Open two macro calculators, type in the same body, and you will often get two different protein targets. Feed one of them a 70 kg adult who just wants to maintain weight, and it hands back 140 grams of protein a day. Feed the other the identical person, and it says 194. Neither calculator is broken. They just use different underlying rules, a detail most apps never mention.

This guide traces every macro number to its origin, the way our BMI, BMR and TDEE guide traced the body-weight math and our heart-rate zones guide traced the "220 minus age" formula. The goal is to separate the figures a government or UN body actually publishes from the ones that got repeated in gyms and diet books until they sounded official.

10–35%
Protein AMDR
US National Academies, a population band
0.8 g/kg
Protein RDA
A deficiency floor, not an optimum
4 / 4 / 9
Atwater factors
kcal per gram, protein / carb / fat (FAO 2003)
< 10%
WHO free sugars
Of total energy; < 5% conditional

The three macros, and the math that turns them into calories

Macronutrients are the three things in food that carry energy: protein, carbohydrate and fat. (Alcohol carries energy too, but it is not a nutrient your body needs, so no authority assigns it a target.) Every calorie count on every label rests on one old piece of arithmetic, the Atwater factors.

MacronutrientEnergy per gram
Protein4 kcal (17 kJ)
Carbohydrate4 kcal (17 kJ)
Fat9 kcal (37 kJ)
Alcohol (ethanol)7 kcal (29 kJ)

Wilbur Atwater worked these values out at the USDA around the turn of the twentieth century, and the citable modern reference is the FAO's 2003 technical report on food energy. Before you trust a calorie count to the gram, know that these are rounded averages. The same factors apply to every food, regardless of source. The FAO report says so and shows the spread. Protein energy runs from roughly 10.2 kJ/g in some vegetable proteins up to about 18.2 kJ/g in eggs, and the report notes the protein in rice combusts around 20 percent hotter than the protein in potatoes. So 4/4/9 is accurate across a mixed diet and approximate for any single item on your plate.

What the authorities actually publish

Most macro apps skip the fact that the main US science body, the National Academies, does not hand out neat splits like "40/30/30 (carbohydrate/protein/fat)." It publishes a band, the Acceptable Macronutrient Distribution Range.

MacronutrientAMDR (% of total energy)
Carbohydrate45–65%
Fat20–35%
Protein10–35%

The definition is key: the AMDR is a range of intakes associated with reduced risk of chronic disease while still meeting your needs for essential nutrients. In the source's own framing it is a population planning and assessment range, not an individual prescription. It is a band you should land somewhere inside, not a target number to hit. The protein band is wide—anything from 10 to 35 percent of calories qualifies. This is why almost any diet split can claim to be "within the guidelines" while offering no specific direction.

The other government number people quote is the protein RDA of 0.8 grams per kilogram of body weight per day. It is routinely misread. It covers the requirements of 97.5% of healthy adults, and the source frames it as a minimum to prevent deficiency rather than the amount for peak health. For an 80 kg adult that is 64 grams a day. It is a floor, not a fitness target, and sitting exactly on it is not the same as optimising.

What is convention dressed as a standard

Now the numbers with no government pedigree at all. Two of them dominate fitness culture.

The first is 40/30/30, the split behind the Zone diet. It comes from The Zone, a 1995 book by Dr. Barry Sears, a private researcher. No health authority derived it and none endorsed it. It is a commercial brand that happens to use round numbers. That does not make it harmful, but it does mean it carries no more official weight than any other book's plan, and it should never be cited alongside the AMDR as though the two sit on the same shelf.

The second is the gym maxim "one gram of protein per pound of bodyweight." The conversion is revealing: one gram per pound is 1 gram divided by 0.4536 kilograms, or 2.2 g/kg. Hold that against the best-sourced athlete guidance available. The International Society of Sports Nutrition, in its 2017 position stand, puts sufficient protein for building or keeping muscle at 1.4 to 2.0 g/kg for most exercising people, rising to 2.3 to 3.1 g/kg only for holding lean mass during a real calorie deficit. The gym rule of 2.2 g/kg sits above the top of the ISSN's general-athlete range.

One label point. The ISSN is a sports-nutrition organisation stating a position for people who train, not a government setting a reference intake. It is the right counterweight to the gym numbers because it is sourced and specific, but it is not a DRI and should not be presented as one. Even so, it caps the everyday athlete at 2.0 g/kg, which already puts the "1 g/lb" rule outside the lines.

Sugar and fibre: the two numbers worth memorising

If macro percentages feel like too much machinery, two intake numbers are simple and well sourced.

Free sugars under 10%The WHO strongly recommends free sugars stay below 10% of total energy for adults and children, with a conditional push toward under 5%. "Free sugars" means sugars added by a manufacturer, cook or you, plus the sugars in honey, syrups and fruit juice.
Fibre, 14 g per 1,000 kcalThe US Dietary Guidelines set fibre intake at 14 grams per 1,000 calories, tied to lower coronary-heart-disease risk. For most adults that lands around 22 to 34 grams a day depending on age and sex.

Both sit inside the carbohydrate picture rather than beside it. Sugar is a subset of your carbohydrate, and fibre is a carbohydrate your body largely does not digest for energy, one more reason the neat "4 calories per gram of carb" rule is an approximation.

