Athlete Energy & Macronutrient Calculator

Where the energy goes How training was costed

Resting metabolic rate — and what the newest research changed

• Metabolic rate does not decline through the 30s and 40s. Pontzer and colleagues analysed doubly-labelled-water measurements in 6,421 people aged 8 days to 95 years and found four distinct life stages: expenditure runs about 50% above adult values at age 1, declines to adult levels by about 20, then stays stable from 20 to 60 — even through pregnancy — before declining in older adults. The widely believed mid-life slowdown is not there once body composition is accounted for.
• Fat-free mass is the real driver, in a power-law relationship. That is the central finding, and it is why this tool now includes the Cunningham equation (500 + 22 × FFM) as a full member of the synthesis rather than a sidebar. Cunningham carries no age term at all, which is exactly what the 20–60 stability implies.
• How the synthesis is built. The four equations are evaluated at a reference age of 30 — inside the stable window — and averaged. That keeps every bit of their body-size sensitivity while freezing the age penalty. A life-stage multiplier is then applied: below 20, 1 + 0.012 × (20 − age), still descending toward adult values; flat from 20 to 60; and 1 − 0.007 × (age − 60) above 60. Anchoring to the validated equations rather than fitting something new keeps the magnitude honest. The post-60 decline is retained because it is real beyond body composition: Fukagawa found older men lower than young men even after adjusting for fat-free mass, and Gallagher found measured expenditure in older adults running 9.5–13% below what their actual organ masses predicted — organ atrophy alone does not account for it, so a body-fat figure cannot capture it either.
• Why the age terms had to go. Mifflin-St Jeor subtracts about 5 kcal per year and Harris-Benedict about 5.7 — roughly 200 kcal across a 20-to-60 span. Those coefficients come from cross-sectional data where older subjects carry less fat-free mass on average. Since this tool asks for body fat directly, leaving them in counts the same decline twice. For a 70 kg athlete at 15% body fat, the earlier version carried about 141 kcal of unsupported decline by age 60.
• Sex differences are mostly — but not entirely — body composition. Unadjusted, the gap is large: Arciero and colleagues measured 522 adults and found men 23% higher (1,740 vs 1,348 kcal/day), and in NCAA athletes Jagim found 2,481 vs 1,553. Almost all of that is body size and composition. Once fat-free mass, fat mass and aerobic fitness are controlled, the gap collapses to about 3% — and in Jagim's athletes it was not statistically detectable at all (31.1 vs 33.6 kcal/kg FFM/day, p = 0.12, women numerically higher).
• Why a small residual survives. It is not that female tissue burns less. Javed and colleagues measured liver, kidney, spleen, heart and brain mass by MRI: adding organ mass to a model already containing age, sex, fat and fat-free mass rendered sex statistically insignificant. Müller's review of organ-specific metabolic rates likewise finds no sex difference in the rates themselves. At matched total fat-free mass, men and women differ slightly in the organ-to-muscle split, and muscle is metabolically cheap (about 13 kcal/kg/day) next to liver and kidney (200–440). So the difference is body composition at a finer grain than a body-fat percentage can resolve.
• How this tool handles it. The four equations carry sex offsets sized for the unadjusted 20%+ gap, which produced roughly 8% here — two to three times too large for a population that supplies its own body-fat figure. Instead the equations are evaluated at age 30 for both sexes and averaged, giving a sex-neutral anchor, and a single explicit ±1.5% factor is applied on top. That reproduces Arciero's 3.0% residual, keeps the assumption visible in one number rather than buried across three regression constants, and mirrors exactly how the age term is handled.

Where body fat actually feeds through

• It is not decorative. Fat-free mass drives four outputs: the Cunningham equation inside the BMR synthesis, energy availability (expressed per kg of fat-free mass) and the fat-mobilisation ceiling that sets the goal slider's limit, the protein target in a deficit (specified per kg FFM, not body mass), and therefore the carbohydrate remainder. Move the body-fat slider and all four move.
• The default follows the sex selection — 12% for male, 20% for female, typical of trained endurance athletes rather than the general population. Both are placeholders, not measurements, and the slider covers 5–35% either way; once you move it, switching sex will not reset it. At a fixed body mass, going from 10% to 25% body fat changes fat-free mass by over 10 kg, which is a meaningful shift in every one of those outputs — so an actual measurement is worth more here than any other single input.

