Macros or calories: which should you count?

What each number actually measures

Calories are the energy in food, and macros are where that energy comes from. The FAO's reference on food energy gives the conversion that nutrition tables use, the Atwater general factors: 4 kcal per gram of protein, 4 per gram of carbohydrate and 9 per gram of fat, all estimated averages. So macros are not a rival to calories. They are the same total, split into three parts, which is why a macro log always contains a calorie log inside it.

That also explains the real trade-off. Counting calories means one number per food. Counting macros means three, plus a target for each, so it takes more time and more places to make a mistake. What you buy with the extra effort is information about the split. Whether the split matters to you is the whole question, and the next section shows what it looks like.

Two days, the same calories, a different plate

The test is simple. Take one adult of 65 kg and two days of food, each close to 2,000 calories. Day A is built from convenience foods: a bagel, a sandwich, chips, a cola, pasta with a side salad, ice cream and a cookie. Day B is built from foods chosen with protein in mind: oats, chicken, salmon, Greek yogurt, rice and potatoes. Both are ordinary, and both land on the same calorie number. 2,000 is a round figure for the test, not a recommendation for anyone.

Added up, the two days are 2,008 and 1,990 calories (estimated), a gap of 18 calories, under 1%. A calorie log would show them as the same day. Row values are rounded, so a total can differ from the sum of its rows by a couple of calories.

What a calorie log sees, and what a macro log sees

Now put the two logs side by side. The calorie log reports one line per day. The macro log reports the split, and we have added the protein in grams per kilogram of body weight for our 65 kg adult, because that is how protein targets are usually stated.

Three things stand out. First, protein: Day B holds about 2.3 times as much as Day A at the same calories (estimated). Second, fat is almost identical, so on this pair the split is a protein story and a carbohydrate story, not a fat story. Third, dietary fiber, which no macro target tracks directly, moves with it: 12.8 g against 27.1 g (estimated), because the foods that carry protein in Day B also come with vegetables, oats and potatoes.

Both days sit inside the official ranges

This is the part that makes the question hard. The National Academies' acceptable macronutrient distribution ranges for adults are 10 to 35% of energy for protein, 45 to 65% for carbohydrate and 20 to 35% for fat (NASEM, 2024). Run the shares from the table through them and both days pass: Day A is 10% protein, 62% carbohydrate and 29% fat, Day B is 23%, 50% and 28% (estimated). A check against the ranges would not have flagged either day.

So what does separate them? The reference intake. The recommended dietary allowance for protein is a floor, and a published review describes it plainly:

The RDA for protein for adults ≥18 y of age (0.8 g/kg) has been essentially unchanged for >70 y.

For a 65 kg adult, 0.8 g/kg is 52 g a day. Day A, at 50.5 g (estimated), sits just under that floor while staying on 2,000 calories and inside every percentage range. That is the case for adding one macro: protein is the target a pure calorie count can quietly miss.

Where more protein stops helping

A second number does not mean a bigger number. A meta-analysis of 49 randomised trials with 1,863 participants found that protein supplementation added to resistance training improved muscle size and strength, and that intakes above about 1.6 g/kg a day gave no further gain in fat-free mass (Morton and colleagues, 2018). That applies to people who lift, in trials that supplemented protein, so treat it as a ceiling for useful effort in that group and not as a target for everyone.

For our 65 kg adult, 1.6 g/kg is 104 g. Day B, at 115.9 g (estimated), is already past it, and it did not take protein powder or an unusual food list. It took one lean protein at each meal. If you want the full method for setting a protein figure, our guide to protein per day works it through, and the macro calculator gives you the numbers for your own body.

Protein per 100 calories: why Day B works

Foods differ a lot in how much protein comes with their calories, and that ratio is what separated our two days. This table divides protein by calories for 14 common foods, from the same USDA entries used above.

Nothing in the bottom half is a bad food, and olive oil in Day B is there on purpose. The point is arithmetic: a day made mostly of foods in the bottom half struggles to reach a protein figure inside 2,000 calories, while a day with a food from the top half at each meal gets there easily. A calorie log never shows you this column.

A quick check on the arithmetic

Because macros and calories are the same total, you can test any log. Multiply protein and carbohydrate grams by 4 and fat grams by 9 and the sum should land near the calorie figure. For Day A that is 4 × 50.5 + 4 × 310.1 + 9 × 64.3 = 2,021 calories, against 2,008 from USDA's own energy values. For Day B it is 4 × 115.9 + 4 × 247.2 + 9 × 62.6 = 2,016, against 1,990 (estimated). The two methods agree within about 1.3%, so counting macros does not change the energy total, it only splits it.

