This is not medical advice. These are findings from population trials, not a plan written for you. If you are under 18, pregnant or breastfeeding, have ever had a difficult relationship with food, or are managing a medical condition, talk to a doctor or a registered dietitian before changing what you eat. The full list is here.

The short answer

No, you don't need all three. The trials that hold calories steady and vary the split find almost nothing on the scale, and the one macro that reliably changes your result is protein. So the useful version of tracking macros for weight loss is two numbers, not four: calories, and protein.

If you want to know what protein, carbohydrate and fat each do, I wrote a separate guide to macros. This one is only about whether the tracking is worth your time, and it comes with the limits of the evidence attached.

Calories decide the weight

The cleanest test of this is POUNDS LOST, published in the New England Journal of Medicine in 2009. 811 overweight adults were randomised to one of four reduced-calorie diets spanning 20 to 40% fat, 15 to 25% protein, and 35 to 65% carbohydrate, then followed for two years with group and individual sessions.1 Every arm lost about 6 kg by six months, and by two years the gaps between them were under a kilogram and statistically indistinguishable.

The detail I find more interesting sits at the end of the results. Attendance at the group sessions was strongly associated with weight loss, at about 0.2 kg per session attended.1 Turning up predicted the outcome. The macro split didn't.

DIETFITS at Stanford ran the harder version of the same question in 2018. 609 adults, twelve months, a healthy low-fat diet against a healthy low-carbohydrate one, and the groups genuinely separated: 48% versus 30% of energy from carbohydrate at twelve months.2 Weight change was 5.3 kg against 6.0 kg, a difference of 0.7 kg with a confidence interval crossing zero. The trial also tested whether a genotype pattern or insulin secretion picked out who should be on which diet. Neither did.

Zoom out to a network meta-analysis of 48 randomised trials covering 7,286 people and the picture holds. Low-carbohydrate diets produced 7.25 kg of loss at twelve months and low-fat diets 7.27 kg.3 Between individual named programmes the largest gap the authors could find was Atkins beating Zone by 1.71 kg at six months, which is roughly what behavioural support was worth on its own.

Three separate lines of evidence give the same answer. Once the calories are matched, tracking macros for fat loss does not change how much fat comes off. What you are managing is the size of the deficit, and the ratio behind it is mostly a question of what you would rather eat.

One limit worth stating, because it cuts against me. These are counselled diets with prescribed targets, and adherence in all of them drifted toward the middle over time. Nobody has run a two-year trial where the splits stayed perfectly separated, because people don't eat that way. What the trials rule out is a large macro effect. A small one could be hiding in the noise.

Protein decides how much of it is muscle

Now the exception, and it comes from a meta-analysis that measured the right thing. Wycherley and colleagues pooled 24 randomised trials of 1,063 people comparing energy-restricted high-protein diets against standard-protein ones, matched for calories and for fat.4

The weight difference was 0.79 kg, which is nearly nothing and consistent with everything above. Underneath it, the higher-protein groups lost 0.87 kg more fat mass, held on to 0.43 kg more fat-free mass, and saw a smaller fall in resting energy expenditure. Nearly the same scale reading, made of different tissue.

That row of results is why protein earns a slot in your tracker while carbohydrate and fat do not. It changes what the loss is made of rather than how fast it happens.

How much protein, and where the number comes from

The figure everyone quotes without sourcing it is 1.6 g per kg of bodyweight. It has a source: a meta-analysis of 49 resistance-training trials with 1,863 participants, which found no further gains in fat-free mass above roughly 1.62 g/kg/day.5 Note what that means. 1.6 is where the benefit stops, not where it starts. Around 112 g a day for a 70 kg person, and going well past it buys you protein powder rather than muscle.

The caveat belongs to the researchers, not to me. A review in Advances in Nutrition concluded that adequate protein during weight loss preserves lean and muscle mass but does not improve muscle strength, and that excessive intakes could have adverse metabolic effects.6 The same review put resistance training alongside protein as the other thing that preserves muscle, and the only one of the two that made people stronger. If you take one action from this article, make it lifting something twice a week rather than adding a shake.

So how much do you actually need to log?

Most macro tracking for weight loss advice starts at the top of a ladder. The evidence supports stopping partway down it.

Track calories alone and you have the variable that decides the direction. That is enough to lose fat, and for a lot of people it is where tracking should end, because six months of rough logging tells you more than three weeks of careful logging.