Daily living outside training

• What the activity multiplier actually contains. It is not a step count. Roughly 40% of it is the thermic effect of food — the energy spent digesting and storing what you eat, which Westerterp's review puts at 5–15% of daily expenditure for a mixed diet at energy balance. The remainder is posture and low-grade movement spread across sixteen waking hours: standing, walking between rooms, cooking, commuting, fidgeting. For a 70 kg desk worker the multiplier adds about 515 kcal, of which roughly 223 is food thermogenesis and 292 is movement.
• Why 1.30 is conservative rather than generous. Comparing it to a walk is misleading — walking is a cheap activity concentrated into a short time, whereas this is a small cost smeared across the whole day. Built bottom-up from MET values, an ordinary desk day (8 h sleeping, 8 h seated work at 1.5 METs, 4 h light household at 1.8, 2 h standing and cooking at 2.2, 2 h seated leisure at 1.3) comes to about 660 kcal above resting before food thermogenesis, implying a multiplier near 1.54. Black and colleagues' analysis of 574 doubly-labelled-water measurements places the floor of human daily expenditure at 1.2 × BMR in non-ambulant subjects and the ceiling at 4.5 × in elite endurance athletes. The 1.30 used for "desk-based" therefore sits one notch above bed-bound, and is deliberately set low so the tool does not inflate intake.

Reading the availability bar

• Energy availability is intake minus exercise energy expenditure, per kg of fat-free mass per day. It asks how much energy is left for everything that is not training. 30 is Loucks and Thuma's measured threshold — below it, luteinising hormone pulsatility was disrupted in their subjects. 45 is the reference for full function and energy balance, carried into the IOC's REDs consensus. 60 is simply the end of the scale, not a target; there is no benefit implied by sitting at the far right.
• It only means something if you train. With little or no training load the calculation collapses to maintenance divided by fat-free mass, which for a genuinely sedentary person legitimately lands around 35–40 — below the 45 reference without anything being wrong. The thresholds presume a training athlete, so the bar is greyed and the warning bands suppressed below about 3 kcal/kg/day of training.

Costing the training

• Running: distance × body mass. The ACSM running equation gives a net oxygen cost of 0.2 ml/kg per metre — 200 ml/kg per km, about 1.0 kcal per kg per km net of resting metabolism. Pace barely changes cost per kilometre, so it is not an input: the net cost per unit distance is approximately independent of speed, and mass-specific cost is independent of body mass across a wide range.
• The model is strictly additive. Daily requirement = resting metabolic rate × the daily-living factor, plus the gross energy cost of training, plus the recovery term. Nothing is subtracted. The work on the bike happens regardless of what resting metabolism is doing underneath it, and the accounting is both simpler and consistent with what head units and training platforms report.
• Cycling: kilojoules ÷ 21% gross efficiency. A power meter records external mechanical work; metabolic energy is that work divided by gross mechanical efficiency, then divided by 4.184 to reach kilocalories. At 21% that is 1 kJ ÷ 0.21 ÷ 4.184 = 1.14 kcal. So 3,000 kJ of work costs about 3,410 kcal. Trained cyclists sit around 20–23% gross; 21% is a reasonable central value and, if anything, slightly conservative.
• Running: 1.08 kcal per kg per km, gross. The ACSM running equation gives gross oxygen cost as 0.2 × speed + 3.5 ml/kg/min, which works out to 1.07–1.11 kcal/kg/km across normal training speeds. Pace is not an input because the cost per kilometre barely moves with it — distance and body mass are what matter.
• Strength: 5 METs gross. Same convention as everything else.
• Cycling by distance is a rough fallback using about 30 kJ per km. It cannot see terrain, wind, drafting or how hard you rode, and unlike running, cycling energy is strongly speed-dependent because drag rises with the cube of velocity. Use a power meter if you have one.
• Strength is costed at 5 METs, net of rest.
• Recovery cost is real but small. Excess post-exercise oxygen consumption runs at 6–15% of the net oxygen cost of the exercise itself, and exercise influences body mass predominantly through energy spent during the session. A 10% term is applied.
• Carbohydrate need is computed from substrate use, not from volume alone. Training does not run on carbohydrate exclusively, so counting every training calorie against the carbohydrate target over-prescribes it. The requirement here is a baseline of 3.6 g/kg/day — covering brain, central nervous system and resting glycogen turnover — plus only the carbohydrate-derived share of session energy, divided by 4 kcal per gram. The 3.5 g/kg baseline is set so the result reconciles with the published volume bands at the 35% default.
• Why 35% from fat is the default. Tracer work by Romijn and colleagues across 25, 65 and 85% VO₂max shows carbohydrate uptake and glycogen oxidation rising directly with intensity while fat's contribution falls. Achten and Jeukendrup put maximal fat oxidation at 64 ± 4% VO₂max, the Fatmax zone at 55–72%, and fat's contribution as negligible above 89 ± 3% VO₂max. Weighting those zone-level fractions by the energy each zone contributes gives roughly 35% fat across a realistic week — and, interestingly, the figure barely moves between polarised, pyramidal and threshold-heavy weeks (35%, 37%, 34%), because harder sessions burn more energy per hour, which offsets their lower fat fraction. A default of 35% therefore sits right on the computed central estimate, and because the figure is so insensitive to how the week is shaped it is fixed at 35% rather than exposed as a control — a slider spanning the realistic 30–40% range moved no prescribed target at all, only the displayed requirement.
• Sense-check against the published bands. This construction reproduces the ACSM volume tiers across the whole range: 0.5 h/day lands at 4.3 g/kg against their 3–5; 1 h at 5.0 against 5–7; 2 h at 6.6 and 3 h at 8.1 against 6–10; 4 h at 9.6 and 5 h at 11.2 against 8–12. It sits inside every band while being continuous rather than stepped, and it responds to how the week is actually ridden or run. The volume-only band is still displayed as a cross-check.

