Protein intake calculator: how much protein per day?

Ten published figures address how much protein you need in a day. Five come from a European authority and five from the sports literature. No authority has set a figure for athletes at all. This calculator asks for your weight, your training and your goal, lines all ten up and picks no number out of them.

Counted by Bodi · Last counted 4 August 2026 · 9 sources

Work out your protein intake

This calculator needs JavaScript. With it switched off, all ten reference values are still printed further down the page, sources and all. Each row is a single multiplication. That is deliberate.

What are you doing?

This field picks no number for you. It decides which of the ten rows apply to you within the scope those rows were written for. Anything not written for your situation drops out of the overview. All ten are still written out in the text below this calculator, sources and all.

Bulking is the phase on an energy surplus. Iraki et al. put 1.6 to 2.2 grams per kilogram against it, on a surplus of roughly 10 to 20 per cent and a gain of 0.25 to 0.5 per cent of body weight a week.⁸

Cutting is the phase on an energy deficit. There is one row for that, and it counts per kilogram of fat-free mass rather than per kilogram of body weight.⁹ Fill in your body fat as well or it cannot join in.

Maintaining is everything else. Note that bulking and cutting only change anything when resistance training is in the mix, both at once included. The two rows this calculator carries for a surplus and for a deficit were both established in people doing resistance training.⁸⁹ Outside that group they say nothing, and what remains is what there is.

What do you train?

Resistance training switches on the rows derived from resistance training: Morton et al.'s break point and, depending on your goal, the bulking or the cutting row.⁴⁸⁹

Distance training switches on one row of its own. Kato et al. measured the crossover point on a day with a 20 km run, after two days of 10 and 5 km: 1.65 to 1.83 grams per kilogram.⁷ That is one heavy training day in six men, and not an average week.

Hybrid switches both sets on, each inside its own scope. On a deficit that produces something worth seeing: the sources for resistance training in a deficit and those for a day of distance running do not overlap, and then the page says they disagree.⁷⁹

Hardly any drops all five sports rows. What remains is what there is: the European reference value for adults.¹² That is not a shortcoming of this calculator but of the literature, and it is exactly what ought to be visible.

Age

Exactly two rows pay any attention to age. EFSA sets the requirement per kilogram equal across every adult age, and the Dutch Health Council follows it.¹²

The DACH countries and the Nordic countries do not: they put a higher figure against people over 65, 1.0 and 1.1 to 1.3 grams per kilogram respectively.² Those two rows appear from 65 on, because below that they were not written for you.

The Dutch Health Council, incidentally, declined to adopt those higher figures: fewer than 40 per cent of the studies it examined found a favourable effect of extra protein on fat-free mass.²

years

Body weight

All of you, not your fat-free mass. That is how nine of the ten rows are written down. EFSA states outright that the requirement per kilogram is the same for men and women and the same at every body weight.¹ Which is why this calculator does not ask your sex. It does ask your age, because two other rows do pay attention to it.

And there, immediately, is the boundary. The Dutch Health Council writes that these values apply to healthy people at a healthy weight with average physical activity, and that its report does not address people far outside a healthy weight and does not address people who train to extremes.² So a calculator that multiplies 0.83 by whatever weight you type is working outside the range the number was written for. Including this one. Now you know.

Energy intake

Leave this empty if you do not know it. One row then drops out, which is more honest than a row resting on a guess. If you do not know it, the TDEE calculator gives you a band.

Why a second unit turns up here. Not every European body counts per kilogram. The Nordic and Baltic countries express their recommendation as a share of your energy intake: 10 to 20 per cent of it from protein.³ The two units do not map onto each other, and at an ordinary intake the Nordic method lands considerably higher than EFSA's 0.83.

We convert using 17 kilojoules per gram of protein, because that is the factor the same Nordic review uses itself.³ The 4 kcal per gram printed on your tub gives an answer about one and a half per cent higher.

kcal

Body fat

Leave this empty if you have never had it measured. The bottom row then drops out, and it is the only one that needs it. Pick "cutting" above and that row is the one written for your situation.

Why one row asks for this. Helms et al. count per kilogram of fat-free mass rather than per kilogram of body weight, and that is their entire argument: protein recommendations were established in people at ordinary or high body fat, and per kilogram of total weight such a figure comes out too low for someone lean.⁹ Fat-free mass is your weight minus your fat, and we do nothing cleverer to it than that.

Above 35 per cent that row stays empty. Their review took in only people below 23 per cent body fat for men and below 35 for women.⁹ We do not ask your sex, so we hold mechanically to the wider of those two bounds. If you are a man between 23 and 35 per cent you will therefore get a number while sitting outside the group that was studied.

What has been written down for your situation

Reference value	Published	Grams per day
European reference values
EFSA, average requirement¹	0.66 g/kg
EFSA and Dutch Health Council¹²	0.83 g/kg
DACH countries, from age 65²	1.0 g/kg
Nordic countries, from age 65²	1.1 to 1.3 g/kg
Nordic and Baltic countries³	10 to 20 E%
Sports literature
Kato et al., day of distance running⁷	1.65 to 1.83 g/kg
Iraki et al., building phase⁸	1.6 to 2.2 g/kg
ISSN, floor for people who exercise⁵	1.4 to 2.0 g/kg
Morton et al., break point, resistance⁴	1.03 to 2.20 g/kg
Helms et al., in an energy deficit⁹	2.3 to 3.1 g/kg fat-free

Where these rows come from, and what is not in them

of the ten rows were written for it.

From their lowest bound to their highest they run from . That spread is not in you. It is in them.

The European reference value for adults in general comes out at grams a day at this weight.¹² It was not written for athletes, and the Dutch Health Council says so in as many words.²

Everything that applies to you runs from together, and the highest is times the lowest. That is not measurement error. These rows are not attempts at the same number: one is the lowest intake that keeps you in nitrogen balance, another is a floor for people who train. Anyone who boils them down to a single figure has decided on your behalf which question you were asking.

The rows from the sports literature come from nobody official, and that is not an omission on our part. EFSA and the Dutch Health Council have set no protein reference value for athletes and both write that their advice does not cover them.¹² What is left is sports science, where the interests sit differently: the distance-running study was paid for by a manufacturer of amino acids with three of its four authors on the payroll,⁷ and of the 23 authors of the sports nutrition position stand, 22 declare a tie to the supplement industry.⁵

Two of the ten rows also lean on one analysis: the 2.2 at the top of the building phase is lifted straight out of Morton et al.'s interval, and the authors say as much.⁸ That interval is a bend in a regression those authors print as not significant, p is 0.079, running from 1.03 to 2.20.⁴

Nothing here gets divided across meals. The Dutch Health Council sets no reference value for how intake is spread through the day, because there is too little research.² And the 25 gram per sitting ceiling that other calculators divide by has been measured: 100 grams in one go put 53 grams of amino acids into the bloodstream over twelve hours, against 16 grams from 25, and that curve had not levelled off when the clock ran out.⁶

All ten reference values are written out below, with the source attached. Each row is a single multiplication. Check it yourself.

