# Macronutrients explained: protein, carbs, and fat

July 16, 2026 · Nutrition & Macros · 10 min read · https://burnweek.fit/blog/macronutrients-explained/

> The 4/4/9 on every label is a 130-year-old estimate of what your body can extract — not what is in the food. Whole almonds miss the prediction by 32%.

**Key takeaways**

- Protein and carbs supply about 4 kcal per gram, fat about 9, alcohol about 7 — Atwater's factors, set at the USDA in the 1890s and still printed on every label.
- 4/4/9 estimates absorbed energy, not energy in food: it predicts ~90% of gross energy as metabolizable, overestimates high-fiber diets, and misses whole almonds by 32%.
- Official ranges are wide on purpose: 10-35% of calories from protein, 45-65% from carbs, 20-35% from fat (IOM AMDR). They're boundaries, not bullseyes.
- Only protein and fat have components you must eat — 9 essential amino acids, plus linoleic and alpha-linolenic acid. Dietary carbohydrate is conditionally essential.
- Digestion costs differ by macro (protein 20-30%, carbs 5-10%, fat 0-3%), but total thermic effect is only ~10% of daily energy expenditure — real, not a shortcut.

## What a macronutrient actually is

Here are macronutrients explained without the jargon: they are the three components of food you eat in gram quantities and that supply energy — protein, carbohydrate, and fat. (Alcohol is a fourth energy source, but nobody needs it, so it rarely makes the list.) Everything else in food — vitamins, minerals, water, fiber's non-caloric fraction — is either a micronutrient, needed in milligrams, or has no calories at all. When people say "tracking macros," they mean splitting a day's calorie total into those three buckets instead of leaving it as one lump number.

The conversion rates are the first thing to learn: protein and carbohydrate supply about 4 kcal per gram, fat about 9, and alcohol about 7 ([FAO, 2003](https://www.fao.org/4/y5022e/y5022e04.htm)). That is why a spoonful of oil outweighs a spoonful of sugar in calories despite looking smaller. But those numbers are averages with a real history and real limits, and understanding where they came from is the difference between using macros well and treating them as physics.

## Where 4/4/9 came from — and what it hides

The factors on every nutrition label are older than the label itself. They come from the Atwater general factor system, developed by W.O. Atwater and colleagues at the USDA Agricultural Experiment Station in Storrs, Connecticut at the end of the nineteenth century ([FAO, 2003](https://www.fao.org/4/y5022e/y5022e04.htm)). Atwater did not simply burn food and read the heat. He took the heats of combustion of protein, fat, and carbohydrate and *corrected* them for losses in digestion, absorption, and the urinary excretion of urea — which is why protein gets 4 kcal/g rather than the roughly 5.6 it releases in a bomb calorimeter. The unrounded factor for alcohol, incidentally, is 6.9 kcal/g.

So 4/4/9 is not the energy in your food. It is an estimate of the energy your body can extract from your food, built on a small number of human subjects studied more than a century ago ([Roberts & Flaherman, 2022](https://pmc.ncbi.nlm.nih.gov/articles/PMC9776627/)). Across a typical mixed diet the system holds up well — roughly 90% of gross food energy is available as metabolizable energy, which is what the factors are calibrated to predict.

> The number on the package is not the energy in the food. It is a century-old estimate of the energy your body will manage to get out of it.

The cracks show at the edges. Atwater factors appear to overestimate available energy in higher-fiber diets, and the measured metabolizable energy of whole almonds came in 32% below what the factors predict, because the intact cell walls shield fat from digestion ([Roberts & Flaherman, 2022](https://pmc.ncbi.nlm.nih.gov/articles/PMC9776627/)). Fiber is a particular oddity: in the original system, carbohydrate was determined by difference, so it simply included fiber ([FAO, 2003](https://www.fao.org/4/y5022e/y5022e04.htm)) — one reason the "carbohydrate" line on a label is a mixed bag rather than a single substance. None of this makes macros useless. It makes them estimates, which is the same reason [calorie counts are ranges](/blog/why-calorie-counts-are-ranges) and why [the number printed on a nutrition label carries its own error bar](/blog/how-accurate-are-nutrition-labels).

## Protein: the one with a floor under it

Protein is a chain of amino acids, and its job is structural before it is energetic — muscle, enzymes, hormones, immune cells. Of the twenty amino acids that build human protein, nine cannot be synthesized by human cells at all and must come from food: histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine ([Lopez & Mohiuddin, StatPearls](https://www.ncbi.nlm.nih.gov/books/NBK557845/)). Those nine are the reason "protein quality" is a real concept rather than marketing — a food's protein is only as useful as its weakest essential amino acid.