Why the goal slider is bounded

• The scale runs −0.75% to +0.5% of body mass per week. Where a limit bites sooner, the unreachable part of the track is hatched and the marker is labelled with the actual cap in kg or lb per week, so you can read the ceiling without dragging to it.
• Three limits, whichever binds first. Supply — fat stores release energy at a finite rate. Alpert derived a ceiling of 290 ± 25 kJ per kg of fat per day (about 69 kcal/kg, or 31 kcal/lb) from underfed subjects, and showed that a deficit exceeding it does not come out of fat at all: it comes straight out of fat-free mass. Function — energy left after training must still cover resting metabolism. Rate — 0.75% of body mass per week, for lean mass preservation.
• Why supply is the honest lean-mass constraint. It scales with fat mass, which is the thing actually being drawn down. A leaner athlete gets a tighter cap because there is less fat to mobilise per day, which is why the last few percent always comes off slowest and why aggressive cuts cost lean tissue precisely in the people who can least afford it. Earlier versions of this tool used a fixed energy-availability floor instead, and it produced the wrong answer: permitted deficit fell as athletes got bigger, so a 120 kg athlete was allowed 207 kcal/day against 587 for a 60 kg athlete at identical body fat. The fault was structural — resting metabolism scales sub-linearly with fat-free mass because every predictive equation carries a large intercept, while 30 × FFM is linear through the origin, so the floor overtook the ceiling as body size rose.
• Why the resting-metabolism floor replaced the fixed availability floor. Loucks derived the 30 kcal/kg FFM/day threshold in regularly menstruating, habitually sedentary young women of normal body composition. At that body size, 30 × FFM sits within a few percent of resting metabolic rate. Expressed against each person's own RMR the same rule is 0.85× at 39 kg of fat-free mass but 1.26× at 102 kg — so ported unchanged to a large athlete it silently becomes a far stricter constraint than Loucks measured. Anchoring to RMR keeps the intent and removes the body-size artefact.
• Loss is capped by lean mass preservation. In a randomised trial in elite athletes, 0.7% of body mass per week increased lean mass by 2.1% while 1.4% per week left it unchanged, for the same fat loss. The −0.75% ceiling sits just above that supported figure.
• Adding easy aerobic volume loosens the limit rather than tightening it. Training energy raises maintenance without raising the resting-metabolism floor, so a bigger week means you can eat more and hold the same deficit.
• Gain is capped at 0.5% per week; faster and the extra is mostly fat, particularly in trained athletes.

