This is not medical advice. General information about food tracking, not a treatment plan. If you are under 18, pregnant or breastfeeding, have ever had an eating disorder or a difficult relationship with food, or are managing a medical condition, talk to a doctor or a registered dietitian before you start counting calories or change your intake. The full list is here.
I build one of these calculators, so at some point I had to pick a protein number and be able to defend it. The uncomfortable part of that job is that there is no single number to pick. There is a set of recommendations from credible bodies, they disagree with each other by a factor of three, and every one of them is right about the question it was actually answering.
This post is about what those questions are. If you want the plain version of what protein, carbohydrate and fat each do, that is macros explained. If you want to know whether the split is worth tracking at all, that is do you need to track macros. This one is about why the number itself keeps moving.
Five sources, five protein numbers, all defensible
Here is what the main bodies actually recommend. The right-hand column is the part that gets dropped whenever one of these numbers is quoted on its own.
| Source | Recommends | Written for | Scaled to |
|---|---|---|---|
| Institute of Medicine RDA (2005) | 0.8 g/kg/day | All healthy adults | Bodyweight |
| Humayun et al. (2007) | 1.2 g/kg/day | Young men, method reappraisal | Bodyweight |
| Leidy et al. (2015) | 1.2 to 1.6 g/kg/day | Adults managing bodyweight | Bodyweight |
| Morton et al. (2018) | about 1.6 g/kg/day | Adults doing resistance training | Bodyweight |
| ISSN position stand (2017) | 1.4 to 2.0 g/kg/day | Exercising individuals | Bodyweight |
| Helms et al. (2014) | 2.3 to 3.1 g/kg/day | Lean trained athletes in a deficit | Fat-free mass |
| PROT-AGE (2013) | 1.0 to 1.2 g/kg/day | Adults over 65 | Bodyweight |
Run one 70 kg person through that table and the answers are 56 g at the bottom1 and, taking 15% body fat so that fat-free mass is 59.5 kg, 184 g at the top.5 That is not a rounding disagreement. One figure is more than three times the other, and every row has a peer-reviewed paper or a national committee behind it.
They stop looking like contradictions once you read the third and fourth columns. The RDA asks how little a population can eat without drifting into deficiency.1 Morton asks where extra protein stops adding muscle to a training programme.3 Helms asks how much a lean athlete needs to hold onto muscle while cutting for a competition.5 Take the RDA to a bodybuilder eight weeks out from a show and it is far too low. Aim a national nutrition policy at the Helms figure and you have told an entire country to eat like a competing athlete.
Per kilogram of what?
Look again at that last column. Six rows are per kilogram of bodyweight and one is per kilogram of fat-free mass, and that single difference does more damage to a protein target than all the others put together.
Take two people. One weighs 100 kg at 40% body fat. The other weighs 70 kg at 15%. Their fat-free mass is 60.0 kg and 59.5 kg, which for any practical purpose is the same amount of lean tissue.
| 100 kg at 40% fat | 70 kg at 15% fat | |
|---|---|---|
| Fat-free mass | 60.0 kg | 59.5 kg |
| 1.6 g per kg of bodyweight | 160 g | 112 g |
| 1.6 g per kg of fat-free mass | 96 g | 95 g |
Scaled to lean mass, the two targets land within a gram of each other, which is what you would expect if the tissue protein maintains is the tissue that is not fat. Scaled to bodyweight, one of them gets 43% more protein than the other for having more fat on them.
The athlete literature scales to fat-free mass for exactly that reason, and it says so on the label: the Helms recommendation is written per kilogram of FFM, for men under 23% body fat and women under 35%, where the two scalars nearly converge anyway.5 The general recommendations scale to bodyweight because bodyweight is a number everybody already has. A fat-free mass target needs a body composition measurement, and most people either do not have one or have one from a bathroom scale they would not bet on.
There is a third scalar in circulation: goal weight. Some calculators take the bodyweight you are aiming at and scale protein to that, which splits the difference for anyone with a lot to lose. It sounds reasonable, so I went looking for the trial behind it. I could not find one that randomised people to a goal-weight target against a current-bodyweight target and then measured what happened to their lean mass. That does not make it wrong. It means nobody has checked, and a calculator using it is making the same guess you would be.
The RDA is a population floor, not your target
0.8 g/kg/day is the figure that turns up most often in protein arguments, and it is almost always doing the wrong job in them. It is a Recommended Dietary Allowance, which in the Institute of Medicine's framework means the intake judged sufficient to meet the requirement of nearly all healthy people in a group.1 It is built to prevent deficiency, and it does that job well.
