Why it exists
Nearly every keto calculator online performs the same trick: take your maintenance calories, carve off a fixed percentage for carbs, another for protein, call the rest fat, and present it as science. It is backward. Ketosis responds to the absolute grams of carbohydrate you eat, not to the fraction of your plate they represent, so deriving carbs from a percentage gives a large man twice the carbohydrate of a small woman and labels both plans ketogenic. One of them probably is not.
The second reason is the first two weeks. Everybody drops a lot of weight in week one on keto, and nearly all of it is the water that was bound to glycogen. Every calculator we looked at either ignores this entirely or buries it in a sentence somewhere below the fold. It is the single most common reason people misread their own progress, either declaring victory in week one or quitting in week three when the water is gone and the real rate finally shows up. We put a number on it, and we draw it.
Who writes this
A small independent team that builds calculators. None of us holds a clinical qualification, and you will not find a stock photo of a physician with a blue check mark anywhere on this site. That pattern is everywhere in health content and it is worth close to nothing, because you cannot verify who the reviewer is or what they actually reviewed.
What you get instead: every equation named on the homepage with its source, and the pieces we constructed ourselves labeled as ours. If you think a number is wrong you can go check it, which is a better guarantee than a name you have no way to audit. Corrections are genuinely welcome at hello@ketomacrocalculator.com.
What we are confident about
- Nutritional ketosis starts at a blood beta-hydroxybutyrate of 0.5 mmol/L, and 20 to 50 g of net carbs a day is the range that produces it. This is consistent across the entire literature.
- Glycogen is stored with water, and losing it accounts for most of what the scale does in week one. The mechanism is not in dispute.
- Gluconeogenesis is demand-driven. Eating more protein does not reliably end ketosis, and the moderate-protein convention costs people lean tissue for a reason that does not survive scrutiny.
- At matched calories and protein, keto does not beat other diets for fat loss. Its real advantage is appetite suppression, which is worth having and is not what most keto marketing claims.
- Keto flu is sodium and potassium depletion, and salting your food properly prevents the large majority of it.
What is genuinely uncertain, and we say so on the page
- The water-per-gram-of-glycogen ratio. The classic figure is 3 grams of water per gram of glycogen, from Olsson and Saltin in 1970. More recent work argues the true ratio is lower and more variable between people. We publish a range rather than a point, and the range is wide on purpose.
- The protein interpolation. The research gives a band of 2.3 to 3.1 g per kg of lean mass, about 1.0 to 1.4 g per pound, and an instruction that leaner people belong higher in it. The exact curve we run between those two bounds is our own construction. It is a defensible reading of the evidence, not a published equation.
- Where your personal carb ceiling sits. Nobody can calculate this from your height and weight. A muscular person who trains hard may hold ketosis at 50 g while a sedentary person loses it at 30. Only a blood meter answers it.
- The fat mobilization ceiling. The 48.5 kcal per kg of fat mass figure comes from Alpert's modeling work, corrected for a unit error that circulates widely online. Alpert's published figure is per pound, and the frequently quoted per-kilogram version is a factor of 2.2 too small. We use the conservative corrected value, and it remains a model rather than a measurement.
- Activity multipliers. They are conventions, not measurements. Being one tier out is worth several hundred calories a day, which is more than all the precision elsewhere on the page put together.
Safety
This is information, not medical advice, and there are three situations where you genuinely should not act on it without talking to your doctor first.
- SGLT2 inhibitors. Dapagliflozin, empagliflozin and canagliflozin combined with very low carbohydrate have produced ketoacidosis at normal blood glucose, which is dangerous precisely because your sugars look reassuring while it happens. There are multiple published case reports of exactly this combination.
- Insulin or sulfonylureas. Cutting carbohydrate sharply while your dose stays where it was is a direct route to a hypo. Doses usually need reducing within days.
- Kidney disease, liver disease, pancreatitis, pregnancy or nursing, or a history of disordered eating. Each of these changes the calculation in ways no web page can account for.
The calculator also refuses to hand you a target below the greater of the usual 1,200 or 1,500 calorie guideline and 30 calories per kilogram of lean mass (about 14 per pound), and it tells you when it has raised your number. Keto suppresses appetite well enough that people undereat by accident, which is exactly why that floor is there.
Privacy
Everything runs in your browser. Nothing you type is sent anywhere, stored, or used to build a profile of you. There is no account, no email capture, and nothing for sale.