How the keto calculator works, where it goes wrong, and how to read the result

The keto calculator estimates calories with the Mifflin–St Jeor equation, takes about 85% of that figure for weight loss, and splits it into carbs (5–10%), protein (30–35%) and fat (55–60%). Each of the three steps is an approximation. Below is how rough each one is, what the calculator does not know about you, and when keto is a conversation for your doctor before the first calculation.

What the calculator actually computes

The form takes sex, age (18 or over), weight, height, activity level and a goal: lose weight or maintain. Then it does three things.

1. Basal metabolic rate from the Mifflin–St Jeor equation: 10 × weight in kilograms + 6.25 × height in centimetres − 5 × age, plus 5 for men or minus 161 for women.
2. Multiplication by an activity factor that the user picks. The result is the calorie level that keeps weight stable.
3. For the “lose weight” goal, the middle of the 80–90% corridor of that level is used, about 85%. The calculator does not go below 1,200 kcal for women or 1,500 kcal for men: it substitutes that floor and says such a calorie level is a matter for a doctor. For “maintain weight” it uses the level itself.

The resulting calories are split by share: carbs 5–10%, protein 30–35%, fat 55–60%. In grams it looks like this:

An example from the form: a 35-year-old woman, 65 kg, 165 cm, activity factor 1.375 (the default), goal “lose weight”. Her maintenance level comes out at 1,850 kcal, the keto base is 1,570 kcal, carbs 20–39 g, protein 118–137 g, fat 96–105 g.

What the assumptions rest on

Calories. In a 2005 systematic review, the Mifflin–St Jeor equation, among the four most widely used, more often than the others landed within 10% of measured resting metabolic rate. The same authors state that noteworthy errors remain for an individual. In one study included in the review, 83 non-obese adults, the equation was within 10% for 82% of participants. In a separate study of women with different body mass index it was 71%. With a resting rate of about 1,350 kcal, 10% is roughly 135 kcal either way, and depending on the sample the miss is larger for one person in five or one in three. The activity factor adds its own error: people choose it by self-assessment.

The split. The shares come from the StatPearls review (Daley et al., updated December 2025), which gives them as typical for a ketogenic diet: fat 55–60%, protein 30–35%, carbs 5–10% of calories. That describes how such a diet is usually put together; it is not the result of a trial that found the best split. The same review separately describes a very-low-carbohydrate version: 20–50 g of carbs a day, moderate protein and 60–75% of calories from fat. It is most often used for obesity, type 2 diabetes and metabolic syndrome. At 2,000 kcal the calculator gives 25–50 g of carbs, which falls in the same corridor.

Where the calculator simplifies and errs

The ranges do not add up to a whole. At 2,000 kcal the lower ends give 1,800 kcal and the upper ends 2,100. The calculator shows three independent corridors, not a ready menu, so values inside them have to be picked so that the total matches the target calories.

It is not a therapeutic ketogenic diet. The classic ketogenic diet used for drug-resistant epilepsy is built on a ratio of fat to protein and carbohydrate combined. The most common ratio, 4:1, puts 90% of calories into fat. The calculator uses 55–60% fat, so its output is closer to a low-carbohydrate diet than to a medical protocol. Therapeutic versions are run by a team of doctors and dietitians, and the calculator does not replace them.

Protein comes out high at low calories. In the example above, 118–137 g of protein for 65 kg is 1.8–2.1 g per kilogram. By the EFSA opinion, the reference intake for adults is 0.83 g per kilogram, and twice that is considered safe, about 1.7 g/kg. Above that line harm has not been shown: EFSA could not set an upper intake level for protein because the data are insufficient. In the example the whole corridor sits above it, and the lower the weight and calories, the more so. StatPearls adds a warning that ketosis and high protein can increase the load on the kidneys, especially where kidney disease already exists.

Carbs here means all carbs, vegetables and dairy included. That is stricter than counting “net” carbs after subtracting fibre, and figures from other sources may not match.

