Methodology
How the projection is calculated, and how far to trust it.
Every assumption the calculator makes, where each one comes from, and a side-by-side test against the model behind the NIH Body Weight Planner — including where this calculator comes out too optimistic.
Last updated
The daily loop
Most weight loss calculators do one division: the deficit, into the energy stored in a kilogram, gives a rate, and the rate gives a date. This one runs a simulation instead, one day at a time for up to 365 days. Each simulated day does five things:
- Recalculates BMR from yesterday's weight, using Mifflin-St Jeor. This is the step that makes the curve bend: a lighter body costs less to run, so the same intake becomes a smaller deficit.
- Builds total expenditure by multiplying BMR by an activity factor for everyday movement, then adding any planned exercise, itself recalculated from the day's weight.
- Applies the NEAT penalty once you have spent 30 consecutive days in a deficit, taking 5% of BMR off expenditure for the drop in unconscious movement that sustained dieting brings.
- Converts the day's gap into weight at 7,700 kcal per kilogram.
- Checks the day against the safety thresholds and the plateau rule below.
Alongside it, the calculator draws the fixed-rate projection most other tools give you: the day-one deficit held constant for the whole horizon. The distance between the two lines is the error that assuming a constant deficit builds up over time.
Every assumption, and where it comes from
Some of these figures come straight from published research. Others are my own modelling choices, informed by research but not taken from it. The table says which is which.
| Parameter | Value | Basis |
|---|---|---|
| BMR | 10 × kg + 6.25 × cm − 5 × age, plus 5 (male) or -161 (female) | Published: Mifflin-St Jeor (1990) |
| Activity factor | 1.2 to 1.75 | My own ladder. It is adapted from standard activity multipliers but narrower, because exercise is counted separately |
| Exercise | (MET − 1) × 3.5 × kg ÷ 200 per minute | Published: Compendium of Physical Activities. The engine subtracts the one MET you would have burned at rest anyway |
| Energy per kg lost | 7,700 kcal | The conventional figure. Real loss varies with how much is fat and how much is lean tissue (Hall, 2008) |
| NEAT penalty | −5% of BMR after 30 deficit days | My estimate. The research documents the effect (Levine, 2002; Rosenbaum and Leibel, 2010) but gives no single onset day or size |
| Diet breaks | 3 deficit days refunded per maintenance day, ×0.5 for each further consecutive day | My own model. The direction comes from the MATADOR trial; the numbers do not |
| Intake floors | 90% of BMR; 1,200 kcal (female) or 1,500 kcal (male) | Commonly cited clinical figures (Jensen et al., 2014) |
| Rate of loss | Caution above 0.75%, danger above 1% of bodyweight a week | Informed by Garthe et al. (2011) |
| Plateau | Under 0.1 kg of change over a trailing week, for 7+ days | My definition, chosen to sit below what a bathroom scale can reliably detect |
Compared with the NIH Body Weight Planner
The NIH Body Weight Planner is the most rigorous free projection tool available. It is built on a model by Kevin Hall and colleagues at the National Institute of Diabetes and Digestive and Kidney Diseases, published in The Lancet in 2011, and it is the benchmark I would want any calculator, including this one, measured against.
| This calculator | NIH Body Weight Planner | |
|---|---|---|
| You enter | A daily intake; it returns the date | A goal and a date; it returns the intake |
| Body model | Total weight, with BMR from Mifflin-St Jeor | Fat and lean mass tracked separately |
| Energy per kg lost | Fixed at 7,700 kcal | Varies with the fat and lean mix being lost |
| Activity | Everyday-movement factor, with exercise counted separately from METs | One physical activity level, 1.4 to 2.5, that includes exercise |
| Horizon | Up to 365 days | Years, including the maintenance phase |
| Extras | Plateau timing, diet breaks, per-day safety flags, macros, export | Calories to maintain the goal once reached |
The same scenario, three ways
Hall's paper includes a worked example: a 100 kg sedentary man who cuts his intake by 480 kcal a day and holds it there. I ran the same scenario through this calculator's engine. The paper does not state height or age, so the reference run uses 180 cm and 40 years. Across heights from 165 to 190 cm and ages from 25 to 60, the result moves by 0.6 kg at most.
| Method | Lost in year one |
|---|---|
| Fixed 3,500 kcal per pound rule | 22.8 kg |
| This calculator | 14.2 kg |
| Hall model (NIH planner) | about 11 kg |
The engine closes about 70% of the gap between the fixed rule and the NIH model. It is still about 3.2 kg more optimistic than the NIH model after a year, so a date from this calculator is more likely to be too early than too late. The NIH model also has the weight levelling off near 75 kg over about three years. This calculator stops at one year, so it cannot show that far.