A few more are *conditionally* essential: your body normally makes them, but under illness or particular life stages it cannot keep up ([Lopez & Mohiuddin, StatPearls](https://www.ncbi.nlm.nih.gov/books/NBK557845/)). That nuance matters because it is the pattern across all three macros — essentiality is not a fixed property, it depends on who you are and what is happening to you.

The official guidance for protein is 0.8 g/kg/day as an RDA and 10–35% of energy as an acceptable range ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)). Those two numbers answer different questions: the RDA is a deficiency floor, the range is a window of intakes considered acceptable, and neither is a target. Working out what to actually aim at is the whole of [how much protein you actually need per day](/blog/how-much-protein-per-day).

## Carbohydrate: the essential nutrient that isn't

Carbohydrate is the body's most convenient fuel and — this surprises people — the only macro without a single dietary component you strictly must eat. The RDA is set at 130 g/day, and it was derived from the average amount of glucose the brain uses in a day ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)). But the brain's need for glucose is not the same as a need for dietary glucose: the body manufactures it. A review applying formal nutrient-essentiality criteria concluded that dietary carbohydrate is best classified as *conditionally essential* — not required in the diet for the general population, but genuinely required for specific subpopulations, such as people with glycogen storage disease ([Tondt et al., 2020](https://pubmed.ncbi.nlm.nih.gov/32102704/)). That conclusion is worth reading next to the paper's own disclosure, which records that senior author Eric Westman "has ownership of weight loss and nutritional supplement companies" while his two co-authors declare none ([conflict-of-interest statement](https://www.cambridge.org/core/journals/nutrition-research-reviews/article/application-of-nutrient-essentiality-criteria-to-dietary-carbohydrates/248C77609903561AE0F4F1B454C648D7)). The biochemistry above is not in dispute, but a finding this useful to low-carbohydrate marketing has earned the disclosure printed beside it.

That is not an argument for cutting carbs. "Not strictly essential" is a biochemical fact, not a dietary recommendation, and the acceptable range is 45–65% of energy for a reason — carbohydrate-rich foods carry [fiber](/blog/fiber-benefits-and-targets) and micronutrients that are hard to replace ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)). The practical question is how much suits your training, appetite, and preferences, which is the subject of [how many carbs per day](/blog/how-many-carbs-per-day).

## Fat: dense, and partly non-negotiable

Fat carries 9 kcal/g — more than double the other two — which makes it the macro where portion errors cost the most calories. It is also the macro with hard essential requirements. Two fatty acids cannot be made by the body: linoleic acid (an omega-6) and α-linolenic acid (an omega-3). The DRI tables give them their own ranges: 5–10% of energy for linoleic acid and 0.6–1.2% for α-linolenic acid, inside a total fat range of 20–35% ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)).

So the framing that fat is "the fattening one" gets it backwards twice: fat is energy-dense but not uniquely fattening, and it is the macro you cannot zero out without eventually creating a deficiency. [How much fat per day](/blog/how-much-fat-per-day) works through the range.

## The three macros, side by side

| | Protein | Carbohydrate | Fat |
| --- | --- | --- | --- |
| Energy | ~4 kcal/g | ~4 kcal/g | ~9 kcal/g |
| Acceptable range (% energy) | 10–35% | 45–65% | 20–35% |
| Official floor | RDA 0.8 g/kg/day | RDA 130 g/day | none set; LA 5–10%, ALA 0.6–1.2% |
| Must come from food | 9 essential amino acids | none for most people | linoleic + α-linolenic acid |
| Energy burned digesting it | 20–30% | 5–10% | 0–3% |

Sources for this table: energy factors ([FAO, 2003](https://www.fao.org/4/y5022e/y5022e04.htm)); ranges and floors ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)); essential components ([StatPearls](https://www.ncbi.nlm.nih.gov/books/NBK557845/); [Tondt et al., 2020](https://pubmed.ncbi.nlm.nih.gov/32102704/)); thermic effect ([Westerterp, 2004](https://pmc.ncbi.nlm.nih.gov/articles/PMC524030/)).

Those acceptable ranges are wide on purpose. The National Academies describe the AMDR as guidance built from intervention trials and observational data, on the principle that intakes outside the range raise either chronic-disease risk or the risk of nutrient insufficiency ([NASEM, 2024](https://www.ncbi.nlm.nih.gov/books/NBK610333/)). They are boundaries, not bullseyes — and the same report stresses that qualitative choices ([the kind of fat](/blog/saturated-vs-unsaturated-fat), [the kind of carbohydrate](/blog/good-carbs-vs-bad-carbs)) still matter alongside the percentages.

## Not all calories cost the same to digest — but don't overspend the point

The last row of that table is the one people get excited about. Digesting food burns energy, and the burn is not equal across macros: reported values are 0–3% for fat, 5–10% for carbohydrate, 20–30% for protein, and 10–30% for alcohol ([Westerterp, 2004](https://pmc.ncbi.nlm.nih.gov/articles/PMC524030/)). Eat 100 calories of protein and you keep meaningfully fewer than 100.