Add protein and you are covering both effects the evidence supports, at the cost of watching one extra figure. This is the rung I would put almost everyone on. In practice it means glancing at a number once a day and deciding whether dinner needs eggs in it.

Tracking all four is for a narrower group than the internet implies: people prepping for a physique competition, people with a medical reason to hold carbohydrate or fat at a set level, and people who simply enjoy the completeness. None of those is a weight-loss argument. If you fall outside them and you are logging fat grams to three significant figures, you are buying precision the trials say you cannot spend.

Where the split does change what you eat

There is a real case against the argument I have just made. Macros can move your calorie intake without you deciding to, which makes them matter indirectly even where they don't matter arithmetically.

A controlled feeding study at the NIH admitted 20 adults as inpatients and let them eat as much as they liked of either a minimally processed plant-based low-fat diet (10% fat, 75% carbohydrate) or a minimally processed animal-based ketogenic one (76% fat, 10% carbohydrate), two weeks each, then swapped them.7 On the low-fat diet, people ate 689 kcal a day less. Nobody was counting anything. The food composition did it.

Two things about that result. It is 20 people for two weeks each, so it is a mechanism study rather than a weight-loss recommendation. And the direction is the reverse of what low-carbohydrate marketing predicts, which is the point the authors were making: the carbohydrate-insulin model expected the ketogenic arm to eat less, and it ate more.

So composition does something, through appetite rather than through arithmetic. That is an argument for noticing which meals leave you hungry at 4pm, which is a thing you can observe without a macro target.

What a food tracker for weight loss actually buys you

If the split isn't the lever and the deficit is, then the value of any food tracker for weight loss is whether it keeps you aware of the deficit. The dose-response evidence on that is more specific than you would expect.

142 participants in a 24-week online weight-control programme logged their intake through a web tool that recorded how often and how long they used it.8 By month six, the people who had lost 5% or more of their bodyweight were opening their food log 2.4 times a day against 1.6 for those who hadn't; for 10% or more it was 2.7 against 1.7. Time spent logging showed no such difference, and average time fell from 23 minutes a day in month one to about 15 by month six.

Frequency mattered, duration didn't, and it got faster with practice. Opening the log a few times across the day did more than one careful session at midnight, which is roughly the opposite of how most people set out to track.

The broader literature is consistent but weaker than it is usually reported to be. A systematic review of 22 self-monitoring studies found a significant association with weight loss across the board, then graded the evidence weak on methodological grounds: mostly paper diaries, reliance on self-report, and samples that in all but two studies were predominantly white and predominantly women.9 Anyone telling you tracking is proven to work is overstating what those 22 studies can carry.

What MyPlate does with this, and what it doesn't

My own app takes the position this article argues. It gives you a calorie target and a protein target from your profile, using Mifflin-St Jeor with an activity multiplier and a goal factor, and it splits the rest at 28% protein, 45% carbohydrate and 27% fat when your goal is weight loss. Those shares sit inside the Acceptable Macronutrient Distribution Ranges published by the National Academies, which are 10 to 35% for protein, 45 to 65% for carbohydrate and 20 to 35% for fat.10 The calorie calculator runs the same arithmetic in your browser if you want your own numbers without installing anything.

Whatever that arithmetic produces, the target never goes below 1,200 kcal for women or 1,500 for men, following the NIH clinical guidelines.11 Below that belongs with a clinician rather than an app.

Three limitations you should know before you trust any of it. MyPlate has no food database and no barcode scanner, so a logged meal is an AI estimate from a photo or a description, not a measurement, and the fat column is the least reliable one because cooking oil is invisible in a photograph. Logging a weight does not recompute your calorie target; you re-run the goal setup for that. And every target here is a population formula, which is a starting point to adjust from rather than a fact about your metabolism.

If tracking is making you anxious, or you find yourself logging to punish a meal rather than to understand it, stop and talk to someone. A number on a screen is a tool for a fortnight of information, not a verdict on the day. This applies more, not less, if you are good at it.

Common questions

Do I need to track macros to lose fat?

No. Calories decide whether you lose weight, and trials that matched the calories and varied the split found differences under a kilogram over one to two years.12 Protein is the exception worth tracking, because at the same calorie intake a higher-protein diet preserved 0.43 kg more fat-free mass and lost 0.87 kg more fat.4 Calories and protein cover both effects the evidence supports.

