Is the 3,500 calories per pound rule true?

No, and it is the most persistent error in weight loss advice. It assumes your metabolism stays constant while you shrink, so it predicts endless linear loss. In reality energy expenditure falls as body mass falls, and the deficit closes itself. A better rule of thumb from the Hall model is that each permanent reduction of about 10 kcal a day produces roughly one pound of eventual weight loss — with half of it reached after about a year and 95% after about three.

Why did my weight loss stop even though I did not change anything?

Because you did change something: you got smaller. A lighter body costs less to run, so the deficit that produced loss at your starting weight is much closer to maintenance at your new one. A plateau is usually the arithmetic catching up rather than a failure of effort — the same intake that once created a 500 kcal gap may now create almost none.

How long does it take to lose weight?

Far longer than most calculators suggest. Modelling from the Hall group indicates roughly half of the eventual weight change happens in the first year and around 95% within three years of a sustained intake change. Predictions of a pound a week continuing indefinitely are a modelling artefact, not a biological finding.

What is the Katch-McArdle equation and when should I use it?

It estimates basal metabolic rate from lean body mass rather than total weight, so it is more accurate for people who know their body fat percentage reasonably well — particularly the very muscular or the very lean, where weight-based equations misfire. Without a measured body fat percentage, the Mifflin-St Jeor equation is the better default.

Does insulin resistance change how many calories I should eat?

Not in the way it is often claimed. There is no established adjustment factor that lowers your calorie target because of an elevated HbA1c. The DiRECT remission trial used a very low calorie phase of roughly 825 to 853 kcal a day in people with type 2 diabetes and achieved remission in 46% at one year. What an elevated HbA1c should change is supervision, not arithmetic — particularly if you take insulin or a sulfonylurea.

Weight Loss Timeline Calculator — How Long It Really Takes

Why the 3,500-calorie rule is wrong

You have met this rule even if you have never heard its name. A pound of fat holds about 3,500 calories, so cut 500 a day and lose a pound a week. It is arithmetic, it is memorable, and it is one of the most misleading things in popular nutrition.

The problem is the assumption underneath: that your metabolism sits still while your body shrinks. It does not. A smaller body costs less to run. Less tissue to maintain, less mass to carry up the stairs. So the 500-calorie gap you opened in January is not a 500-calorie gap by June — the target moved toward you.

Applied honestly over a year, the 3,500-calorie rule predicts 52 pounds lost. Almost nobody loses 52 pounds. That gap between the promise and the scale is where most people conclude they have failed, when what actually happened is that a bad model met a normal body.

The better rule of thumb

Modelling by Kevin Hall and colleagues, published in The Lancet, replaced it with something that matches what actually happens:¹

Every permanent reduction of about 10 kcal a day produces roughly one pound of eventual weight loss.
Half of that change arrives in about a year.
95% takes about three years.

So a sustained 500 kcal reduction heads toward about 50 lb — but over years, not one year, and along a curve that flattens rather than a line that keeps falling.

Your plateau is usually not a failure of willpower. It is the arithmetic catching up with you, and it was always going to.

What the curve above is doing

The chart is not a straight line because bodies do not lose weight in straight lines. It uses Hall’s exponential approach to a new steady weight, anchored on the roughly one-year half-time from the same work. That is why it drops steeply in the first months and then bends.

One simplification I should own: the full model has fast and slow components, and a single exponential fitted to the one-year half-time reaches about 88% of the change by three years rather than Hall’s 95%. I have anchored on the half-time because it governs the first year, which is the part you will actually experience and judge the plan by. The curve is therefore slightly conservative at the far right — which is the direction I would rather err in.

Understanding the shape changes what you do with it. Rapid early loss is not evidence that your plan is brilliant, and the flattening is not evidence it has stopped working. Both are expected. The people who quit tend to quit at the bend, having been promised a straight line by a calculator.

Which equation the tool uses, and why

With no body fat percentage entered, it uses Mifflin-St Jeor⁴ — the most accurate of the general population equations, and the same one behind the basal metabolic rate calculator.

Enter a measured body fat percentage and it switches to Katch-McArdle, which works from lean body mass instead of total weight.² Fat tissue is metabolically quiet; muscle and organs are not. Two people at the same weight with different composition genuinely have different requirements, and Katch-McArdle sees that where weight-based equations cannot.

One caveat: it is only better if the body fat figure is real. A measured result from DEXA, or a reasonable estimate from calipers or a good bioimpedance scale, will improve the answer. A guess will make it worse, and the tool will not know the difference. If you are unsure, leave it blank.

Why the number this tool gives you may still be too high

Here is the part most calculators never admit, and it is worth more than the rest of the page.

A predicted maintenance requirement is a population average, and the activity multiplier is where it goes wrong. For a 100 kg, 45-year-old woman of average height, the choice between “sedentary” and “moderately active” moves the answer by about 600 kcal a day — more than every other input on the form put together. And people reliably pick the wrong one. Two gym sessions and a dog does not make you moderately active.

What the trials actually prescribed

The studies that produced real, measured weight loss did not calculate anyone’s TDEE and subtract from it. They handed out fixed intakes.