Protein, carbohydrate and fat

• Protein is one continuous surface, not a set of cases. Earlier versions switched formula at the maintenance boundary, so an 80 kg athlete jumped from 144 g to 181 g the instant the goal slider moved off zero. That was an artefact of the code, not the physiology. The target is now a sum of terms, each of which is zero at its own origin and rises smoothly: 1.50 baseline + aerobic load + strength work + energy deficit, in g/kg fat-free mass. Nothing steps anywhere.
• The aerobic term is deliberately shallow, and here is the honest reason. The intuition that more endurance volume means more protein is weakly supported at best. Moore and colleagues ran the direct within-subject test — indicator amino acid oxidation on a rest day, a 10 km run day and a 20 km run day — and found no difference between 10 km and 20 km, with training days if anything lower than the rest day. Kato's often-quoted 1.83 g/kg/day for endurance athletes is a safe intake (upper 95% confidence bound); the estimated average requirement was 1.65. The tiered figures attributed to Tarnopolsky are cross-study and confounded by energy intake and training status rather than a measured dose-response. So this tool encodes a saturating term that gives the expected direction without pretending to a slope nobody has measured. The better-evidenced modifier is energy and carbohydrate availability, which the deficit term handles directly.
• Strength work is what carries you to the top of the range. Morton's meta-analysis of 49 trials and 1,863 participants put the plateau for resistance-training gains in fat-free mass at 1.62 g/kg/day, with a 95% confidence bound of 2.20 — and the authors' own recommendation is to aim for ~2.2 to be sure of capturing it. That is a resistance-training finding, so strength hours drive it here rather than the goal slider.
• A surplus does not raise the protein requirement. This is counter-intuitive and worth stating plainly: there is no evidence the requirement scales with the size of a surplus, and extra energy is if anything protein-sparing. An athlete gaining lean mass needs protein because they are lifting, not because they are in a surplus. The goal slider therefore has no upward effect on protein — which is also why the old jump from 144 g to 160 g on entering a surplus was wrong twice over.
• In a deficit the requirement genuinely does rise. Helms and colleagues recommend 2.3–3.1 g/kg fat-free mass for energy-restricted resistance-trained athletes, explicitly "scaled upwards with severity of caloric restriction and leanness" — so both of those drive the deficit term. Two caveats worth carrying: the population was lean, resistance-trained and male-dominated, and the scaling is the authors' inference from between-study patterns rather than a tested dose-response. No equivalent quantitative figure exists for endurance athletes in a deficit.
• Ceiling and floor. Outside a deficit the ceiling is Morton's 2.20 g/kg body mass; inside one it blends smoothly toward Helms' 3.10 g/kg fat-free mass as the restriction deepens. There is no demonstrated benefit above either. A one-year crossover at about 3.3 g/kg/day in resistance-trained men found no adverse renal, hepatic or lipid effects, so the ceiling is about futility rather than safety. The floor is 15% of maintenance energy — capped at the plateau, and computed against maintenance rather than target intake so the goal slider cannot drag it around.
• Why protein-as-a-percentage stops working at high volume. No position stand expresses athlete protein needs as a share of energy; they all use g/kg, and for good reason. The 15% floor holds comfortably up to roughly 9,000 kJ of riding a week. Past that, carbohydrate intake grows so large that protein falls to 12%, then 10%, then under 10% of energy — while still sitting at 2.20 g/kg body mass, which is the top of the evidence. Holding 15% at 7,400 kcal would mean 276 g, or 3.45 g/kg, well past any measured requirement. The percentage is a display artefact at that end of the range; the g/kg figure is the one that means something.
• Fat's share of energy falls as training rises — from about 33% at rest toward 25% at high volume. This is not an evidence-based fat recommendation and should not be read as one. It is the arithmetic consequence of carbohydrate need scaling with the work while protein is set per kg of body composition: at low training loads there is simply no call for high carbohydrate, so fat can take more of the plate. The whole range stays inside the ACSM/AND/DC band of 20–35% of energy, and the joint stand's caution against chronic intake below 20% sets the floor. The commonly cited 0.8–1.0 g/kg minimum is practitioner convention rather than a position-stand figure, and is retained here only as a secondary floor. Protein is set first because it is least negotiable, fat then takes its share, and carbohydrate takes the remainder — except where carbohydrate need would go unmet, in which case fat gives way toward its floor rather than the athlete simply being told they are short.
• A caution on the fat share. Raising it lowers the carbohydrate requirement, which makes a deficit look more comfortable than it is. It describes the fuel mix of the training you are already doing; it is not a lever for making an aggressive plan pass. Pre-exercise carbohydrate feeding also shifts the mix — ingesting 75 g of glucose 45 minutes beforehand cut maximal fat oxidation by 28% and lowered Fatmax by 14%.
• Where the deficit sits in the week matters as much as its size. Every figure here is a daily average, and averaging hides the most useful degree of freedom a coach has. A big aerobic day both raises expenditure and mobilises a large amount of fat: a 3,000 kJ ride costs roughly 2,850 kcal, of which about 1,000 kcal comes from fat at a 35% share. An 800–1,000 kcal deficit on that day is therefore drawing on fuel the session has already liberated, and tends to feel unremarkable — even though as a percentage of that day's expenditure it is modest. A rest day inverts every part of that. Expenditure is low, so the same absolute deficit is a much larger fraction of it; appetite is typically elevated from the previous day's stimulus; and recovery and adaptation are the actual job of the day. Running the deficit flat across the week puts the hardest restriction exactly where the body is least equipped for it.
• So use the weekly average as the plan and the daily number as a variable. Bias the shortfall toward high-expenditure days, eat closer to maintenance on recovery days, and judge progress over two to three weeks rather than day to day. Appetite is a real signal here rather than a nuisance — persistent hunger on easy days usually means the distribution is wrong, not that the weekly target is.