Short answer

How much protein do you need per day?: That depends entirely on which committee you ask. Ten published rows land anywhere between 53 and 176 grams a day at 80 kg. The European figure is 0.83 g per kilogram, which is 66 grams.¹²
How do you calculate your protein intake?: Your body weight, times a number somebody else wrote down. That is the whole calculation: 0.83 times 80 kg is 66 grams.¹ One row declines to work in kilograms and counts per cent of energy intake instead, which at 2,500 kcal gives 62 to 123 grams.³
How much protein if you train?: No European authority has a figure for that at all. EFSA and the Dutch Health Council both write that their advice does not cover it.¹² So five of the ten rows come from the sports literature. On a day of distance running the figure came out at 1.65 to 1.83 g per kilogram, measured in six men.⁷
How much protein for resistance training?: The famous figure is 1.62 g per kilogram, and it is not significant. Morton et al. print p is 0.079 beside it, with an interval running from 1.03 to 2.20.⁴ At 80 kg that is 82 to 176 grams.

How we count on the rest of this site

How much protein per day do you need?

Nobody has a number for you. There are ten of them, they run from 53 to 176 grams a day at 80 kg, and the top one is 3.3 times the bottom one.

Hand the thing a body fat percentage and the ceiling climbs again. At 15 per cent it reaches 211 grams, and the top is then 4 times the bottom. One of the ten rows refuses to count all of you.

The official European figure is the short one. EFSA set a recommended amount of 0.83 g per kilogram per day in 2012, and the Dutch Health Council adopted it for all adults in 2021.¹² At 80 kg: 66 grams.

Then EFSA quietly takes the wind out of it. Measured average intakes across Europe already sit at or above 0.83 g/kg.¹ Among Dutch men of 65 and over, the top 5 per cent eat 1.7 g per kilogram per day or more.¹

How much protein for your training and goal?

Your training does most of the work here. Your goal counts the moment resistance training is in the mix, and that includes doing both at once. Three fields sit at the top: what you are doing, what you train, and how old you are.

None of the three picks a number for you. They switch rows on and off according to the scope each source writes down for itself.

Here is what that leaves at 80 kg. The third column is the stretch every surviving row contains. The fourth runs from the lowest bound to the highest.

What you enter	Rows	Where they overlap	Lowest to highest
Hardly any training, any goal	none	none	none
Resistance training, maintaining	2⁴⁵	112 to 160 g	82 to 176 g
Resistance training, bulking	3⁴⁵⁸	128 to 160 g	82 to 176 g
Resistance training, cutting, at 15 per cent body fat	3⁴⁵⁹	156 to 160 g	82 to 211 g
Distance training, any goal	2⁵⁷	132 to 146 g	112 to 160 g
Both, maintaining	3⁴⁵⁷	132 to 146 g	82 to 176 g
Both, bulking	4⁴⁵⁷⁸	132 to 146 g	82 to 176 g
Both, cutting, at 15 per cent body fat	4⁴⁵⁷⁹	none	82 to 211 g

Hardly any training produces no sports row whatsoever. What stands instead is the European reference value, and at 80 kg that is 66 grams.¹² These ten rows have nothing else for you.

Bulking and cutting only move anything once resistance training is in the mix. Two of the ten rows address a surplus or a deficit, and both were written down in people who lift.⁸⁹ Pick distance training on its own, and cutting gives you precisely what maintaining does: 132 to 146 grams.

Nothing else in this list was written for that combination. The gap sits in the sources and not in the arithmetic. We are not going to plug it with a row meant for somebody else.

Both at once: two sources that flatly disagree

Choose both and each set applies, inside its own scope. Morton et al. wrote for resistance training, Kato et al. for a day of distance running.⁴⁷ Do both and you sit under each. That is not a compromise. It is addition.

You can hand it maintaining, watch three sources overlap at 132 to 146 grams, hand it bulking, watch a fourth join in without budging it, and then hand it cutting.⁴⁵⁷⁸ The overlap vanishes.

Start with bulking, because that fourth row could never have moved anything. Iraki et al. print 128 to 176 grams at 80 kg, which swallows Kato et al.’s 132 to 146 whole.⁷⁸ The narrower row sets both bounds, with or without bulking.

Cutting is another matter: Helms et al. start at 156 grams and Kato et al. finish at 146.⁷⁹ No figure satisfies both, so the calculator prints their full range instead: 82 to 211 grams.

That is two sources disagreeing, and not the literature as a whole. None of these ten rows was written for resistance training and distance training together in an energy deficit.⁷⁹ The calculator does not average the gap away. Averaging would print a figure nobody has published.

Which rows drop out, and why

Pick cutting without filling in your body fat and the one row written for that situation drops straight out.⁹ Cutting then reads exactly like maintaining, and the calculator says so where you can see it. Above 35 per cent body fat that row falls outside its own range, with the same result.

The calculator keeps only what falls inside its own stated scope for you. Do not train and the five sports rows go. Type an age under 65 and the two rows for people over 65 go with them.² Exactly 65 counts. 64 does not.

All ten are written out below regardless, figure and source attached, because you are allowed to see what was written for somebody else.

Protein intake calculator: one multiplication

That is genuinely all it is. Your body weight, times the number in a reference value, and out comes a figure in grams.

Eight of the ten rows work that way. 0.83 multiplied by 80 kg gives 66 grams, and 2.20 gets you 176.¹⁴ The arithmetic is not the difficult bit. The difficult bit is which row you print, because anybody handing you a single figure has decided on your behalf what you were asking.

The other two want something extra. The Nordic and Baltic countries write their recommendation as a share of your energy intake, so that one wants kilocalories off you.³ Helms et al. count per kilogram of fat-free mass, which is your weight minus your fat, so that row wants a body fat percentage.⁹

The ten rows at 80 kg, side by side

Here they all are, with the source attached to each. The right-hand column is one multiplication, so go ahead and check it.

First the five European reference values. Those were written for adults in general.

Row	As published	At 80 kg
EFSA 2012, average requirement¹	0.66 g/kg	53 g
EFSA 2012 and Dutch Health Council 2021, recommended amount¹²	0.83 g/kg	66 g
DACH countries 2017, from age 65²	1.0 g/kg	80 g
Nordic countries 2012, from age 65, back-calculated²	1.1 to 1.3 g/kg	88 to 104 g
Nordic and Baltic countries 2023, recommended range³	10 to 20 E%	62 to 123 g

The bottom row hangs on energy intake rather than on body weight, and it is printed here at 2,500 kcal a day. E% is the share of your daily kilocalories that arrives as protein.³

Then the five rows out of the sports literature. Not one of them was set by an authority, which is exactly why they sit in a table of their own.

Row	As published	At 80 kg
Kato et al. 2016, on a day of distance running⁷	1.65 to 1.83 g/kg	132 to 146 g
Iraki et al. 2019, off-season on an energy surplus⁸	1.6 to 2.2 g/kg	128 to 176 g
ISSN 2017, floor for people who exercise⁵	1.4 to 2.0 g/kg	112 to 160 g
Morton et al. 2018, break point under resistance training⁴	1.03 to 2.20 g/kg	82 to 176 g
Helms et al. 2014, in an energy deficit⁹	2.3 to 3.1 g/kg fat-free mass	156 to 211 g

That last row is printed here at 15 per cent body fat. At 80 kg that leaves 68 kg of fat-free mass, which is what goes into the sum.⁹ Bulking and cutting do now sit in a dropdown, and it takes nothing off this page. Inside the calculator, whatever was written for somebody else disappears. Here it stays.

You can buy a tub with a scoop, a chart on the back and a figure set in a typeface chosen to look scientific, and none of it says which of these ten rows it answers. That is the whole difficulty.