The calculator does not know the most important things about you: diagnoses, medication, pregnancy, a history of an eating disorder. Its whole response to that is a warning under the result.

Finally, it produces targets, not what is on the plate. Carbs from vegetables, nuts and dairy are easy to underestimate, and fat in the pan and in sauces goes unnoticed, so for the first weeks it helps to weigh and log food rather than judge it by eye.

What keto adds beyond calories

The calculator page says that weight comes off through a calorie deficit, not through ketosis itself. That rests on comparisons.

In the experiment by Hall and colleagues (2016), 17 men with overweight or obesity lived in a metabolic ward: 4 weeks on a high-carbohydrate diet, then 4 weeks on keto with the same calories and the same protein. Fat loss slowed during the first two weeks of keto, and energy expenditure rose by 151 ± 63 kcal a day by the doubly labelled water method, which is near the limit of what the method can detect. The authors conclude that at equal calories the ketogenic diet did not produce more fat loss.

In the DIETFITS trial, 609 adults aged 18–50 with a body mass index of 28–40 and no diabetes followed either a “healthy low-carbohydrate” or a “healthy low-fat” diet for a year. The difference in weight change was 0.7 kg (95% CI −0.2 to 1.6), not statistically significant. A caveat: by the end of the year the low-carbohydrate group was eating on average about 30% of calories as carbs, so this was not keto. The same applies to a meta-analysis of 38 studies, where low-carbohydrate meant no more than 40% carbs: at 6–12 months it showed 1.3 kg more weight loss, and LDL cholesterol was 0.07 mmol/L higher.

For real keto with 5–10% carbs, long comparisons are scarce. StatPearls notes that most clinical trials of keto last under a year and that long-term outcome data are limited. The large water loss in the first week and the cholesterol question are covered in our article on the keto diet.

Who should not do keto without a doctor

The calculator gives numbers to anyone who asks. Below are the conditions where the question “should I start” is for a doctor to answer, not a formula.

• Type 1 diabetes. The main risks of keto here are hypoglycaemia and ketoacidosis, and according to a 2020 review a diabetes specialist should assess them.
• Diabetes medication. The FDA linked SGLT2 inhibitors (gliflozins) to ketoacidosis in 2015: its database held 73 cases between March 2013 and May 2015 in people with type 1 and type 2 diabetes, all ending in hospitalisation or an emergency visit, often with glucose lower than usual for ketoacidosis. Among the potential risk factors identified in those cases the FDA listed a low-carbohydrate diet or an overall reduction of caloric intake, along with infection, a reduced insulin dose and alcohol use. A review in Annals of Medicine describes a patient in whom such ketoacidosis developed one week into keto on an SGLT2 inhibitor, and notes that doses of insulin and sulfonylureas have to be reviewed on keto.
• Pregnancy and breastfeeding. According to StatPearls, keto is not recommended in pregnancy: severe carbohydrate restriction may impair fetal growth and raise the risk of deficiencies, and safety has not been proven. The Annals of Medicine review adds lactation to its list of situations requiring particular caution.
• Pancreas and liver. StatPearls lists acute and chronic pancreatitis and liver failure as contraindications.
• Kidneys. For kidney disease StatPearls advises regular checks of kidney function, and Annals of Medicine calls advanced stages of chronic kidney disease a relative contraindication.
• Digestive tract. Nausea, vomiting, constipation and diarrhoea are common side effects of keto in the first weeks, per StatPearls. Gallstone disease and a removed gallbladder are on the list of conditions for which Annals of Medicine advises particular caution.
• Eating disorders. StatPearls lists a history of an eating disorder among the contraindications: rigid rules and rapid weight changes can trigger a relapse. If counting food causes anxiety, read about when calorie counting becomes a problem.
• Blood pressure medication, diuretics and electrolyte disturbances. Annals of Medicine includes them among the situations requiring caution.
• Rare inherited disorders of fat metabolism, and porphyria. In the same review these are absolute contraindications.

The calculator is for adults. For children, keto is prescribed as treatment, for example in epilepsy, and managed by a team of doctors and dietitians.