Now the caveat, because this is where the internet oversells. On a normal mixed diet at energy balance, the whole thermic effect of food is about 10% of daily energy expenditure, ranging roughly 5–15% depending on how much protein and alcohol the diet contains ([Westerterp, 2004](https://pmc.ncbi.nlm.nih.gov/articles/PMC524030/)). Shifting your macro split moves a few percentage points of that 10% — real, worth having, and nowhere near enough to outrun a surplus. Protein's bigger contribution to weight control runs through satiety, not through the thermic bonus.

## Why macros beat a calorie total alone

Here is the practical case for tracking macros, and it is not about optimization. It is about resolution. Two 2,000-calorie days — one with 60 g of protein, one with 140 g — are the same number and completely different days: different hunger, different muscle retention, different odds you finish the week. A calorie total cannot see that difference. A macro split can.

This is also where the estimate quality gets interesting. The three macros are not equally hard to guess. Protein in a weighed chicken breast is fairly tight; the fat in a restaurant stir-fry depends on how heavy the cook was with the oil, and at 9 kcal/g that single unknown can move a meal by 200 calories. An honest tool should show you that asymmetry rather than hide it — which is why BurnWeek reports estimates as ranges and lets you correct the number that matters instead of pretending all three are equally knowable. The workflow is covered in [how to track macros and calories together](/blog/how-to-track-macros-and-calories-together).

Start with protein as your one firm target, let fat and carbs fill the rest of the calorie budget according to what you actually enjoy eating, and you have captured most of the value of macro tracking without the spreadsheet. The ranges are wide because the evidence is wide. That is permission, not imprecision.

## FAQ

### What are macros in simple terms?

Macros are the three energy-supplying parts of food — protein, carbohydrate, and fat — measured in grams. Protein and carbs give about 4 calories per gram, fat about 9. Tracking macros means seeing how a day's calories divide between those three, instead of only seeing the total.

### What is the ideal macro split?

There isn't one. The official acceptable ranges are 10–35% of calories from protein, 45–65% from carbohydrate, and 20–35% from fat ([Institute of Medicine, 2002](https://www.ncbi.nlm.nih.gov/books/NBK208874/)) — deliberately wide, because a lot of splits work. Set a protein target you can hit, then divide the remainder by preference.

### Are the 4/4/9 calorie values accurate?

They are good averages, not exact values. The factors already subtract estimated digestive and urinary losses ([FAO, 2003](https://www.fao.org/4/y5022e/y5022e04.htm)), and they predict roughly 90% of gross energy as metabolizable for typical diets — but they overestimate energy from high-fiber foods, and whole almonds deliver about 32% less than predicted ([Roberts & Flaherman, 2022](https://pmc.ncbi.nlm.nih.gov/articles/PMC9776627/)).

## Sources

- [FAO. Food energy — methods of analysis and conversion factors. FAO Food and Nutrition Paper 77, Chapter 3: Calculation of the energy content of foods. 2003.](https://www.fao.org/4/y5022e/y5022e04.htm)
- [Roberts SB, Flaherman V. Dietary Energy. Adv Nutr. 2022.](https://pmc.ncbi.nlm.nih.gov/articles/PMC9776627/)
- [Institute of Medicine. Dietary Reference Intakes: Reference Tables (AMDR, RDA for macronutrients). National Academies Press, 2002/2003.](https://www.ncbi.nlm.nih.gov/books/NBK208874/)
- [National Academies of Sciences, Engineering, and Medicine. Rethinking the Acceptable Macronutrient Distribution Range for the 21st Century: A Letter Report. 2024.](https://www.ncbi.nlm.nih.gov/books/NBK610333/)
- [Westerterp KR. Diet induced thermogenesis. Nutr Metab (Lond). 2004.](https://pmc.ncbi.nlm.nih.gov/articles/PMC524030/)
- [Lopez MJ, Mohiuddin SS. Biochemistry, Essential Amino Acids. StatPearls, 2024.](https://www.ncbi.nlm.nih.gov/books/NBK557845/)
- [Tondt J, Yancy WS, Westman EC. Application of nutrient essentiality criteria to dietary carbohydrates. Nutr Res Rev. 2020.](https://pubmed.ncbi.nlm.nih.gov/32102704/)
- [Tondt J, Yancy WS, Westman EC. Application of nutrient essentiality criteria to dietary carbohydrates: full text, financial support and conflicts-of-interest declaration. Nutr Res Rev. 2020;33(2):260-270.](https://www.cambridge.org/core/journals/nutrition-research-reviews/article/application-of-nutrient-essentiality-criteria-to-dietary-carbohydrates/248C77609903561AE0F4F1B454C648D7)