Last reviewed August 2026

Scientific Sources

The published research and public health guidelines behind every calculation Gravl Macros makes.

1. About These Sources

Gravl Macros estimates your energy expenditure, calorie targets, macronutrients and micronutrient goals using published formulas and public health guidelines. This page lists each recommendation the app makes and the sources it is based on.

All of the app's numbers are estimates, not medical advice. Individual needs vary. Read our Health Disclaimer and consult a qualified healthcare professional before making changes to your diet, especially if you have a medical condition, are pregnant or breastfeeding, or have a history of disordered eating.

Published sources explain the evidence behind a calculation. Where Gravl adds a product-specific rule, this page labels it as an app heuristic rather than implying that a study validated the exact setting.

2. Energy Expenditure

2.1. Basal Metabolic Rate

The app estimates your basal metabolic rate with the Katch-McArdle equation, which derives resting energy expenditure from lean body mass (body weight minus estimated body fat). Lean-mass-based equations account for differences in body composition that weight-and-height formulas miss.

  • Cunningham JJ (1980). A reanalysis of the factors influencing basal metabolic rate in normal adults. The American Journal of Clinical Nutrition, 33(11), 2372–2374. pubmed.ncbi.nlm.nih.gov/7435433
  • McArdle WD, Katch FI, Katch VL. Exercise Physiology: Nutrition, Energy, and Human Performance. Wolters Kluwer. The lean-mass equation popularized in this textbook (370 + 21.6 × lean body mass in kg) is the one the app uses. shop.lww.com/Exercise-Physiology/p/9781975217297

2.2. Activity and Daily Movement

Daily energy expenditure is estimated by multiplying basal metabolic rate by a physical activity factor between 1.2 and 1.9, built from your non-exercise daily movement (or your average step count when available) plus additions for exercise frequency and cardio. This range follows the physical activity levels defined by the FAO/WHO/UNU expert consultation on human energy requirements, and the treatment of non-exercise movement follows research on non-exercise activity thermogenesis (NEAT).

  • FAO/WHO/UNU (2004). Human Energy Requirements. Report of a Joint FAO/WHO/UNU Expert Consultation. fao.org/4/y5686e/y5686e00.htm
  • Levine JA (2002). Non-exercise activity thermogenesis (NEAT). Best Practice & Research Clinical Endocrinology & Metabolism, 16(4), 679–702. pubmed.ncbi.nlm.nih.gov/12468415
  • Tudor-Locke C, Bassett DR (2004). How many steps/day are enough? Preliminary pedometer indices for public health. Sports Medicine, 34(1), 1–8. pubmed.ncbi.nlm.nih.gov/14715035

2.3. Thermic Effect of Food

The app accounts for the energy your body spends digesting food, using 25% of protein calories, 8% of carbohydrate calories and 3% of fat calories — within the ranges reported in the research literature (protein 20–30%, carbohydrate 5–10%, fat 0–3%).

2.4. Adaptive Expenditure

After onboarding, the app refines expenditure from logged food and weight. It uses a 21-day intake window, fills the days between weigh-ins, smooths weight with an exponentially weighted average (alpha 0.08), and fits a 20-day trend. It converts the trend to stored-energy change with a body-composition-weighted density between 1,800 and 9,400 kcal/kg, then gradually blends the result into the prior estimate. Updates pause when recent food or weight data are missing. This is Gravl's implementation of the energy-balance principle in the cited models; the exact smoothing, confidence and noise limits are app heuristics.

  • Hall KD, et al. (2011). Quantification of the effect of energy imbalance on bodyweight. The Lancet, 378(9793), 826–837. pubmed.ncbi.nlm.nih.gov/21872751
  • Hall KD (2010). Predicting metabolic adaptation, body weight change, and energy intake in humans. American Journal of Physiology-Endocrinology and Metabolism, 298(3), E449–E466. pubmed.ncbi.nlm.nih.gov/19934407
  • Hall KD (2012). Modeling metabolic adaptations and energy regulation in humans. Annual Review of Nutrition, 32, 35–54. pubmed.ncbi.nlm.nih.gov/22540251

3. Calorie Targets and Rate of Change

3.1. Energy Density of Weight Change

To convert your chosen weekly weight-change pace into a daily calorie adjustment, the app uses the conventional value of ~7,700 kcal per kilogram of body weight change.

3.2. Safe Rates and Deficit Limits

The app caps calorie deficits at 25% of your total expenditure and surpluses at 20%, and limits the selectable pace to at most 1.0 kg per week. Gradual weight loss — commonly 0.25–1.0 kg (about 0.5–2 lb) per week — is associated with better retention of lean mass and performance than aggressive deficits.