Two ways to set protein, and why they disagree

This brings us to the two common methods for setting a protein target, which can produce very different results for the same person.

Method one anchors to body weight. You pick grams per kilogram and multiply by your weight. This is the method behind the ISSN numbers and most evidence-based sports nutrition, because protein needs track lean tissue more closely than they track calories.

Method two anchors to calories. You take a percentage of your daily energy as protein, the way the AMDR and most diet-app presets are written. The trouble is that the same percentage gives very different grams depending on how much you eat, and it can drift far from anything the per-kilogram evidence supports.

Watch what happens to one person under each method. Take a 70 kg adult eating about 2,585 calories a day to maintain weight.

Protein for one 70 kg adult, grams per day — by method
Same person, same calories. The only thing that changes is which rule sets the target.
RDA floor (0.8 g/kg)
56 g
ISSN athlete top (2.0 g/kg)
140 g
"1 g/lb" gym rule (2.2 g/kg)
154 g
30%-of-calories preset (2.77 g/kg)
194 g
Source: RECATOOLS macro-method comparison, 21 Jul 2026. Calorie figure from a Mifflin–St Jeor estimate.

The percentage method lands highest here. Thirty percent of 2,585 calories is 776 protein calories, which at 4 calories a gram is 194 grams, or 2.77 g/kg. That is more than three times the RDA and above the ISSN's general-athlete ceiling, produced by a preset labelled "balanced." A body-weight method set to 2.0 g/kg gives 140 grams for the same person. The 54-gram daily gap—a 39% difference for the same person—is purely an artifact of the two methods.

We should be straight about our own house, because this guide is about method honesty. RECATOOLS ships two macro tools, and they use these two engines. Our macro calculator uses the body-weight method: you give it your weight, your calories and a grams-per-kilogram figure (default 2.0 g/kg), and it sets protein from that, fat from a percentage, and carbohydrate from the remainder. Our other tool, the plural-named macros calculator, uses the percentage method, and its default "balanced" preset is the 30% that produced the 194-gram figure above. For the same person the two disagree by the margin you just saw. Neither is broken. A calculator's answer is only as meaningful as the method behind it, and most tools never name their method.

For the hands-on math here we link one tool, the macro calculator, because its per-kilogram method is the one the protein evidence supports and it states what it is doing. Set the grams-per-kilogram figure yourself using the ISSN range as a guide rather than leaving it on any default, and you get a target you can reason about.

The ASEAN version: a plate, not a spreadsheet

Singapore's Health Promotion Board makes the same point without a single gram. Its My Healthy Plate guidance, published through HealthHub, frames a balanced meal by area on the plate rather than by percentage of calories: half the plate fruit and vegetables, a quarter wholegrains, and a quarter meat and other protein.

It is a deliberately low-math tool, and for most people it is enough. You do not need to know that carbohydrate runs 4 calories a gram to fill a quarter of your plate with wholegrains. The plate model and the macro model answer the same question at different resolutions.

When macros stop being general advice

Two conditions move macro targets out of the population guidance covered here and into medical nutrition therapy, which is individualised and clinician-directed. Diabetes changes how carbohydrate is planned, with no one-size split; a care team sets it per person. Chronic kidney disease changes protein needs, again individually and with a clinician. Neither the AMDR bands nor any calculator on this site is built for those situations, and none of the numbers here should be used to manage a diagnosed condition.

Both of our macro tools carry disclaimers — one routes you to a physician, the other to a registered dietitian — and that deference is the right framing. These are planning aids for generally healthy adults, not treatment.

Frequently asked questions

Is there one correct macro split?

No. The nearest thing to an official position is the AMDR, and it is a wide range rather than a single split: 45–65% carbohydrate, 20–35% fat, 10–35% protein. Almost every popular split fits inside it, which is exactly why "within the guidelines" tells you very little. Your right target depends on your goals, your training and your calories, not on a number a diet book made memorable.

Why do two macro calculators give me different protein numbers?

Because they usually set protein by different methods. One anchors protein to your body weight in grams per kilogram; another takes a fixed percentage of your calories. For the same person those can differ by 50 grams a day or more, as our own two tools do. Before trusting any calculator, find out which method it uses.

Should I eat one gram of protein per pound of bodyweight?

That rule works out to 2.2 g/kg, which is above the top of the ISSN's 1.4–2.0 g/kg range for most exercising people. It is a gym convention, not a sourced recommendation. Higher intakes up to about 3.1 g/kg have support specifically for holding lean mass during a calorie deficit, but that is a narrow case from a sports-nutrition body, not general advice.

Do calorie counts on labels use real per-food energy values?

Mostly they use the rounded Atwater averages of 4, 4 and 9 calories per gram, which the FAO applies across all foods. Actual energy varies by source, so the number is reliable across a mixed diet and approximate for any single item. Fibre and hydration matter too, which is why our water intake guide makes a similar point about round numbers that get treated as precise.

This guide is general information, not medical advice. Macro ranges are population guidance, not personal prescriptions, and none of the figures here should be used to manage diabetes, kidney disease or any other condition. If you have a medical condition or specific performance goals, consult a qualified doctor or registered dietitian.

Sources & verification