Why calories come first
The most useful thing anyone has done for this argument was run it for two years. Researchers assigned 811 overweight adults to one of four diets, differing only in the split: 20% fat with 15% or 25% protein, and 40% fat with 15% or 25% protein — which put carbohydrate anywhere from 35% to 65% of calories. Same foods, same counselling, all reduced-calorie.[1]
At six months every group had lost about 6 kg — 7% of bodyweight. By two years, weight loss was 3.0 kg and 3.6 kg in the 15% and 25% protein groups, 3.3 kg in both the 20% and 40% fat groups, and 2.9 kg and 3.4 kg in the 65% and 35% carbohydrate groups. None of those differences was statistically significant. Hunger, satisfaction and attendance were similar across all four.[1]
Which reorders the whole conversation. Calories first. Protein second. Fat third. Carbohydrate as the remainder. Everything else — fasting windows, meal timing, food order, the order you eat your vegetables — is a tiebreaker at best, and worth nothing if the first number is wrong.
How to set the calorie number
Start with resting metabolic rate. The best-validated prediction is the Mifflin–St Jeor equation: a systematic review found it predicted RMR within 10% of measured in more non-obese and obese people than Harris-Benedict, Owen or the WHO/FAO/UNU equations, and with the narrowest error range of the four.[5]
Then multiply by activity. These factors are convention, not trial results — no study has randomly assigned people to activity multipliers — but they are a sane place to start:
| Situation | Multiply RMR by |
|---|---|
| Desk job, no training | 1.3 |
| Training 2–3 days a week | 1.45 |
| Training 4–5 days a week | 1.6 |
| Training daily, on your feet all day | 1.75 |
That gives a starting target. The real number comes from two or three weeks of scale data: weigh yourself daily, compare weekly averages, and if the average has not moved in the direction you want, change intake by 100–200 kcal and watch again. One bad weigh-in is noise. A flat three-week trend is information.
Protein, fat, carbohydrate — in that order
Protein
About 1.6 g/kg of bodyweight is where the dose–response for muscle gain flattens out for most people — the full argument is in how much protein you actually need.[3] In a deficit the target rises, because the job changes from building to protecting muscle. Helms and colleagues recommend 2.3–3.1 g/kg of lean body mass per day during contest preparation, which for a lean lifter works out to roughly 2 g/kg of total bodyweight, or a little above.[2] That is lean mass, not bodyweight — a distinction that can be worth tens of grams of protein a day if you are carrying much fat.
Fat
15–30% of calories is the range recommended for physique athletes,[2] and there is a reason to respect the floor rather than cut fat to whatever is left. A meta-analysis of six intervention studies, 206 men in total, found low-fat diets lowered total testosterone compared with higher-fat diets (standardised mean difference −0.38, 95% CI −0.75 to −0.01) and free testosterone (−0.37, CI −0.63 to −0.11).[4]
Six studies is a thin base for a strong claim, and none of them was a bodybuilding prep. But the direction is consistent, the cost of a fat floor is low, and “as little fat as possible” is a diet strategy you can stop worrying about.
Carbohydrate
Whatever is left. Carbohydrate is the macro that fuels hard training and the one whose number moves when calories change — not because it is uniquely fattening, but because the other two have floors and something has to give. Keeping carbohydrate high enough to train well is a performance decision, not a metabolic one.
A worked example
An 80 kg lifter with a maintenance estimate of about 2,600 kcal, cutting at 2,200 and then gaining at 2,900. Protein is set from bodyweight first, fat from the percentage floor, carbohydrate gets the remainder:
| Cutting — 2,200 kcal | Gaining — ~2,900 kcal | |
|---|---|---|
| Protein | 180 g · 720 kcal · 33% | 160 g · 640 kcal · 22% |
| Fat | 60 g · 540 kcal · 25% | 70 g · 630 kcal · 22% |
| Carbohydrate | 235 g · 940 kcal · 43% | 407 g · 1,628 kcal · 56% |
Protein goes up as calories go down, because the point of a cut is to keep the muscle you have. Carbohydrate takes the whole difference. That is the entire art of macro setting: three numbers, in a fixed order, and the last one absorbs every adjustment.