Look AHEAD — 1,200–1,500 kcal/day for people starting under 250 lb, and 1,500–1,800 for those above it.⁵
DiRECT — 825–853 kcal/day as total diet replacement for three to five months, medically supervised, which produced remission in 46% at one year.³

If this calculator hands you 2,700 kcal as a weight-loss target for a 100 kg body, hold that against the 1,200–1,500 that Look AHEAD used for the same person. The results panel now shows both, side by side, for exactly this reason.

If you are not losing weight on the number a calculator gave you, the calculator is wrong. It does not know how much you move, and it is guessing about the biggest variable in the sum.

So use the computed figure as a ceiling to work down from, not a target to eat up to. The most common single error is choosing an activity level one notch too high, which is why this page now defaults to sedentary rather than moderate. If in doubt, take the lower option: the cost of underestimating your requirement is slower progress, and the cost of overestimating it is none at all.

About the HbA1c field

I want to be direct about what this does and does not do, because a version of this tool circulating elsewhere gets it wrong.

Entering an elevated HbA1c does not change the calorie maths, and it should not. You will find calculators that quietly shrink your target on the grounds that insulin resistance makes weight loss harder. I know of no evidence supporting a specific adjustment factor, and the best trial evidence points the other way: DiRECT used a very low calorie phase of roughly 825–853 kcal a day in people with type 2 diabetes and achieved remission in 46% at one year.³ These are not patients who needed a gentler deficit.

What an elevated HbA1c genuinely changes is supervision. If you take insulin or a sulfonylurea, eating substantially less at an unchanged dose is how hypoglycaemia happens. That is a conversation with your prescriber before you start, and it is the subject of a separate page.

So the field is here as a clinical flag rather than a multiplier. Enter it and the tool tells you what the number means and what to do about it.

If your HbA1c is above 6.5% The remission calculator shows how your profile compares with the trials, and what weight loss they actually required.

Check my profile →

What to do with all this

Judge the plan by months, not weeks. Week-to-week noise is mostly water and gut contents.
Expect the bend. Slowing down is the model working, not the diet failing.
Decide in advance what happens at the plateau — accept the new weight, or make a further sustainable change. Deciding in the moment usually means quitting.
Protect muscle on the way down. Adequate protein and resistance training twice a week. Lean tissue is metabolically expensive, which is exactly why you want to keep it.
Know your actual intake before setting a target — most people underestimate it by several hundred calories.

Find out what you are really eating Three ordinary days, honestly logged. Free, no account, and it tracks 40+ nutrients as well as calories — which matters more once you are eating less.

Track free →

Glossary

BMR — basal metabolic rate: What your body burns at complete rest: breathing, circulation, temperature, brain and organs. Usually 60–70% of daily energy use.
TDEE — total daily energy expenditure: BMR multiplied by an activity factor. Roughly what you need to hold your current weight.
Metabolic adaptation (adaptive thermogenesis): The fall in energy expenditure during weight loss — partly because there is less of you, and partly because the body becomes somewhat more efficient. The first part is much larger than the second, and it is the part this tool models.
Lean body mass: Everything that is not fat: muscle, bone, organs, water. The metabolically active portion, and what Katch-McArdle works from.
Mifflin-St Jeor: The general-population BMR equation using weight, height, age and sex. The most accurate of the widely used formulas when body composition is unknown.
Katch-McArdle: A BMR equation using lean body mass alone: 370 + (21.6 × lean kg). More accurate when body fat percentage is genuinely measured, worse when it is guessed.

This is educational information, not medical advice, and it does not make me your physician. The tool applies a floor of 1,200 kcal for women and 1,500 for men because intakes below that are difficult to make nutritionally adequate and should be medically supervised — as they were in the trials that used them. If you take insulin or a sulfonylurea, speak to your prescriber before reducing what you eat.

Dr. Gily Ionescu, MS MD

References

The model and every figure above come from these.

Hall KD, Sacks G, Chandramohan D, et al. Quantification of the effect of energy imbalance on bodyweight. The Lancet. 2011;378(9793):826–837. https://doi.org/10.1016/S0140-6736(11)60812-X Source of the trajectory model: the 10 kcal/day per pound rule of thumb, the one-year half-time and the three-year 95% figure, and the demonstration that the 3,500-calorie rule overpredicts.
Katch FI, McArdle WD. Nutrition, Weight Control, and Exercise. 3rd ed. Philadelphia: Lea & Febiger; 1988. Source of the lean-body-mass BMR equation used when a body fat percentage is entered.
Lean MEJ, Leslie WS, Barnes AC, et al. Primary care-led weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. The Lancet. 2018;391(10120):541–551. https://doi.org/10.1016/S0140-6736(17)33102-1 Used for the 825–853 kcal/day intervention and 46% remission at one year in people with type 2 diabetes.
Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition. 1990;51(2):241–247. https://doi.org/10.1093/ajcn/51.2.241 The default BMR equation when no body fat percentage is supplied.
The Look AHEAD Research Group. The Look AHEAD study: a description of the lifestyle intervention and the evidence supporting it. Obesity. 2006;14(5):737–752. https://doi.org/10.1038/oby.2006.84 Used for the prescribed intakes: 1,200–1,500 kcal/day under 250 lb and 1,500–1,800 above it.

Weight Loss Timeline Calculator

Your trajectory

What this actually looks like