Two of the rows we know second hand. The Nordic countries advised 15 to 20 E% for people over 65 in 2012.² The Health Council worked that back to 1.1 g per kilogram at average physical activity and 1.3 for a mostly seated life.² That back-calculation is the committee’s arithmetic, and we have not inspected the two original documents ourselves.²

The Health Council printed those two rows and then declined to adopt them. For adults over 60, fewer than 40 per cent of the studies it examined found a favourable effect of extra protein on fat-free mass.² So the reference value stayed at 0.83.² Fat-free mass is everything you are made of apart from the fat.

Change the inputs and the gap gets wider. At 50 kg and 3,000 kcal the band runs from 33 to 148 grams, and the top is 4.5 times the bottom. The energy row takes over the top end there, at 148 grams against the 110 the sports rows stall at.³⁴⁸

Why 0.83 sits above the average requirement

Because 0.83 is not an average. The average requirement is 0.66 g per kilogram per day, or 53 grams at 80 kg.¹ 0.83 is the point at which 97.5 per cent of healthy adults have plenty, and it was put there deliberately.¹

A figure built to cover 97.5 per cent therefore sits above what almost everybody needs. Most people’s own requirement is below it.¹ The spread is not vast either: EFSA works with a variation of 12 per cent around the mean.¹

Underneath both figures sits a piece of bookkeeping called nitrogen balance. Measure the nitrogen arriving through food, and the amount leaving in urine and faeces. The lowest intake at which the two match is the requirement.¹ The median of that is 105 mg nitrogen per kilogram per day, and 0.83 corresponds to 133 mg.¹

The method has weak spots, and the panel says so itself.¹ One figure shows how much is still moving. The efficiency with which dietary protein is used for maintenance came out at 47 per cent.¹ The literature had been assuming 70 per cent.¹

Two European units that refuse to line up

EFSA counts per kilogram of body weight. The Nordic and Baltic countries count in per cent of energy intake, and the two do not map onto each other.¹³

Put them side by side at 80 kg and 2,500 kcal a day.

Row	Arithmetic	Result
EFSA 2012 and Dutch Health Council 2021¹²	0.83 × 80 kg	66 g
Nordic and Baltic countries 2023, at 15 E%³	15 per cent of 2,500 kcal, at 17 kJ per gram	92 g

That is 39.4 per cent apart, for one person on one day. The 15 per cent is our worked example and not their planning value. Their 2023 text gives a range of 10 to 20 per cent, and 15 sits in the middle of it.³

We convert at 17 kilojoules per gram, because that is the factor the Nordic review uses itself.³ There is a sting in the tail as well: if energy intake drops below 8 megajoules a day, the protein share has to rise to match.³ A percentage of not very much is not very much.

Why the reference values disagree so widely

Because they are not answers to the same question. Six different questions are sitting in those two tables, and there is the entire spread.

What keeps an adult in balance? That is EFSA’s question and the Health Council’s, answered at 0.66 as an average requirement and 0.83 as a reference value.¹²

What protein share goes into a diet planned for a population? That is the Nordic and Baltic question, answered in per cent of energy.³ A range for composing diets is not one person’s requirement, and the review does not present it as one.³

Where do the gains stop under resistance training? That is the question Morton et al. asked, and the answer is a bend in a line drawn through their data.⁴

At what intake does the body stop burning amino acids? That is the question Kato et al. asked, on a day of distance running, and their answer is 1.65 g per kilogram.⁷ Amino acids are the parts protein is built from.

How much protein holds on to fat-free mass in an energy deficit? That is the question Helms et al. asked, and their answer is 2.3 to 3.1 g per kilogram of fat-free mass.⁹

What do we recommend? That is the question the ISSN asked and Iraki et al. asked, and the answers are 1.4 to 2.0 and 1.6 to 2.2 g per kilogram.⁵⁸

Only the top two questions were answered by anybody official.¹²³ The other four come out of the sports literature. Who wrote them down, and on what, is below.

How much protein if you train?

No European authority has a figure for it. There is a European reference value for protein, worked out by a panel, adopted by a national council and printed in a report with a publication number, and neither document says a word about athletes.¹² That silence is the whole problem.

EFSA assessed fat-free mass and body weight as outcomes and found the data insufficient to base a reference value on.¹ The Dutch Health Council adds that its report does not address people who train to extremes.²

What is left is sports science, which is the second table: five rows, and not one was ever set. Who wrote each one, and with what attached, is printed alongside it.

Pick distance training on its own and two of those five apply, whatever your goal.⁵⁷ At 80 kg the overlap is 132 to 146 grams. Together they run from 112 to 160.

On a day of distance running: 1.65 g per kilogram

Underneath that figure are six men. Kato et al. recruited nine, but two dropped out and one managed a single trial day.⁷ Between them those six produced 34 trial days, at seven protein levels from 0.2 to 2.8 g per kilogram per day.⁷

Kato et al. used the indicator amino acid oxidation method rather than nitrogen balance. From the carbon breathed out they derive the intake at which the body stops burning amino acids.⁷ The crossover point landed at 1.65 g per kilogram per day and the recommended intake at 1.83.⁷ At 80 kg that is 132 to 146 grams.

A second calculation on the same measurements gave 1.53 and 1.70, and both sit above the US recommended daily allowance of 0.8 g per kilogram.⁷ Both also sit above the 1.2 to 1.4 standing for distance athletes at the time.⁷ The amount they found works out at roughly 12 per cent of energy intake in this study.⁷

What it was measured on is in the paper too. A day with a 20 km run, on top of two days of 10 and 5.⁷ That is a heavy training day and not an average week, and the authors say so themselves.⁷ The protein was also not food but a free amino acid mixture, which behaves differently.⁷

Who paid for it is stated openly. Three of the four authors are employed by Ajinomoto, a manufacturer of amino acids.⁷ That company funded the study, and the mixture they were fed came from there as well.⁷

The 1.4 floor comes from a membership society

The ISSN holds to 1.4 to 2.0 g per kilogram per day for people who exercise, explicitly as a floor and not as a ceiling.⁵ At 80 kg that is 112 to 160 grams.

It is a position stand and not a systematic review with a fixed search. There is no search strategy in it, no selection criterion, no weighting of the studies, and the society’s own research committee reviewed the thing.⁵

Then the authorship. Of the 23, some 22 declare a tie to the supplement industry, and exactly one declares no interest whatsoever.⁵ Two are employed by a supplement manufacturer, and the last author is co-founder and chief executive of the society.⁵ The society states that brands selling protein supplements help fund it.⁵

None of which makes 1.4 untrue. It does make it worth knowing who wrote it down.

How much protein for resistance training?

The figure you meet everywhere is 1.62 g per kilogram per day. It comes from one regression, in one paper, whose own authors describe the analysis as not significant: the pattern could be chance.⁴

In the calculator, resistance training on its own with no surplus or deficit sits on two rows.⁴⁵ The society’s range fits entirely inside Morton et al.’s. So the overlap is one row, 112 to 160 grams, and the full range is the other, 82 to 176.

Morton et al. added up 49 studies from 17 countries, with 1,863 participants averaging 35 years old.⁴ The figures in brackets below are the margin the authors print beside each result.

Extra protein alongside resistance training gave 0.30 kg more fat-free mass (0.09 to 0.52).⁴ On the heaviest weight for a single repetition it was 2.49 kg (0.64 to 4.33).⁴

That 0.30 kg is the gain across the entire run of those studies. In people already training the difference was 1.05 kg (0.61 to 1.50).⁴ In people with no experience, 0.15 kg, and not significant.⁴ The effect shrank by 0.01 kg for every year of age.⁴

1.62 is a bend, not a reference value

The authors went looking for the point where the line changes direction. Two straight lines with a bend between them, and the data decide where it goes.⁴ It landed on 1.62 g per kilogram per day.

