should I eat less on rest days?
You can, as long as the weekly total hits your target. Many people find a flat daily target easier to follow.
Sources
- Shcherbina et al. (2017), Journal of Personalized Medicine - Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure
- Church et al. (2009), PLoS One - Changes in weight, waist circumference and compensatory responses with different doses of exercise
- Martin et al. (2019), American Journal of Clinical Nutrition - The E-MECHANIC randomized controlled trial
- Levine (2002), Best Practice and Research Clinical Endocrinology and Metabolism - Non-exercise activity thermogenesis (NEAT)
- Helms et al. (2014), Journal of the International Society of Sports Nutrition - Evidence-based recommendations for natural bodybuilding contest preparation
Ready to apply this to your own program?
MBM builds 1:1 coaching programs around the exact principles above - weekly check-ins, weekly adjustments, and a plan that fits your life.