Is tracking calories and protein really enough?

For fat loss, yes. Carbohydrate and fat are what is left over once calories and protein are set, so tracking them separately adds bookkeeping rather than information. The reasons to track all four are a physique competition, a medical constraint on carbohydrate or fat, or enjoying the completeness. None of those is about losing fat faster.

Does a low-carb split burn more fat than a high-carb one?

Not at matched calories. The DIETFITS trial ran 609 adults for twelve months at 30% against 48% of energy from carbohydrate and found a 0.7 kg difference in weight change, with a confidence interval crossing zero.2 A network meta-analysis of 48 trials put low-carbohydrate and low-fat diets within 0.02 kg of each other at twelve months.3 Where composition does change things is appetite: in a controlled feeding study people ate 689 kcal a day less on a low-fat diet than on a ketogenic one when both were unlimited.7

How much protein should I eat while losing fat?

Around 1.6 g per kilogram of bodyweight, which is about 112 g a day at 70 kg. That figure comes from a meta-analysis of 49 resistance-training trials which found no further gain in fat-free mass above roughly 1.62 g/kg/day, so it is a ceiling rather than a floor.5 Protein preserves muscle during weight loss but does not make you stronger, and resistance training does both.6

What makes a good food tracker for weight loss?

How often you will open it, more than what is in it. In a 24-week programme, people who lost 5% or more of their bodyweight opened their food log 2.4 times a day against 1.6 for those who lost less, while time spent logging showed no difference at all.8 So judge an app on how fast one meal takes, whether it shows you a protein figure, and whether it tells you which formula produced your target. A thorough tracker you abandon in week three loses to a rough one you still open in month six.

Can I lose fat without tracking anything?

Yes, and people do. Tracking is a way of making the deficit visible, not the thing that creates it, and the review evidence linking self-monitoring to weight loss was graded weak because the studies leaned on self-report and their samples were mostly white women.9 If a fixed eating pattern keeps you in a deficit without logging, you are not missing anything a tracker would give you.

Track the two numbers that do something

Set a calorie target, set a protein target near 1.6 g/kg, and stop there. That covers the variable that decides whether fat comes off and the variable that decides how much of the loss is muscle. Carbohydrate and fat sort themselves out in the space left over, and the trials say they were never going to change your result anyway.

What is left is the part no macro split helps with: opening the app tomorrow, and the day after. The tracking research is much clearer about frequency than about precision,8 and a two-number system is far easier to keep up than a four-number one.

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.

  1. Sacks FM, Bray GA, Carey VJ, et al. (2009). Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. New England Journal of Medicine, 360(9):859-873. Read on PubMed →
  2. Gardner CD, Trepanowski JF, Del Gobbo LC, et al. (2018). Effect of low-fat vs low-carbohydrate diet on 12-month weight loss in overweight adults and the association with genotype pattern or insulin secretion: the DIETFITS randomized clinical trial. JAMA, 319(7):667-679. Read on PubMed →
  3. Johnston BC, Kanters S, Bandayrel K, et al. (2014). Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis. JAMA, 312(9):923-933. Read on PubMed →
  4. Wycherley TP, Moran LJ, Clifton PM, Noakes M, Brinkworth GD (2012). Effects of energy-restricted high-protein, low-fat compared with standard-protein, low-fat diets: a meta-analysis of randomized controlled trials. American Journal of Clinical Nutrition, 96(6):1281-1298. Read on PubMed →
  5. 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 →
  6. Cava E, Yeat NC, Mittendorfer B (2017). Preserving healthy muscle during weight loss. Advances in Nutrition, 8(3):511-519. Read on PubMed →
  7. Hall KD, Guo J, Courville AB, et al. (2021). Effect of a plant-based, low-fat diet versus an animal-based, ketogenic diet on ad libitum energy intake. Nature Medicine, 27(2):344-353. Read on PubMed →
  8. Harvey J, Krukowski R, Priest J, West D (2019). Log often, lose more: electronic dietary self-monitoring for weight loss. Obesity, 27(3):380-384. Read on PubMed →
  9. Burke LE, Wang J, Sevick MA (2011). Self-monitoring in weight loss: a systematic review of the literature. Journal of the American Dietetic Association, 111(1):92-102. Read on PubMed →
  10. Institute of Medicine (2005). Acceptable Macronutrient Distribution Ranges, in Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. The National Academies Press. Read the report →
  11. 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.