How fast should the scale move?
Rate matters as much as direction. Twenty-four athletes were randomised to lose either 0.7% or 1.4% of bodyweight per week. They actually lost 0.7% and 1.0% per week respectively. Fat mass fell 31% in the slow group and 21% in the fast group. Lean mass rose 2.1% in the slow group and was unchanged (−0.2%) in the fast group — a significant difference between them.[6]
For an 80 kg lifter, 0.7% is roughly 560 g a week. Helms and colleagues put the target at 0.5–1% of bodyweight per week to maximise muscle retention,[2] which for most people is slower than they want it to be — and faster than the alternative, once you count what you lose along the way.
Gaining has far less to go on. I could not find a trial that randomised the rate of gain and measured lean mass. That slower gains put more of the surplus into muscle and less into fat is a reasonable inference from energy partitioning, not a demonstrated result — worth saying plainly, because the internet states it as fact.
Why your calculator is wrong
Two reasons, and neither is a reason to skip the maths.
Equations describe averages. The best one is within 10% of measured for most people — and 10% of 2,500 kcal is 250 kcal a day, which is the difference between a deficit and maintenance over a month.[5]
Your body pushes back. After weight loss, energy expenditure falls by more than the loss of tissue alone would predict. This adaptive thermogenesis is part of why more than 80% of people return to their pre-loss body fatness after otherwise successful weight loss.[7] The deficit that worked at the start of a cut is rarely the one that works at the end.
How this shows up in your program
The macro calculator runs this exact order — bodyweight to protein, then a fat percentage, then carbohydrate from whatever remains. The weight timeline takes a deficit or surplus and shows what that pace looks like week by week, which is the quickest way to find out whether the rate you have in mind is realistic before you spend a month on it.
From there the training side takes over: the workout builder sets weekly volume, and how many sets actually build muscle explains why those numbers are what they are, while progressive overload covers how they get harder week to week.
What we still don’t know
- Most of this evidence comes from overweight adults in trials running 8 weeks to 2 years — not from lean physique athletes at 8% body fat, who are the people most likely to be reading it.
- Nobody has run a long trial comparing macro ratios in trained lifters at genuinely equal protein and equal calories. The big trial matched food and protein; it did not isolate the split from everything else.[1]
- The size of adaptive thermogenesis varies enormously between people and is hard to predict. The 80% regain figure is an average across studies, not a sentence for any individual.[7]
- Food records are systematically wrong. Any study built on self-reported intake — and any diet built on your memory — carries an error nobody has measured for you.
- The rate-of-gain side is far less studied than the rate-of-loss side, which is why the advice you read about bulking is usually inference dressed up as evidence.
References
- Sacks FM, Bray GA, Carey VJ, et al. Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N Engl J Med. 2009;360(9):859–873. PubMed 19246357
- Helms ER, Aragon AA, Fitschen PJ. Evidence-based recommendations for natural bodybuilding contest preparation: nutrition and supplementation. J Int Soc Sports Nutr. 2014;11:20. PubMed 24864135
- Morton RW, Murphy KT, McKellar SR, et al. A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults. Br J Sports Med. 2018;52(6):376–384. PubMed 28698222
- Whittaker J, Wu K. Low-fat diets and testosterone in men: systematic review and meta-analysis of intervention studies. J Steroid Biochem Mol Biol. 2021;210:105878. PubMed 33741447
- 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 15883556
- Garthe I, Raastad T, Refsnes PE, Koivisto A, Sundgot-Borgen J. Effect of two different weight-loss rates on body composition and strength and power-related performance in elite athletes. Int J Sport Nutr Exerc Metab. 2011;21(2):97–104. PubMed 21558571
- Rosenbaum M, Leibel RL. Adaptive thermogenesis in humans. Int J Obes (Lond). 2010;34 Suppl 1:S47–55. PubMed 20935667
Want this applied to your own numbers? Bring your current intake and your scale data from the last month, and I will set the four numbers with you and rebuild them around the rate you can actually hold.
Work with me 1:1