How it was arrived at matters, because a different technique moves it a long way. The estimated average requirement of 0.66 g/kg and the RDA of 0.8 both come from nitrogen balance studies. In 2007 a group at the Hospital for Sick Children in Toronto ran seven graded protein intakes through the indicator amino acid oxidation technique in eight healthy young men and got a mean requirement of 0.93 g/kg and a population-safe intake of 1.2 g/kg, which they reported as 41% and 50% above the standing figures.2
Eight men, a crystalline amino acid mixture rather than food, one laboratory, one technique. The authors did not claim the RDA was wrong. They concluded that current protein requirements "are too low and require reassessment", and the RDA you will see quoted today is still 0.8.
Either way, the RDA answers a question almost nobody opening a macronutrient tracker is asking. Deficiency is not the thing you are trying to avoid when you set a protein target during a diet. The same report is worth knowing for a different number: it sets the Acceptable Macronutrient Distribution Range for protein at 10 to 35% of energy,1 and that ceiling is what the rest of this post keeps bumping into.
Where the curve flattens, and how wide the error bars are
1.6 g/kg is the figure this site has quoted twice, and I owe it a correction.
It comes from Morton and colleagues in 2018: 49 randomised trials, 1,863 participants, resistance training of at least six weeks, comparing protein supplementation against control. Two findings get quoted from it. The first is the size of the effect, which is smaller than the discourse around protein suggests: supplementation added 0.30 kg of fat-free mass (95% CI 0.09 to 0.52) and 2.49 kg to one-repetition maximum strength (95% CI 0.64 to 4.33). The second is the breakpoint, that above roughly 1.62 g/kg/day more protein bought no further fat-free mass.3
What I left out of the earlier posts is how that breakpoint was arrived at. It came from a two-phase segmental regression on 42 study arms and 723 participants. The 95% confidence interval around 1.62 runs from 1.03 to 2.20 g/kg/day, the model explained 19% of the variance, and the authors present the whole thing, in their words, "despite not being statistically significant (p=0.079)".3
So 1.6 g/kg is a reasonable central estimate from the best meta-analysis available, and it is also a number whose interval is wide enough to contain both 1.0 and 2.2. Quoting it to two decimal places, as I have done, claims more precision than the analysis holds. Quoting it as a hard ceiling quotes a line the authors themselves flagged as not significant.
The International Society of Sports Nutrition's position stand puts the practical range at 1.4 to 2.0 g/kg/day for most exercising people, and points out that this sits inside the AMDR.4 That is the band I would now defend. It covers most of the confidence interval and it does not pretend to a precision nobody has.
Two other results in the Morton analysis set up the next two sections. The benefit of supplementation shrank with age, by 0.01 kg of fat-free mass per year (p=0.002), and it was larger in people who already trained, by 0.75 kg (p=0.03).3
A calorie deficit moves the number up
Everything above assumed you were eating enough. Restrict calories and protein takes on a second job: not building tissue so much as stopping the body dismantling what is already there. That mechanism is why tracking macros for fat loss comes down to protein and very little else.
Helms and colleagues reviewed six studies covering 13 groups of resistance-trained athletes in a deficit and concluded that their protein needs are "likely 2.3-3.1g/kg of FFM scaled upwards with severity of caloric restriction and leanness".5 The ISSN position stand carries the same range across.4
Read the inclusion criteria before you take that number anywhere. It is six studies, in adults over 18 with more than six months of resistance training, men at or under 23% body fat and women at or under 35%. Body fat fell in all 13 groups and fat-free mass fell in nine of them. The one group that was neither high in body fat nor on a gentle deficit and still held its lean mass was eating the most protein in the entire review, at 2.5 to 2.6 g/kg.5
That is a small and unusually lean population, and the recommendation is scaled to fat-free mass rather than bodyweight. For someone with 30 kg to lose who does not lift, it is the wrong reference class in two directions at once.
The general-population version is gentler and better hedged. Leidy and colleagues, reviewing the weight-management literature, landed on 1.2 to 1.6 g/kg/day with roughly 25 to 30 g per meal, and were careful about why the evidence looks the way it does: the tightly controlled short-term feeding studies were consistent, the longer-term ones produced "limited and conflicting findings", and adherence appeared to be the reason. The benefit turned up in the people who actually stuck to the higher-protein diet, and not in the ones who drifted off it.6
After about 65 it moves up again
Older muscle responds less to the same amount of protein. The PROT-AGE study group, convened by the European Union Geriatric Medicine Society, put the recommendation for people over 65 at an average of 1.0 to 1.2 g/kg/day, at least 1.2 for those who exercise, and 1.2 to 1.5 for most older adults with acute or chronic illness.7
That is above the RDA, which is the whole point of the paper: 0.8 is not age-adjusted, and the group's argument is that it is not enough to hold onto lean mass and physical function later in life.