Now the small print. The interval runs from 1.03 to 2.20 and p is 0.079, so the analysis was not significant.⁴ The authors say so themselves, and printed the break point anyway.⁴

That interval is more than twice as wide as the figure suggests. So at 80 kg there is no single number at all, only 82 to 176 grams.

The floor under it is narrower than it looks, too. The break point rests on 42 study groups with 723 participants, at intakes from 0.9 to 2.4 g per kilogram per day.⁴ Outside that range the data say nothing.

One sentence from that paper rarely survives the quoting. The participants were already eating around 1.4 g per kilogram per day at baseline, and the average extra dose was about 35 grams a day.⁴ The comparison therefore started well above the European reference value.

The provenance belongs on the page as well. The last author declares research funding, travel expenses and speaking fees from the US National Dairy Council.⁴ That body funded trials included in this analysis, and the paper says so itself.⁴

How much protein when bulking?

1.6 to 2.2 g per kilogram per day, which at 80 kg is 128 to 176 grams.⁸ Iraki et al. wrote it for the off-season of natural bodybuilders, on an energy surplus of roughly 10 to 20 per cent.⁸ A weekly gain of 0.25 to 0.5 per cent of body weight goes with it.⁸

Pick bulking and that row joins in wherever resistance training is involved. On resistance training alone, three apply, and where they overlap sits 128 to 160 grams.⁴⁵⁸ The lower bound moves up. The upper one does not.

Add distance training as well and that row stops mattering: the overlap stays where maintaining left it, at 132 to 146 grams.⁷⁸

It is a narrative review. There is no search strategy in it and no weighting of the studies, exactly as with the position stand above.⁸

Two of the ten rows also lean on one analysis. The 2.2 at the top is lifted straight out of Morton et al.’s interval, and Iraki et al. say as much.⁸ That interval is the bend that was not significant.⁴

Their argument against the other unit is worth having, though. A recommendation in energy per cent works out badly for a light person with a high energy requirement.⁸ They land far above what is needed, and a recommendation like that can squeeze out room for carbohydrate and fat.⁸

What the group actually eats is in there as well: men go up to 4.3 g per kilogram per day and women to 2.8.⁸ That is well above what the article itself recommends.⁸

The authors declare no interests and report no external funding, though the first two do run nutrition consultancies of their own.⁸

How much protein when cutting?

One row here refuses to count the fat you are carrying. Helms et al. land on 2.3 to 3.1 g per kilogram of fat-free mass per day.⁹ At 80 kg and 15 per cent body fat your fat-free mass is 68 kg, and the row prints 156 to 211 grams.

Pick cutting on resistance training alone and three rows apply. Where all three overlap, very little survives: 156 to 160 grams.⁴⁵⁹ Their full range is 82 to 211 grams, and that is the honest picture.

Add distance training and the overlap stops existing altogether: this row starts at 156 grams and Kato et al. finish at 146.⁷⁹ What is left is that 82 to 211.

Their argument is about the arithmetic itself: protein recommendations were established in people at ordinary or high body fat.⁹ Per kilogram of total weight such a figure then comes out too low for someone lean.⁹

It is a systematic review with a fixed search, running to July 2013, over six studies with thirteen groups between them.⁹ Well-matched ones existed at only four protein levels: 0.8, 1.0, 1.6 and 2.3 g per kilogram of body weight.⁹

In all thirteen groups the body fat percentage fell, by 0.5 to 6.6 percentage points.⁹ In nine of the thirteen the fat-free mass fell with it, by 0.3 to 2.7 kg.⁹ Within the range you move up the leaner the person and the larger the deficit, the authors add.⁹

There is a counter-finding in there too. In one study outside the analysis, the people on 1.6 g per kilogram held on to slightly more fat-free mass than those on 2.4.⁹ That difference was not significant, and the two groups ate very different amounts of carbohydrate.⁹

The authors name something that may weigh heavier than the protein choice.⁹ A slower rate: about half a kilogram, or 0.7 per cent of body weight, a week.⁹

A declaration of interests is missing. The text we read names no interests and no funder.⁹ That is not evidence there were none.

Why that row goes blank above 35 per cent body fat

Because the review had nobody up there. Helms et al. took in only adults of 18 and over who were in an energy deficit and had been resistance training for at least six months.⁹ For body fat they held to two bounds: men up to and including 23 per cent, women up to and including 35.⁹

We do not ask your sex. EFSA states outright that the requirement per kilogram is the same for men and women.¹ So we hold mechanically to the wider of the two Helms bounds. If you are a man between 23 and 35 per cent you get a number anyway, from outside the group that was studied.

At that bound of 35 per cent, 80 kg gives 120 to 161 grams.⁹ The top of that, 161 grams, sits below Morton et al.’s 176.⁴ The highest of the ten is not always the same row.

It works just as hard the other way. Put the body fat field at its lower limit of 1 per cent and out come 182 to 246 grams.⁹ The whole band then runs from 53 to 246 grams, and the top is 4.6 times the bottom.

Does protein have to be spread across the day?

There is no reference value for it. The Dutch Health Council declines to set one, because there is too little research.²

You can take a daily total, chop it into four, cap every sitting at 25 grams and print the result as a tidy grid, and the assumption underneath the whole exercise was measured in Maastricht between 2019 and 2020. It did not hold.

That 25 grams had counted, until then, as the portion above which nothing more would be added.⁶ Trommelen et al. went and checked.

36 healthy, recreationally active young men were given 0, 25 or 100 grams of milk protein once after a bout of resistance exercise, twelve per group.⁶ Over twelve hours the researchers took thirteen blood samples and four pieces of muscle tissue from the thigh.⁶

Of the 100 grams, 53 grams of amino acids reached the bloodstream within twelve hours; of the 25 grams it was 16.⁶ At 25 grams, 66 per cent of what was eaten had been released after twelve hours; at 100 grams, 53 per cent, and it was still going.⁶

Incorporation into muscle tissue was higher at 100 grams than at 25.⁶ The difference was larger in the late phase, four to twelve hours in, than during the first stretch.⁶ Burning off barely rose at all.⁶

What it does not say: 36 healthy young men, once, with the measurement stopping at twelve hours while outcomes were still running.⁶ It is one portion and not a daily total.⁶ So it hands this page no number whatsoever. It only takes an assumption away.

The protein on a label is itself a conversion

You are counting towards a figure that was estimated in the first place. Somewhere between a bean and a bottle of milk, a laboratory measures nitrogen, multiplies it by 6.25 and prints the answer on the packaging as protein.³ Nobody weighed the protein.

That factor is not right for every food. It runs from 5.70 for soya to 6.32 for milk, which can put the real content 15 to 20 per cent out.³

So you are working towards a reference value that varies 12 per cent between people.¹ And you are counting it in grams that can be 15 to 20 per cent out per product.³ Precision is not on the menu here.

What this figure is not

It is not a measurement. It is your body weight times a number somebody else wrote down.

It also applies to a defined group. The reference values were written for people in good health at a healthy weight, with average physical activity.² The Health Council adds that its report does not address people far outside a healthy weight, and not those who train to extremes.²

The five rows from the sports literature are not reference values. The break point under resistance training rests on 42 study groups.⁴ The range in an energy deficit comes from thirteen.⁹ The crossover point on a day of distance running comes from six men.⁷ All three were measured in different people.