Limitations

• The fat share of training fuel is a week-level aggregate. Within a week it varies enormously between an easy long ride and a threshold session, which is why the guidance is still to fuel the hard days properly and take any shortfall on easy ones.
• Prediction equations carry meaningful individual error, and metabolic adaptation during a sustained deficit lowers actual expenditure below the calculated figure. Adjust from two to three weeks of measured response.
• The 1 kcal/kg/km running cost is for level ground; substantial climbing raises it.
• Gross cycling efficiency is fixed at 21% and is not exposed as an input, to keep the number of controls down. Real values run roughly 20–23%, and that spread moves the cycling figure by about ±7% — still the largest single uncertainty in the training estimate.
• Because the model is additive and uses gross costs, resting metabolism during training hours is counted in both the baseline and the session. At typical volumes that is a few per cent of the daily total, well inside the error of the underlying equations, and it keeps the arithmetic transparent.
• The 10% recovery term is a midpoint of the 6–15% range, applied uniformly.
• Rate targets use the conventional 7,700 kcal per kg. Real weight change includes water and glycogen, so short-term scale movement will not match.
• Does not account for pregnancy, growth in adolescent athletes, clinical conditions or medication effects.

References

Literature retrieved via PubMed.

• Moore DR, Gillen JB, West DWD, Kato H, Volterman KA. Protein requirements may be lower on a training compared to rest day but are not influenced by moderate training volumes. Appl Physiol Nutr Metab 2024. doi:10.1139/apnm-2023-0297
• Witard OC, Hearris M, Morgan PT. Protein requirements for endurance athletes. Sports Med 2025;55:1361–76. doi:10.1007/s40279-025-02203-8
• Antonio J, et al. A high protein diet has no harmful effects: a one-year crossover study in resistance-trained males. J Nutr Metab 2016;2016:9104792. PMID 27807480
• Delany LV, et al. Nutrition recommendations in sport: a scoping review of consensus and position statements. Sports Med 2025;55:2445–87. doi:10.1007/s40279-025-02285-4
• Alpert SS. A limit on the energy transfer rate from the human fat store in hypophagia. J Theor Biol 2005;233(1):1–13. doi:10.1016/j.jtbi.2004.08.029
• Loucks AB, Thuma JR. Luteinizing hormone pulsatility is disrupted at a threshold of energy availability in regularly menstruating women. J Clin Endocrinol Metab 2003;88(1):297–311. doi:10.1210/jc.2002-020369
• Westerterp KR. Diet induced thermogenesis. Nutr Metab (Lond) 2004;1(1):5. doi:10.1186/1743-7075-1-5
• Black AE, Coward WA, Cole TJ, Prentice AM. Human energy expenditure in affluent societies: an analysis of 574 doubly-labelled water measurements. Eur J Clin Nutr 1996;50(2):72–92. PMID 8641250
• Pontzer H, et al. Daily energy expenditure through the human life course. Science 2021;373(6556):808–12. doi:10.1126/science.abe5017
• Arciero PJ, Goran MI, Poehlman ET. Resting metabolic rate is lower in women than in men. J Appl Physiol 1993;75(6):2514–20. doi:10.1152/jappl.1993.75.6.2514
• Ferraro R, et al. Lower sedentary metabolic rate in women compared with men. J Clin Invest 1992;90(3):780–4. doi:10.1172/JCI115951
• Javed F, et al. Brain and high metabolic rate organ mass: contributions to resting energy expenditure beyond fat-free mass. Am J Clin Nutr 2010;91(4):907–12. doi:10.3945/ajcn.2009.28512