One exception is written into the recommendation, and it is the reason this belongs with a doctor rather than a calculator. Older people with severe kidney disease who are not on dialysis, meaning an estimated GFR under 30, may need to limit protein instead.7
For older adults deliberately losing weight the case for more protein is stronger still. A meta-analysis of 20 randomised trials in adults over 50 compared higher protein intakes, defined as at least 25% of energy or 1.0 g/kg/day, against lower ones during energy restriction, and found the higher-protein groups retained more lean mass and lost more fat.8 Notice the threshold: 1.0 g/kg counted as the high arm there. In the athlete literature that would be the control group.
What macro calculator apps are actually doing
Almost every calculator you will meet uses one of two methods, and they are not the same method with different constants.
The first sets protein per kilogram of bodyweight and lets the calorie split fall where it falls. The second sets protein as a percentage of the calorie target. MyPlate, the app I build, does the second: protein is 28% of your daily calories while you are losing weight and 25% otherwise, with the remainder split between carbohydrate and fat. Both shares sit inside the AMDR, which is deliberate, and the formula is published with its citations on the health sources page.
Run four profiles through it and this is what comes out, with each gram figure converted back to grams per kilogram so it can be read against everything above.
| Profile, all losing weight | Daily calories | Protein | Per kg bodyweight |
|---|---|---|---|
| Woman, 70 kg, 165 cm, 35, lightly active | 1,630 | 114 g | 1.63 |
| Man, 100 kg, 180 cm, 40, lightly active | 2,255 | 157 g | 1.57 |
| Man, 70 kg, 175 cm, 25, very active | 2,703 | 189 g | 2.70 |
| Woman, 110 kg, 165 cm, 45, lightly active | 2,039 | 142 g | 1.29 |
The first two land almost exactly on 1.6 g/kg, and that is not luck. Calories scale with bodyweight through the Mifflin-St Jeor equation, so a fixed percentage of calories tracks bodyweight for anyone near the middle of the distribution. For most people the two methods quietly agree, which is why nobody notices the difference.
The edges are where they part. The very active 25-year-old gets 2.70 g/kg, past the point at which Morton's analysis found no further fat-free mass,3 because his activity multiplier raised his calories and the percentage came along for the ride. Nothing about that harms him, but about 77 g of it is there because of the arithmetic rather than because of the evidence.
The 110 kg profile is where the two readings come apart. Her 142 g is 1.29 g/kg of bodyweight, which looks low beside every recommendation in this post. Against her fat-free mass, if she is 45% body fat, it is 2.35 g/kg, which sits inside the range Helms recommends for athletes cutting.5 The same 142 g is either short or generous depending entirely on which kilogram you divide by, and neither reading is a mistake.
Two things are worth knowing about any calculator, mine included. It cannot see your body composition, so it is guessing at the scalar in that last column. And MyPlate does not recompute your targets when you log a weight: the plan is written when you run setup and stays there until you run it again. Lose 10 kg without revisiting it and your protein number still belongs to the person you were. That is a design choice I am reporting rather than defending.
There is a related collision at the top end that macros explained works through in detail: hold a heavy person on a low calorie target at 1.6 g/kg and the resulting percentage can run straight past the AMDR ceiling of 35%.
What the evidence does not settle
A lot of protein advice is delivered with more confidence than the work under it supports. Three questions in particular come up constantly and do not have settled answers.
The first is distribution across the day. The study behind that advice put eight healthy adults through a seven-day crossover, feeding the same total protein either evenly across three meals, about 30 g each, or skewed towards dinner at 11, 16 and 63 g. The even pattern produced a 24-hour muscle protein synthesis rate 25% higher, and the difference held after a week of habituation.9 It is an elegant result. It is also eight people, one week, and a synthesis rate rather than a measurement of anyone's muscle. Schoenfeld and Aragon reviewed the acute and long-term work together and landed on 0.4 g/kg per meal across a minimum of four meals as a way of reaching 1.6 g/kg/day, offered as what the current data support rather than as an established requirement.10 Spreading protein out across the day is a sensible default on that evidence, which is thinner than the confidence it usually gets quoted with.