3.3. Minimum Calorie Floors

Regardless of goal, the app will not set targets below 1,500 kcal/day for men and 1,200 kcal/day for women by default, consistent with the low-calorie diet ranges used in clinical obesity treatment guidelines. An optional lower floor (1,200/1,000 kcal) can be enabled explicitly and the app warns to monitor energy levels closely; intakes at or below these levels are best undertaken with professional supervision. The actual floor is the highest of the selected sex-based floor, its hard minimum, and 80% of estimated BMR (scaled by the lower-floor preference). The 80% BMR term is a conservative Gravl safeguard, not a separately validated clinical threshold.

  • National Heart, Lung, and Blood Institute / NIH (1998). Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. ncbi.nlm.nih.gov/books/NBK2003

3.4. Metabolic Adaptation and Refeeds

The calculator can model a 1% weekly reduction in expenditure, up to 15%, when a duration in deficit is supplied. Onboarding starts that duration at zero; after onboarding, the adaptive expenditure method above responds to the user's logged data instead of assuming a fixed reduction. For loss plans, the app also shows an optional higher-calorie planning day at 130% of estimated expenditure, with the same protein, about 15% of calories from fat, and the remainder from carbohydrate. Suggested cadence ranges from every one to four weeks based on sex and estimated body-fat range.

The cited intermittent-restriction studies support planned breaks at maintenance intake; they do not validate Gravl's exact one-day 130%, macro split, or cadence. Those settings are disclosed app heuristics, not medical treatment and not a requirement for weight loss.

  • Rosenbaum M, Leibel RL (2010). Adaptive thermogenesis in humans. International Journal of Obesity, 34(S1), S47–S55. pubmed.ncbi.nlm.nih.gov/20935667
  • Trexler ET, Smith-Ryan AE, Norton LE (2014). Metabolic adaptation to weight loss: implications for the athlete. Journal of the International Society of Sports Nutrition, 11, 7. pubmed.ncbi.nlm.nih.gov/24571926
  • Byrne NM, et al. (2018). Intermittent energy restriction improves weight loss efficiency in obese men: the MATADOR study. International Journal of Obesity, 42(2), 129–138. pubmed.ncbi.nlm.nih.gov/28925405

3.5. Weekly Target Adjustments

Weekly check-in recommendations recalculate targets from the latest adaptive expenditure estimate and the user's chosen pace. To avoid reacting fully to one noisy week, the app applies 60% of the raw calorie change and keeps the protein target from the calculator. For maintenance goals, no correction is made while trend weight is within 0.7 kg of the target; outside that band, the app applies a small 0.15% body-weight-per-week correction using 7,700 kcal/kg. The buffer and 60% step are Gravl smoothing heuristics; the energy conversion is based on the weight-change literature cited in 3.1.

4. Protein

Protein targets range from 1.3 to 2.4 g per kg of lean body mass depending on the preference you choose, with small additions when dieting, when highly active, and from age 40 onward — and never below the Recommended Dietary Allowance of 0.8 g per kg of body weight. These ranges are consistent with the protein intakes supported by meta-analyses and position stands for people doing resistance or endurance training, with lean-mass-based targets drawn from evidence-based contest-preparation recommendations, and the age adjustment from expert recommendations for older adults.

5. Fat, Carbohydrate and Fiber

In coached diet-style presets, the calories left after protein are split between fat and carbohydrate. These presets enforce at least 20% of calories from fat plus a height-based absolute floor, and warn when non-keto carbohydrate falls below 50 g/day. The 20% bound comes from the Acceptable Macronutrient Distribution Range; the height floor and 50 g warning are Gravl safeguards. Collaborative custom-macro mode is user-directed and can go below the coached preset floor, so it should not be read as a health recommendation. The plan-results fiber target uses 14 g per 1,000 kcal and clamps the result to 25–40 g/day. Auto nutrient goals use at least 25 g and do not apply the 40 g results-screen cap.

6. Hydration

The water goal starts at 33 ml per kg of body weight and scales up to 1.4× with activity — in line with the adequate intakes for total water established by the European Food Safety Authority (2.0 L/day for adult women, 2.5 L/day for adult men, more with physical activity and heat).

7. Micronutrients

Micronutrient goals (fiber, sugars, sodium, saturated and trans fat, cholesterol, potassium, calcium, iron, vitamins A, C and D, magnesium, phosphorus and zinc) follow the Dietary Reference Intakes published by the National Academies and NIH. Where those reference types exist, the app generally maps floor to the Estimated Average Requirement, target to the RDA or Adequate Intake for sex and age, and ceiling to the Tolerable Upper Intake Level. Fiber and guideline-based limits are exceptions. The 300 mg cholesterol value is a legacy tracking ceiling from earlier guidance, not a current individual treatment target. These values are complemented by WHO guidance for sodium, free sugars and trans fat, the NHS reference intake for total sugars, and earlier dietary guidance for cholesterol.

If you spot an error on this page or want to suggest an additional source, contact us at support@gravl.ai.