How to read the result

• The carb line is the limit for the day; the rest are guides. The limit counts all carbs, vegetables and dairy included.
• The calorie figure is a starting point, not a norm. It can differ from your real expenditure by 10% or more, so after three or four weeks compare the number with how you feel and your appetite, not with the table value.
• At a small body weight and about 1,500–1,600 kcal, the protein corridor goes above 1.7 g/kg. Discuss it with a doctor or dietitian, especially if you have kidney disease.
• A sharp drop in weight in the first week is mostly water: glycogen is stored with water, three to four parts per part of glycogen.
• Weakness, headache and dizziness in the first days are common: per StatPearls they pass within days or weeks. If they do not, or vomiting and marked weakness appear, see a doctor: the same review lists dehydration, electrolyte disturbances and hypoglycaemia among the causes of hospital admissions.
• It makes sense to have a lipid panel before starting and after three months: cholesterol on keto behaves differently in different people.

If keto still looks like the right choice after that, you can work out the split in the keto calculator. Your basic calorie level without the split is in the calorie calculator, and how protein, fat and carbs relate outside keto is in the article on the macro ratio.

Count it from a photo

Frequently asked questions

How does the keto calculator work out calories?

With the Mifflin–St Jeor equation and an activity factor the user chooses. For the “lose weight” goal it takes the middle of the 80–90% corridor of the maintenance level, but not below 1,200 kcal for women or 1,500 kcal for men. The equation more often than others lands within 10% of measured resting metabolic rate, but noteworthy errors remain for an individual.

Where do the 5–10%, 30–35% and 55–60% shares come from?

From the StatPearls review (Daley et al., updated December 2025), which gives them as the typical split of a ketogenic diet: fat 55–60%, protein 30–35%, carbs 5–10% of calories. It describes usual practice, not the result of a trial comparing splits.

Why do the lower and upper ends not add up to the calorie target?

The calculator shows three independent corridors. At 2,000 kcal the lower ends give 1,800 kcal and the upper ends 2,100. Values inside the corridors are picked so that the total matches the target.

Is the protein in the result too high?

At a small body weight and about 1,500–1,600 kcal the protein corridor goes above 1.7 g per kilogram, which is twice the EFSA reference intake and the level EFSA calls safe for adults. Above it harm has not been shown, but there are no data for an upper limit either. With kidney disease, the amount of protein is a doctor’s decision.

Who should not switch to keto without a doctor?

People with diabetes (especially on insulin or SGLT2 inhibitors), during pregnancy and breastfeeding, with pancreatic, liver or kidney disease, gallstone disease, a past or present eating disorder, those treated for blood pressure with diuretics, and those with rare disorders of fat metabolism.

Does the calculator guarantee ketosis?

No. It counts grams from shares of calories and does not measure ketones. StatPearls gives 20–50 g of carbs a day for the very-low-carbohydrate version, but whether a person enters ketosis depends on the person.

Read next

• Keto diet: what happens in the body and what comparisons have shown — The keto diet promises a non-standard mechanism: not counting calories, but switching metabolism to a different fuel.
• How many carbs do you need per day? — For an illustrative 2,000-calorie diet, the European reference range of 45–60% of energy works out to about 225–300 g of carbohydrates per day.
• Macro ratio: where percentages come from and what to change first — Proportions like "30/30/40" look like an optimal formula someone has discovered.
• Calorie deficit: how to calculate it and how to hold it — A calorie deficit is the one thing without which weight does not come down.

Topics: Carbohydrates · Protein · What the research shows

This article is for general information. It is not medical advice, a diagnosis, or a prescription for treatment or a diet, and it does not replace a consultation with your doctor. If you have a health condition, are pregnant, take medication, or follow a diet prescribed to you, decisions about food belong with your doctor.

Figures from regulations, guidelines and studies are given as they stood when this article was prepared and may since have changed; check them against the primary sources. This article is not advertising, an offer, or individual advice, and neither the author nor the site owner is responsible for decisions taken on the basis of it.