The second is the goal-weight scalar from earlier, which as far as I can tell has never been tested head to head against current bodyweight.
The third is the upper limit, and the kidney question has the best answer of the three. A meta-analysis of 28 randomised trials and 1,358 healthy adults compared higher protein intakes, defined as at least 1.5 g/kg or at least 20% of energy or at least 100 g a day, against lower or normal ones. Post-intervention glomerular filtration rate was slightly higher on the higher-protein diets, which the authors called a trivial effect; the change in GFR from before to after did not differ between the groups. Their conclusion was that higher intakes "do not adversely influence kidney function on GFR in healthy adults", and the limitation they named was an unclear risk of selection bias in the trials they pooled.11 Healthy adults is doing a lot of work in that sentence. It says nothing about existing kidney disease, which is the case PROT-AGE carves out.7
How to pick one number and stop moving it
The practical answer is much narrower than the literature, because most of the spread in that first table comes from populations you are not in.
Start from bodyweight unless you have a body composition measurement you would actually bet on. It is what every general recommendation is scaled to and it is the number you already have. If you are neither training nor dieting, 1.2 to 1.6 g/kg covers both the reappraised RDA and the weight-management range with room to spare.26 If you are in a deficit, training, or both, sit at the top of that band or a little above, at 1.6 to 2.0 g/kg, which is the ISSN range and where most of Morton's confidence interval lives.4
If you carry a lot of fat, bodyweight will overshoot. Fat-free mass is the more defensible scalar when you can estimate it, and when you cannot, the low end of the bodyweight band is the safer place to sit. If you are over 65, take at least 1.2 g/kg unless a doctor has told you to limit protein.7
Then check the answer against your calorie target. Past 35% of energy you are outside the AMDR and crowding out carbohydrate and fibre you would rather have kept.1
The gap between 1.6 and 1.9 g/kg is about 20 g of protein for a 70 kg person, which is one pot of Greek yoghurt. The gap between hitting your number four days a week and seven is far larger, and it is the one you have any control over. Pick a figure inside the band and stop re-deciding it.
Common questions
How much protein do I need per day?
For most adults, somewhere between 1.2 and 2.0 grams per kilogram of bodyweight, and where you sit in that band depends on what you are doing with your body rather than on which calculator you opened. The Institute of Medicine's RDA of 0.8 g/kg is a deficiency-prevention figure for a population, not a performance target.1 A meta-analysis of 49 resistance-training trials put the point where extra protein stopped adding fat-free mass at about 1.62 g/kg/day, though its 95% confidence interval ran from 1.03 to 2.20 and the authors reported that analysis as not statistically significant.3 The International Society of Sports Nutrition's practical range of 1.4 to 2.0 g/kg/day for exercising adults is the one I would use.4
Should I calculate protein from bodyweight or lean body mass?
Bodyweight, unless you have a body composition measurement you trust and a lot of fat to lose. The two scalars diverge sharply as body fat rises: a person at 100 kg and 40% body fat and a person at 70 kg and 15% have almost identical lean mass, around 60 kg, but scaling 1.6 g/kg to bodyweight hands them 160 g and 112 g. The athlete recommendations that use fat-free mass were written for men under 23% body fat and women under 35%, where the two numbers nearly converge anyway.5 If you carry a lot of fat and cannot measure lean mass, the low end of the bodyweight band is the safer place to sit.
Why do macro calculator apps give me different protein targets?
Because they scale to different things and draw on recommendations written for different people. Some calculators set protein per kilogram of bodyweight, some as a percentage of your calorie target, and a few use your goal weight. Underneath them the published figures genuinely differ: 0.8 g/kg from the RDA,1 1.2 to 1.6 for weight management,6 1.4 to 2.0 for exercising adults,4 2.3 to 3.1 grams per kilogram of fat-free mass for lean athletes in a deficit,5 and 1.0 to 1.2 for adults over 65.7 Run one 70 kg person through all of those and the answers span 56 g to 184 g. The useful check is whether the number you ended up with was written for the population you are actually in.
Is a high-protein diet bad for your kidneys?
Not in healthy adults, on the evidence available. A meta-analysis of 28 randomised trials and 1,358 participants compared higher protein intakes, defined as at least 1.5 g/kg or at least 20% of energy or at least 100 g a day, against lower or normal ones, and found no difference in the change in glomerular filtration rate, although post-intervention GFR was slightly higher on the higher-protein diets. The authors named an unclear risk of selection bias in the pooled trials as their main limitation.11 None of this extends to existing kidney disease: the PROT-AGE recommendations explicitly carve out older people with an estimated GFR under 30 who are not on dialysis, who may need to limit protein instead.7 If you have any kidney condition, that is a conversation with a doctor rather than with a calculator.