The off-season row is one example of that. It was written for natural bodybuilders on an energy surplus, and not for everybody who trains.⁸

There is no upper limit in any of it either. EFSA could not derive one, and that is a shortage of data rather than a licence.¹ The panel calls twice the recommended amount safe for adults.¹ Intakes of three to four times it have been observed, showing neither harm nor benefit that could be demonstrated.¹

And it is not dietary advice.

Sources

Open a source and you see exactly what was used from that study, which sentence on this page it supports, and what it explicitly does not support.

1EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific Opinion on Dietary Reference Values for protein. EFSA Journal 2012;10(2):2557. DOI 10.2903/j.efsa.2012.2557. Open access.opinion of a European scientific authority, resting on a meta-analysis of nitrogen balance research · full text read, 66 pages, PDF retrieved from a public copy at cambridgefoodscience.com and extracted with pdftotext; the publisher's version at Wiley returned HTTP 403 on 2026-08-04 · read 2026-08-04

Design: Not an experiment of its own. The adult value comes from the 2003 meta-analysis of nitrogen balance research by Rand et al., which EFSA adopts following the WHO. Nitrogen balance works like this: measure how much nitrogen goes in through food and how much leaves through urine and faeces, then find the lowest intake at which the two are equal, at energy balance. The 97.5th percentile was then taken from the spread of that requirement across the population. For children and for pregnant and lactating women a factorial sum was used as well: maintenance plus the laying down of new tissue.
Studied in: 235 healthy adults from 19 studies in the meta-analysis the adult value rests on. The resulting reference value applies, according to the Panel, to adults of all ages, to both sexes and to all body weights, provided they are healthy and in energy balance.
Compared with: The earlier reference values of WHO/FAO/UNU (2007) and of the US Institute of Medicine, plus the values individual European countries and regions used at the time. The opinion tabulates them side by side: average requirements from 0.60 to 0.66 and recommended amounts from 0.75 to 0.83 g/kg/d.
Dose and duration: Not applicable as a dose. Intake is itself the exposure here, expressed in grams of protein per kilogram of body weight per day.
What was measured: The lowest protein intake at which an adult stays in nitrogen equilibrium at energy balance, expressed per kilogram of body weight per day.
What was found: The average requirement is 0.66 g/kg/d, that is, the median of 105 mg nitrogen per kilogram per day. The 97.5th percentile of that same distribution is 133 mg nitrogen, or 0.83 g/kg/d, and that is the recommended amount. The coefficient of variation between people is 12 per cent. The efficiency with which dietary protein is used for maintenance came out at 47 per cent, where the literature had previously assumed 70 per cent. For adults of 60 and over the requirement was set equal to that of younger adults, because there were insufficient data to establish anything else. No upper level can be derived. Twice the recommended amount the Panel calls safe for adults, and intakes of three to four times it have been observed with no demonstrable harm and no demonstrable benefit. In Europe measured mean adult intakes sit at or above 0.83 g/kg/d; among Dutch men aged 65 and over the 95th percentile is 1.7 g/kg/d.
Supports on this page: The two numbers the table in this calculator opens with, 0.66 and 0.83 grams per kilogram per day, and the fact that they apply per kilogram of body weight to both sexes and to all body weights. Above all it carries the point that 0.83 is not an average but a percentile: it was deliberately set high enough to be ample for 97.5 per cent of healthy adults.
Explicitly does not support: This is a reference value for maintenance, not for training. The Panel explicitly assessed outcomes such as fat-free mass, body weight and bone strength and concluded the data were insufficient to base a reference value on; it therefore makes no statement whatsoever about what a higher intake delivers. The derivation rests on nitrogen balance, and the Panel notes itself that the method has weak spots. Nor is 0.83 a target for an individual: by construction it covers almost everyone's requirement, so most people's own requirement sits below it. On athletes this opinion says nothing. Nor on an upper level, because none could be derived, and 'no upper level set' is a shortage of data rather than a licence. The opinion says nothing about a supplement and nothing about a brand.
Interests and funding: EFSA is an agency of the European Union. The Panel consists of external experts with publicly declared interests. No commercial funder named.

https://doi.org/10.2903/j.efsa.2012.2557

2Gezondheidsraad. Voedingsnormen voor eiwitten: referentiewaarden voor de inname van eiwitten. Den Haag: Gezondheidsraad, 2 maart 2021, publicatienr. 2021/10. Open access.advisory report of a national scientific council · full text read, 40 pages, PDF retrieved from gezondheidsraad.nl and extracted with pdftotext · read 2026-08-04

Design: The committee set the existing reference values of five bodies side by side: EFSA (2012), the WHO with FAO and UNU (2007), the Dutch values of 2001, the Nordic countries (2012) and the DACH countries (2017). For adults under 60 it adopts EFSA's derivation. For adults from 60 it found the EFSA report insufficient, because many new intervention trials have appeared since 2012, and had an additional systematic literature review carried out. It also searched again for recent nitrogen balance studies in older adults; that turned up nothing beyond what EFSA already had.
Studied in: Healthy people at a healthy weight who need no medical care involving special dietary measures, and with average physical activity. The values are worked out per age group into grams per day using reference weights from Dutch population research: men aged 18 to 29 at 75.6 kg, women of the same age at 64.6 kg.
Compared with: The reference values of EFSA, of the WHO with FAO and UNU, of the Nordic countries and of the DACH countries, and the Dutch values of 2001.
Dose and duration: Not applicable as a dose. Intake is itself the exposure, in grams of protein per kilogram of body weight per day.
What was measured: For the reference value itself: nitrogen equilibrium. For the additional question about older adults: fat-free mass, muscle strength, physical functioning and a range of other measures, each judged on a scale of six possible conclusions.
What was found: The committee adopts the EFSA values: an average requirement of 0.66 g/kg/d and a recommended amount of 0.83 g/kg/d, for adults of every age. For older adults it sets no higher value, because fewer than 40 per cent of the studies examined found a favourable effect of more protein on fat-free mass, and the same holds for more protein combined with physical activity in relation to muscle strength; the majority of studies therefore show no effect. It does print the other European values alongside, and those are higher. The DACH countries set an adequate intake of 1.0 g/kg/d for people over 65 in 2017. The Nordic countries advised 15 to 20 energy per cent for people over 65 in 2012, and the committee works that back to 1.1 g/kg/d for people with average physical activity and 1.3 g/kg/d for people who mostly sit. The committee does not rule out that frail or undernourished older people benefit from more, but those groups fall outside this report.
Supports on this page: The two rows in the table that come from somewhere other than EFSA: 1.0 g/kg/d from the DACH countries and 1.1 to 1.3 g/kg/d as a back-calculation of the Nordic value, both for people over 65. Above all it carries the boundary of this calculator: the values apply to healthy people at a healthy weight with average physical activity, and the report states in as many words that it does not address people with obesity and does not address people who train to extremes. It also carries the fact that no reference value exists for how intake is spread across the day, because there is too little research.
Explicitly does not support: The DACH value and the Nordic value appear in this report as a description of what others have set, not as the Health Council's own. We have not inspected the 2017 DACH reference values or the 2012 Nordic recommendation ourselves; we know them as this report renders them, which makes them second-hand. The back-calculation from energy per cent to 1.1 and 1.3 g/kg/d is moreover the committee's arithmetic and not the Nordic committee's. Beyond that: this report positively declines to adopt the higher values for the Netherlands. It says nothing about athletes, nothing about a supplement and nothing about a brand.
Interests and funding: The Health Council of the Netherlands is an independent advisory body established by law that advises government and parliament. This report was issued to the State Secretary for Health, Welfare and Sport. No commercial funder named.