• Jagim AR, et al. Sex differences in resting metabolic rate among athletes. J Strength Cond Res 2019;33(11):3008–14. doi:10.1519/JSC.0000000000002813
• Müller MJ, et al. Advances in the understanding of specific metabolic rates of major organs and tissues in humans. Curr Opin Clin Nutr Metab Care 2013;16(5):501–8. doi:10.1097/MCO.0b013e328363bdf9
• Gallagher D, et al. Smaller organ tissue mass in the elderly fails to explain lower resting metabolic rate. Ann N Y Acad Sci 2000;904:449–55. doi:10.1111/j.1749-6632.2000.tb06499.x
• Fukagawa NK, Bandini LG, Young JB. Effect of age on body composition and resting metabolic rate. Am J Physiol 1990;259(2 Pt 1):E233–8. doi:10.1152/ajpendo.1990.259.2.E233
• Thomas DT, Erdman KA, Burke LM. ACSM Joint Position Statement: Nutrition and Athletic Performance. Med Sci Sports Exerc 2016;48(3):543–68. doi:10.1249/MSS.0000000000000852
• Kato H, et al. Protein requirements are elevated in endurance athletes after exercise. PLoS One 2016;11(6):e0157406. doi:10.1371/journal.pone.0157406
• Morton RW, et al. Protein supplementation and resistance training-induced gains in muscle mass and strength. Br J Sports Med 2018;52(6):376–84. doi:10.1136/bjsports-2017-097608
• Helms ER, et al. Dietary protein during caloric restriction in resistance trained lean athletes. Int J Sport Nutr Exerc Metab 2014;24(2):127–38. doi:10.1123/ijsnem.2013-0054
• Jäger R, et al. ISSN Position Stand: protein and exercise. J Int Soc Sports Nutr 2017;14:20. doi:10.1186/s12970-017-0177-8
• Aragon AA, et al. ISSN Position Stand: diets and body composition. J Int Soc Sports Nutr 2017;14:16. doi:10.1186/s12970-017-0174-y
• Matsuda T, et al. Within-day amino acid intakes and nitrogen balance in male collegiate swimmers. Nutrients 2018;10(11):1809. doi:10.3390/nu10111809
• Loucks AB, Kiens B, Wright HH. Energy availability in athletes. J Sports Sci 2011;29(Suppl 1):S7–15. doi:10.1080/02640414.2011.588958
• Mountjoy M, et al. 2023 IOC consensus statement on Relative Energy Deficiency in Sport. Br J Sports Med 2023;57(17):1073–97. doi:10.1136/bjsports-2023-106994
• Garthe I, et al. Effect of two different weight-loss rates on body composition and performance in elite athletes. Int J Sport Nutr Exerc Metab 2011;21(2):97–104. doi:10.1123/ijsnem.21.2.97
• LaForgia J, Withers RT, Gore CJ. Effects of exercise intensity and duration on excess post-exercise oxygen consumption. J Sports Sci 2006;24(12):1247–64. doi:10.1080/02640410600552064
• Zamparo P, et al. The energy cost of walking or running on sand. Eur J Appl Physiol 1992;65(2):183–7. doi:10.1007/BF00705078
• Taboga P, et al. Energetics and mechanics of running men: the influence of body mass. Eur J Appl Physiol 2012;112(12):4027–33. doi:10.1007/s00421-012-2389-6
• Romijn JA, et al. Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. Am J Physiol 1993;265(3):E380–91. doi:10.1152/ajpendo.1993.265.3.E380
• Achten J, Gleeson M, Jeukendrup AE. Determination of the exercise intensity that elicits maximal fat oxidation. Med Sci Sports Exerc 2002;34(1):92–7. doi:10.1097/00005768-200201000-00015
• Achten J, Jeukendrup AE. The effect of pre-exercise carbohydrate feedings on the intensity that elicits maximal fat oxidation. J Sports Sci 2003;21(12):1017–24. doi:10.1080/02640410310001641403
• Mifflin MD, et al. A new predictive equation for resting energy expenditure. Am J Clin Nutr 1990;51(2):241–7. PMID 2305711

Disclaimer. Educational tool for coaching use, not medical or dietetic advice. Athletes with a history of disordered eating, or anyone considering a sustained energy deficit, should work with a registered dietitian.