So which should you count?

Here is how the pieces fit, as a way to decide and not as a rule. Whatever you choose, use it for a few weeks before you judge it, because a consistent method beats a precise one you drop.

• You want the simplest start: count calories. One number per food, one daily total. It is enough to know how much you eat, and you can add a second number later.
• You lift or train, or you suspect your protein is low: add protein in grams. Calories plus protein is a two-number method that catches the gap shown above without tracking three targets.
• You have a specific plan from a professional: count what they gave you. That may include carbohydrate, fat, fiber or sodium. Follow their targets, not this page.
• Weighing and logging makes eating stressful: do neither. Planning ahead is a legitimate alternative, covered below.

If you do count macros, the biggest source of error is not the three numbers. It is the weight and the entry disagreeing about whether the food is raw or cooked, which we worked through in count macros raw or cooked. The how to calculate macros guide shows the four-step method that produces the targets.

A caution about counting

Counting can help some people understand portions and feel counterproductive for others. If tracking makes eating stressful, if you skip meals or avoid foods because you cannot log them, or if you have or have had an eating disorder, do not use this page to tighten your tracking. Talk to a doctor or a registered dietitian about a method that suits you. If you have a medical condition such as diabetes, are pregnant or breastfeeding, take a medicine that affects appetite or weight, or are under 18, ask a professional for your numbers before you follow any figure on this page. This is general information, not medical advice.

If you would rather plan than log

This part is written by the team behind MacroRecipe, so read it with that in mind. There is a third way to get what macro counting gives you: decide what you will eat before the week starts, so the numbers are already attached to the meals. Day B above is a small example, a day built backwards from a protein figure, and our guide to planning the week shows the same idea over seven days.

MacroRecipe does that for you. It plans your week of menus around your macros and the goal you choose, from dishes you pick, with ingredient data from USDA and CIQUAL (ANSES), and sizes the portions toward your estimated calorie and protein targets. When you swap a dish, the portions and the grocery list follow. You cook from a menu instead of logging what you ate.

Who it isn't for: MacroRecipe runs on iPhone only, not on Android. It has no barcode scanner, no food database to search and no complete food diary, though a photo scan of a meal eaten elsewhere comes with the subscription. You can preview your first menu for free, after which it is a paid subscription. If you would rather log everything you eat, our MacroFactor alternatives and MyFitnessPal alternatives guides cover trackers.

How we calculated

Every value is a USDA FoodData Central SR Legacy entry per 100 g, read on 2026-10-28 from the April 2018 CSV on the FoodData Central download page, by the FDC ID shown in the protein table (the other entries are 174924 white bread, 172941 sliced turkey breast, 169737 cooked pasta, 173418 cream cheese, 169098 raw orange juice, 173592 ranch dressing, 169247 romaine, 172717 soft-type chocolate chip cookies, 171265 whole milk, 173944 banana, 169967 cooked broccoli, 170093 baked potato with skin). Calories are USDA's own energy values in kcal, not recomputed from macros, and all of them are estimated.

• Each row. Grams × value per 100 g ÷ 100. Day A totals: 2,007.6 kcal, 50.5 g protein, 310.1 g carbohydrate, 64.3 g fat, 12.8 g fiber. Day B: 1,989.7 kcal, 115.9 g protein, 247.2 g carbohydrate, 62.6 g fat, 27.1 g fiber.
• Per kilogram. 50.5 ÷ 65 = 0.78 g/kg; 115.9 ÷ 65 = 1.78 g/kg. Reference points: 0.8 × 65 = 52 g; 1.6 × 65 = 104 g.
• Shares of calories. Protein and carbohydrate grams × 4 and fat grams × 9, divided by the USDA calorie total: Day A 10.1% / 61.8% / 28.8%, Day B 23.3% / 49.7% / 28.3%. Rounded in the table.
• Atwater recount. Day A 4 × 50.5 + 4 × 310.1 + 9 × 64.3 = 2,021; Day B 4 × 115.9 + 4 × 247.2 + 9 × 62.6 = 2,016. USDA SR Legacy applies food-specific factors, so a recount rarely matches to the calorie.
• Not tested. These are composed menus, not logged meals. Nobody ate them and nothing was weighed in a kitchen. Branded foods, restaurant portions and cooking methods will move every figure, and fiber is the USDA total dietary fiber for each food.

Frequently asked questions