https://www.gezondheidsraad.nl/documenten/2021/03/02/voedingsnormen-voor-eiwitten

3Geirsdottir OG, Pajari AM. Protein: a scoping review for Nordic Nutrition Recommendations 2023. Food & Nutrition Research 2023;67:10261. DOI 10.29219/fnr.v67.10261. Open access.scoping review underpinning a regional set of dietary reference values · full text read via the copy at PubMed Central (PMC10770649); the main NNR2023 report itself, at pub.norden.org, returned HTTP 403 on 2026-08-04 and was therefore not read · read 2026-08-04

Design: Scoping review: not an experiment of its own and not a meta-analysis, but an ordered survey of the existing systematic reviews and reference values, intended as the basis for setting and revising the Nordic and Baltic dietary reference values. The recommendation here is expressed as a share of daily energy intake rather than per kilogram of body weight.
Studied in: The populations of the Nordic and Baltic countries, with each country's national dietary survey as the basis. Measured adult intakes run from 15 energy per cent among women in Denmark to 18 energy per cent in Iceland, Norway and Finland; across the region 15 to 19 energy per cent. Among children from about one year old, 13 to 16 energy per cent.
Compared with: The reference values of EFSA (2012) and of the US Institute of Medicine, and the previous Nordic values of 2012.
Dose and duration: Not applicable as a dose. The exposure is ordinary daily intake, expressed as a share of energy intake.
What was measured: The recommended intake range for protein, expressed in energy per cent, plus a survey of what is known about protein intake and a range of health outcomes.
What was found: The recommended intake range is 10 to 20 per cent of daily energy intake, and current intake in the region falls inside it. The review puts the energy content of protein in a mixed diet at 17 kJ per gram. Two things are stated explicitly. First: if energy intake drops below 8 MJ a day, through less movement or through deliberate cutting, the protein share has to rise accordingly. Second, on measurement itself: protein on a label is determined by multiplying measured nitrogen by 6.25, and that factor can vary per food from 5.70 for soya to 6.32 for milk, which can produce a 15 to 20 per cent error in the actual protein content. The 2012 recommendation for food planning in older adults was 18 energy per cent, corresponding to roughly 1.2 g protein per kilogram per day.
Supports on this page: The row in the table that works in energy per cent, 10 to 20 per cent of daily energy intake, and the conversion factor of 17 kJ per gram with which this calculator turns that row into grams. It therefore also carries the point that two European bodies express the same requirement in two units that do not map onto each other. It further carries the fact that a protein figure on a label is a conversion from nitrogen, with an error margin of its own of 15 to 20 per cent.
Explicitly does not support: A recommended intake range for composing a diet is not the same thing as one person's requirement, and this review does not present it as one. The range of 10 to 20 energy per cent is meant for planning diets for a population. The planning value of 15 energy per cent that circulates widely does not appear in this text; it would be in the main NNR2023 report, which we were unable to read, see the comment at the top of this file. The figure of 18 energy per cent and the roughly 1.2 g/kg/d that goes with it come from the previous round in 2012 and appear here as a look back. The review says nothing about a supplement and nothing about a brand.
Interests and funding: Prepared within the NNR2023 project of the Nordic Council of Ministers, run by the Norwegian and Swedish food authorities with experts from the Nordic and Baltic countries. Open access. No commercial funder named.

https://doi.org/10.29219/fnr.v67.10261

4Morton RW, Murphy KT, McKellar SR, Schoenfeld BJ, Henselmans M, Helms E, Aragon AA, Devries MC, Banfield L, Krieger JW, Phillips SM. 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. British Journal of Sports Medicine 2018;52(6):376-384. DOI 10.1136/bjsports-2017-097608.systematic review with meta-analysis and meta-regression · full text read via the copy at PubMed Central (PMC5867436), including the results section and the declaration of interests · read 2026-08-04

Design: Systematic review of randomised trials in which extra protein alongside resistance training was compared with resistance training without it. Random-effects meta-analyses and meta-regressions on four variables set in advance. For the relationship between total protein intake and change in fat-free mass a two-phase break point analysis was used: two straight lines with a bend between them, where the location of the bend comes out of the data. Studies with three or more domains at high or unclear risk of bias were left out of the analyses.
Studied in: 49 studies from 17 countries with 1,863 participants, mean age 35 with a standard deviation of 20. Ten of those studies were in people already doing resistance training, and fourteen study groups consisted entirely of women. Publications run from 1962 to 2016. The break point analysis itself rests on a smaller subset: 42 study arms with 723 participants, at protein intakes from 0.9 to 2.4 g/kg/d.
Compared with: Extra protein against a control group doing the same resistance training without it, and alongside that the earlier meta-analyses on the same subject, which had reached differing conclusions.
Dose and duration: The extra protein dose varied per study; on average it came to about 35 grams of additional protein per day. The mean protein intake of the 1,863 participants at baseline was already about 1.4 g/kg/d.
What was measured: Change in fat-free mass, measured by DEXA, hydrodensitometry or air displacement, and change in the heaviest weight for a single repetition. Alongside those, cross-sectional area of muscle fibres and of the mid-thigh.
What was found: Extra protein alongside resistance training gave on average 0.30 kg more fat-free mass (95 per cent interval 0.09 to 0.52) and 2.49 kg more on the heaviest weight for a single repetition (0.64 to 4.33). Among people already doing resistance training that difference was 1.05 kg (0.61 to 1.50); among people without experience 0.15 kg, and that was not significant. The effect shrank with age, by 0.01 kg per year. The break point in the two-phase analysis sat at 1.62 g protein per kilogram per day, with a 95 per cent interval of 1.03 to 2.20. The authors note themselves that the analysis was not significant, p is 0.079, and that they print it anyway. They further write that given the top of the interval it may be prudent to aim for about 2.2 g/kg/d for anyone wanting to get the most out of it.
Supports on this page: The row in the table that turns the range 1.03 to 2.20 g/kg/d into grams, and the figure 1.62 that sits in the middle of it. Above all it carries how thinly that famous number is supported: it is a break point from a regression the authors themselves report as not significant, with an interval more than twice as wide as the number suggests.
Explicitly does not support: This is not a reference value and not a recommendation from an authority, but the output of a regression on group means from studies not designed for this question. The break point rests on 42 study arms, and outside the range of 0.9 to 2.4 g/kg/d the data say nothing. The outcome is fat-free mass and not body weight, and the measured differences are small: 0.30 kg across the whole duration of these studies. Among people without training experience the difference was not significant. The study says nothing about health, nothing about a brand and nothing about whether a supplement adds anything over ordinary food: it compared extra protein with no extra protein, whatever the protein came from.
Interests and funding: The last author declares research funding, travel expenses and speaking fees from the US National Dairy Council, and that body funded trials included in this analysis. The article states so itself. The remaining authors declare nothing.

https://doi.org/10.1136/bjsports-2017-097608

5Jager R, Kerksick CM, Campbell BI, e.a. International Society of Sports Nutrition Position Stand: protein and exercise. Journal of the International Society of Sports Nutrition 2017;14:20. DOI 10.1186/s12970-017-0177-8. Open access.position stand of a membership society, not of an authority · full text read at the publisher (jissn.biomedcentral.com), including the declaration of interests in full · read 2026-08-04