What should I look for in a macro tracking app?
Whether it tells you which formula produced your number, and whether you will still be opening it in month three. Any macro tracking app can print a protein target; the useful ones show you the arithmetic behind it and let you change it, because the published range is wide and no single figure is right for everyone. After that, weigh speed of logging above depth of database, since a rough log you keep beats a precise one you abandon. For the record, MyPlate sets protein at 28% of your calorie target while you are losing weight, publishes that formula with its citations on the sources page, and has no food database and no barcode scanner, which is a real limitation if most of what you eat comes out of a packet.
Pick a number you will actually hit
The spread between the credible recommendations is real and it is wide, but nearly all of it comes from the populations they were written for. Strip out the contest-prep athletes at one end and the deficiency floor at the other and what is left for most people is 1.2 to 2.0 grams per kilogram of bodyweight, sitting toward the top if you are training, dieting or over 65.
Pick a figure inside that band, check it does not push protein past 35% of your calories, and then leave it alone. Nothing in the evidence supports choosing between 1.6 and 1.8 with any confidence, and the difference between them is 14 g of protein a day for a 70 kg person.3
One thing not to do: raise the protein percentage by cutting calories further. If you are eating in a deficit, the floor worth respecting is 1,200 calories a day for women and 1,500 for men, which is where the NIH clinical guidelines put it and where MyPlate's own targets stop no matter what the formula produces.12
And if a post like this one leaves you wanting to recompute everything tonight and re-weigh your dinner, that is the point to step back rather than refine further. Tracking that has turned into something you have to get right, rather than something that helps, is worth raising with a doctor or a registered dietitian.
Sources
Every figure above traces to one of these. If you find a number that doesn't match the source it claims, tell me and I'll correct it.
- Institute of Medicine (2005). Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. The National Academies Press, Washington DC. Read the report →
- Humayun MA, Elango R, Ball RO, Pencharz PB (2007). Reevaluation of the protein requirement in young men with the indicator amino acid oxidation technique. American Journal of Clinical Nutrition, 86(4):995-1002. Read on PubMed →
- Morton RW, Murphy KT, McKellar SR, et al. (2018). 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, 52(6):376-384. Read on PubMed →
- Jager R, Kerksick CM, Campbell BI, et al. (2017). International Society of Sports Nutrition Position Stand: protein and exercise. Journal of the International Society of Sports Nutrition, 14:20. Read on PubMed →
- Helms ER, Zinn C, Rowlands DS, Brown SR (2014). 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, 24(2):127-138. Read on PubMed →
- Leidy HJ, Clifton PM, Astrup A, et al. (2015). The role of protein in weight loss and maintenance. American Journal of Clinical Nutrition, 101(6):1320S-1329S. Read on PubMed →
- Bauer J, Biolo G, Cederholm T, et al. (2013). Evidence-based recommendations for optimal dietary protein intake in older people: a position paper from the PROT-AGE Study Group. Journal of the American Medical Directors Association, 14(8):542-559. Read on PubMed →
- Kim JE, O'Connor LE, Sands LP, Slebodnik MB, Campbell WW (2016). Effects of dietary protein intake on body composition changes after weight loss in older adults: a systematic review and meta-analysis. Nutrition Reviews, 74(3):210-224. Read on PubMed →
- Mamerow MM, Mettler JA, English KL, et al. (2014). Dietary protein distribution positively influences 24-h muscle protein synthesis in healthy adults. Journal of Nutrition, 144(6):876-880. Read on PubMed →
- Schoenfeld BJ, Aragon AA (2018). How much protein can the body use in a single meal for muscle-building? Implications for daily protein distribution. Journal of the International Society of Sports Nutrition, 15:10. Read on PubMed →
- Devries MC, Sithamparapillai A, Brimble KS, Banfield L, Morton RW, Phillips SM (2018). Changes in kidney function do not differ between healthy adults consuming higher- compared with lower- or normal-protein diets: a systematic review and meta-analysis. Journal of Nutrition, 148(11):1760-1775. Read on PubMed →
- National Heart, Lung, and Blood Institute, NIH (1998). Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. NIH Publication 98-4083. Read the guidelines →
The formulas and limits behind MyPlate's own numbers, with their citations, are on the health sources page.