Design: A position stand and not a systematic review: a narrative survey of the literature in which the society sets out its position, reviewed by its own research committee and not by an independent body. No search strategy is printed, no selection criterion and no weighting of the studies included.
Studied in: Not applicable, as there is no sample of its own. The stand addresses people who exercise, and says of that group that the majority would do well to hold at least to this.
Compared with: The US recommended daily allowance and the Institute of Medicine's acceptable macronutrient distribution range, inside which the society says its own recommendation falls.
Dose and duration: 1.4 to 2.0 grams of protein per kilogram of body weight per day, explicitly presented as a floor and not as a ceiling. The paper also mentions preliminary indications above 3 g/kg/d.
What was measured: Not applicable as a measured outcome; this is a position stand and not a study. The outcomes cited in the underlying literature are fat-free mass and strength.
What was found: The stand is that the majority of exercising people should consume at minimum roughly 1.4 to 2.0 g of protein per kilogram of body weight per day. The society adds that this amount is a floor and that more may be needed for anyone cutting energy intake while wanting to hold on to fat-free mass, and that there is preliminary evidence that considerably more than 3 g/kg/d does something for body composition.
Supports on this page: The top row in the table, 1.4 to 2.0 g/kg/d, and with it the upper end of the spread across what has been published about protein.
Explicitly does not support: This is the position of a membership society and not the reference value of an authority, and it is not underpinned by a systematic search. Heavier still is who wrote it down: of the 23 authors, 22 declare a tie to the supplement industry, ranging from research funding and advisory boards to employment at a manufacturer and appearing as expert witnesses for supplement companies. The society itself states that it is funded in part by raw goods suppliers and by brands that sell protein supplements. That does not make the number untrue, but it makes its provenance something that belongs on the page. The stand further says nothing about health and nothing about any specific brand.
Interests and funding: This is the heaviest declaration of interests on this list and it runs to nearly two pages. Two authors are employees of a supplement manufacturer. The last author is co-founder and CEO of the society issuing the stand. Of the 23 authors, exactly one declares no interest at all. The society states that it is funded in part by raw goods suppliers and by brands that sell protein supplements.

https://doi.org/10.1186/s12970-017-0177-8

6Trommelen J, van Lieshout GAA, Nyakayiru J, e.a. The anabolic response to protein ingestion during recovery from exercise has no upper limit in magnitude and duration in vivo in humans. Cell Reports Medicine 2023;4(12):101324. DOI 10.1016/j.xcrm.2023.101324. Open access.randomised, double blind trial with a placebo group and isotope tracers · full text read via the copy at PubMed Central (PMC10772463), including the methods section and the declaration of interests · read 2026-08-04

Design: After a bout of resistance exercise, participants received a single drink containing 0, 25 or 100 grams of milk protein, split across three groups of twelve. The protein itself was labelled with stable isotopes, so the amino acids from that exact protein could be traced. Alongside it ran an infusion of three further labelled amino acids. Over twelve hours, thirteen blood samples and four muscle biopsies from the thigh were taken. Registered in advance with the Netherlands trial register as NL7700.
Studied in: 36 healthy, recreationally active young men aged 18 to 40 with a body mass index between 18.5 and 30, twelve per group. Excluded were smokers, people exercising fewer than one or more than three times a week, and people who do not tolerate lactose, among others. Carried out between June 2019 and March 2020 at Maastricht University.
Compared with: 100 grams of protein against 25 grams, and both against a placebo without protein. The 25 grams stands for the amount that until then counted as the portion above which nothing more would be added.
Dose and duration: A single serving of 0, 25 or 100 grams of milk protein in 800 ml of water, immediately after training, followed by twelve hours without food.
What was measured: How much of the amino acids from the ingested protein appeared in the bloodstream and was incorporated into muscle tissue, followed across twelve hours, plus how much was burnt off.
What was found: Of the 25 grams, 16 grams of amino acids reached the bloodstream in twelve hours; of the 100 grams it was 53 grams. At 25 grams, 66 per cent of the ingested protein had been released after twelve hours; at 100 grams, 53 per cent, and that curve had not yet levelled off. Incorporation into muscle tissue was higher at 100 grams than at 25, and the difference was larger in the late phase, from four to twelve hours, than in the first four. Burning off of the amino acids barely rose. The authors write that the capacity to take protein up therefore turns out to be larger than had been assumed, and that the duration of the response moves with the size of the portion.
Supports on this page: The reason this calculator does not split the daily total across meals and carries no ceiling per portion. That is not laziness: the assumption such a split rests on, that a portion above roughly 25 grams adds nothing further, was measured here and did not hold.
Explicitly does not support: 36 healthy young men, once, after a bout of resistance exercise, and the measurement stopped at twelve hours while several outcomes were still running. The authors write themselves that they do not know whether this extends to other groups, and that people with lower physical activity or poorer health may well hit a ceiling. The study concerns a single portion and not a daily total, so it supplies no number for this calculator: it only removes an assumption that other calculators do make. It says nothing about how much protein anyone needs per day, nothing about health and nothing about a brand.
Interests and funding: One of the authors is employed by the dairy company FrieslandCampina. The article states that the company had no role in funding, data collection, analysis or writing. For the first and last authors the article points to a public page listing their full career research funding. Carried out at Maastricht University, with independent monitoring by the Clinical Trial Center Maastricht.

https://doi.org/10.1016/j.xcrm.2023.101324

7Kato H, Suzuki K, Bannai M, Moore DR. Protein Requirements Are Elevated in Endurance Athletes after Exercise as Determined by the Indicator Amino Acid Oxidation Method. PLoS ONE 2016;11(6):e0157406. DOI 10.1371/journal.pone.0157406. Open access.controlled feeding study with isotope tracers, six participants · full text read via the copy at PubMed Central (PMC4913918), including the methods section, the table of seven protein levels and the declaration of interests · read 2026-08-04

Design: Not nitrogen balance but the indicator amino acid oxidation method. Across a series of separate trial days each participant was fed seven different protein amounts, randomly assigned, each after two adaptation days on fixed food and fixed training. On the trial day, a 20 km run first, then eight hourly meals of an amino acid mixture modelled on egg protein with labelled phenylalanine in it. From the carbon breathed out, the intake at which the body stops burning amino acids was derived. The crossover point of a two-phase regression is the average requirement, the top of the 95 per cent interval the recommended intake. Registered in advance, though by the authors' own account only after recruitment had begun.
Studied in: Six trained men, mean age 28 with a standard deviation of 4, mean weight 64.5 kg, self-reporting 45 to 130 km of running a week. Nine were recruited; two dropped out and one completed only a single trial day. Between them those six produced 34 trial days, at protein intakes from 0.2 to 2.8 g/kg/d.
Compared with: The US recommended daily allowance of 0.8 g/kg/d, and the 1.2 to 1.4 g/kg/d that was the standing recommendation for distance athletes at the time, which rests on nitrogen balance.
Dose and duration: Seven protein levels: 0.2, 0.9, 1.3, 1.65, 1.8, 2.3 and 2.8 g/kg/d, as a free amino acid mixture, split across eight hourly meals on a day with sufficient energy and about 9 g of carbohydrate per kilogram.
What was measured: The amount of labelled carbon in expired breath at steady state, from which the crossover point in amino acid burning follows.
What was found: The crossover point sat at 1.65 g/kg/d and the recommended intake at 1.83 g/kg/d, with an explained variance of 0.86. A second calculation on phenylalanine oxidation gave 1.53 and 1.70. Both sit above the recommended daily allowance of 0.8 g/kg/d and above the 1.2 to 1.4 g/kg/d then standing for this group. In this study the amount found works out at roughly 12 per cent of energy intake.
Supports on this page: The row in the table for distance training, 1.65 to 1.83 g/kg/d, and the fact that a method other than nitrogen balance lands on a higher figure.
Explicitly does not support: Six men, and that is the smallest sample on this list. No women. It applies to a day with a 20 km run on top of two days of 10 and 5 km, so to a heavy training day and not to an average week; the authors say so themselves. The protein was a free amino acid mixture and not food, which behaves differently. The method measures the point at which amino acid burning turns over, and that is a different measure from nitrogen equilibrium; the two figures are therefore not directly comparable with EFSA's. Of the nine participants recruited, three fell away. The study says nothing about a supplement and nothing about a brand.
Interests and funding: Heavy, and stated openly. Three of the four authors are employed by Ajinomoto, a manufacturer of amino acids, and that company funded this study. The amino acid mixture used came from the same company's North American arm. The fourth author is affiliated with the University of Toronto, where ethical approval was granted alongside that of the company itself.

https://doi.org/10.1371/journal.pone.0157406

8Iraki J, Fitschen P, Espinar S, Helms E. Nutrition Recommendations for Bodybuilders in the Off-Season: A Narrative Review. Sports 2019;7(7):154. DOI 10.3390/sports7070154. Open access.narrative review article, no systematic search · full text read via the copy at PubMed Central (PMC6680710), including the summary table and the statement on funding and interests · read 2026-08-04

Design: Narrative review: the authors set out the existing literature and derive from it recommendations for the off-season of natural bodybuilders, that is, the period outside contest preparation in which training happens on an energy surplus. No measurement of their own, no search strategy printed and no weighting of the studies included.
Studied in: Not applicable, as there is no sample of its own. The article addresses natural bodybuilders in the off-season. It does report what that group eats in practice: up to 4.3 g/kg/d among men and 2.8 g/kg/d among women, well above what the article itself recommends.
Compared with: The earlier guideline for this group, which was written in energy per cent: 25 to 30 per cent of energy intake from protein.
Dose and duration: 1.6 to 2.2 grams of protein per kilogram of body weight per day, on an energy surplus of roughly 10 to 20 per cent and a weekly gain of 0.25 to 0.5 per cent of body weight.
What was measured: Not applicable as a measured outcome; this is a review and not a study.
What was found: The recommendation is a minimum of 1.6 g/kg/d in the off-season, with 2.2 g/kg/d as the target for anyone wanting to be sure. The 2.2 comes straight from the top of Morton et al.'s interval, and the authors say so. Notable is their argument against the other unit: a recommendation in energy per cent works out badly for a light person with a high energy requirement, who then lands far above what is needed, and it can squeeze out room for carbohydrate and fat. Which is why they count per kilogram of body weight.
Supports on this page: The row in the table for a building phase, 1.6 to 2.2 g/kg/d. It also carries the argument that a unit is a choice and not a detail, and that a recommendation in energy per cent and one per kilogram diverge for the same person.
Explicitly does not support: A narrative review is neither a systematic review nor a reference value: there is no search strategy, no selection criterion and no weighting. The upper bound of 2.2 is moreover not independent confirmation of Morton et al., because it comes from that same interval; two rows in the table therefore lean on one analysis. The recommendation is for natural bodybuilders in an off-season on an energy surplus, not for anyone who simply trains. The article covers a great deal more than protein and those other recommendations are not reproduced here, including the ones about supplements. It says nothing about a brand.
Interests and funding: The authors declare no interests and state that there was no external funding. Against that stands what appears under their names: the first author runs a nutrition consultancy of his own, and so does the second. The last author is the same as on source 9, and both papers recommend at the upper end of the ranges they find.

https://doi.org/10.3390/sports7070154

9Helms ER, Zinn C, Rowlands DS, Brown SR. A Systematic Review of Dietary Protein During Caloric Restriction in Resistance Trained Lean Athletes: A Case for Higher Intakes. International Journal of Sport Nutrition and Exercise Metabolism 2014;24(2):127-138. DOI 10.1123/ijsnem.2013-0054.systematic review, six studies · full text read, PDF retrieved from a public copy at paulogentil.com and extracted with pdftotext; the publisher's version at Human Kinetics sits behind a paywall · read 2026-08-04

Design: Systematic review with a search from the earliest record to July 2013. Included were studies in adults of 18 and over who were in an energy deficit, had been doing resistance training for at least six months and carried a low body fat percentage: men up to and including 23 per cent, women up to and including 35. The results were then converted to grams per kilogram of FAT-FREE MASS rather than per kilogram of body weight, and that is the point of the paper.
Studied in: Six studies with thirteen groups between them. In all thirteen the body fat percentage fell, by 0.5 to 6.6 percentage points. In nine of the thirteen the fat-free mass fell as well, by 0.3 to 2.7 kg. The six groups that lost none sat at the higher end for body fat, the lower end for energy deficit, or received a training stimulus that was new to them.
Compared with: The standing protein recommendations for strength athletes, which according to the authors were never established in people in an energy deficit or at a low body fat percentage, and the 1.8 to 2.7 g per kilogram of body weight that Phillips and Van Loon gave in 2011.
Dose and duration: Protein intakes varied across the studies included; well-matched groups existed at only four levels, namely 0.8, 1.0, 1.6 and 2.3 g per kilogram of body weight. Converted to fat-free mass the conclusion lands at 2.3 to 3.1 g/kg.
What was measured: Retention of fat-free mass during an energy deficit, set against protein intake per kilogram of fat-free mass.
What was found: The range of 2.3 to 3.1 grams per kilogram of fat-free mass per day was the most consistently protective against loss of fat-free mass. The authors add that within that range you move up the leaner the person and the larger the deficit, and down when neither holds. Their second argument is about the arithmetic itself: recommendations were traditionally established in people at ordinary or high body fat, and per kilogram of total body weight such a figure comes out too low for a lean person. A counter-finding is in there too: in one study outside the analysis the group on 1.6 g/kg retained slightly more fat-free mass than the group on 2.4 g/kg, although that difference was not significant and the two groups also ate very different amounts of carbohydrate. Finally the authors name something that may matter more than the protein choice: a slower rate of loss, around half a kilogram or 0.7 per cent of body weight a week.
Supports on this page: The one row in the table that counts per kilogram of FAT-FREE MASS, 2.3 to 3.1 g/kg, and with it the reason this calculator asks for a body fat percentage. It also carries the top of the whole band: for someone lean and in a deficit, the highest published figure sits far above anything a European authority has written down.
Explicitly does not support: Six studies, with well-matched groups at only four protein levels. This is not a reference value and not a dose-response curve, but the range that worked out most consistently across those six. The people included had been resistance training for at least six months and sat below 23 per cent body fat for men and 35 for women; outside those bounds these figures say nothing. The review itself names a counter-finding in which more protein did not work out better, and names the rate of loss as possibly mattering more than the protein choice. The text we read carries NO declaration of interests and NO funding statement; that is not evidence that there were none. The study says nothing about a supplement and nothing about a brand.
Interests and funding: The text we read names no declaration of interests and no funder. The acknowledgements thank four people for their guidance and four for their comments, among them two who appear elsewhere in this source list as authors of the sports nutrition society's position stand. University research, carried out in New Zealand.

https://doi.org/10.1123/ijsnem.2013-0054