Why the gap remains
- Expenditure falls more slowly per kilogram lost. In Hall's model, each kilogram of bodyweight costs about 24 kcal a day to maintain. In this engine it costs the Mifflin-St Jeor weight term times the activity factor: 12 kcal at the sedentary setting. As weight comes off, the engine's expenditure falls about half as fast, so the deficit stays open longer. This is the largest single difference.
- Eating less makes digestion cost less, and the engine ignores this. Digestion uses about 10% of what you eat, so a 480 kcal cut also saves about 48 kcal a day on digestion. That saving shrinks the real deficit from the first day.
- Adaptation arrives as a single step. Hall's model builds metabolic adaptation into its equations. This engine adds it all at once on day 30, so the first month of any projection has no adaptation in it.
These gaps are why the calculator's date is best read as the earliest plausible date rather than the most likely one. If your plan is tight, check it against the NIH planner as well.
What the model cannot see
- Individual variation in BMR. Mifflin-St Jeor is the most accurate of the common equations, and it still misses measured BMR by more than 10% for roughly one person in five.
- What you actually eat. The engine assumes the intake you enter is the intake you eat. Self-reported intake usually runs low, and that error is often larger than anything the engine models.
- Water. Glycogen, sodium, training and the menstrual cycle move the scale by more than a week of fat loss. The projection is a smooth trend, and daily weigh-ins will scatter around it.
- Medical factors. Thyroid function, medication, pregnancy and metabolic conditions are all outside the model. None of this is medical advice.
If a figure on this page disagrees with what the calculator shows you, that is a bug, and I would like to hear about it on the contact page.
Sources
- The BMR equation the engine recalculates every dayMifflin MD, St Jeor ST, et al. (1990). A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition 51(2):241–247.
- How accurate Mifflin-St Jeor is for an individualFrankenfield D, Roth-Yousey L, Compher C (2005). Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. Journal of the American Dietetic Association 105(5):775–789.
- Physical activity level multipliers, adapted here to exclude exerciseFAO/WHO/UNU (2004). Human energy requirements: report of a joint FAO/WHO/UNU expert consultation. FAO Food and Nutrition Technical Report Series 1.
- MET values and the MET-to-calorie conversionAinsworth BE, Haskell WL, et al. (2011). 2011 Compendium of Physical Activities: a second update of codes and MET values. Medicine & Science in Sports & Exercise 43(8):1575–1581.
- The energy content of lost weightHall KD (2008). What is the required energy deficit per unit weight loss?. International Journal of Obesity 32(3):573–576.
- Non-exercise activity as a large and variable part of expenditureLevine JA (2002). Non-exercise activity thermogenesis (NEAT). Best Practice & Research Clinical Endocrinology & Metabolism 16(4):679–702.
- The size of adaptive reductions in expenditureRosenbaum M, Leibel RL (2010). Adaptive thermogenesis in humans. International Journal of Obesity 34(Suppl 1):S47–S55.
- Maintenance breaks partly reversing adaptationByrne NM, Sainsbury A, et al. (2018). Intermittent energy restriction improves weight loss efficiency in obese men: the MATADOR study. International Journal of Obesity 42(2):129–138.
- The guideline intake ranges behind the absolute floorsJensen MD, Ryan DH, et al. (2014). 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults. Circulation 129(25 Suppl 2):S102–S138.
- Faster loss costing more lean mass, behind the rate thresholdsGarthe I, Raastad T, et al. (2011). Effect of two different weight-loss rates on body composition and strength and power-related performance in elite athletes. International Journal of Sport Nutrition and Exercise Metabolism 21(2):97–104.
- The thermic effect of food, about 10% of intakeWesterterp KR (2004). Diet induced thermogenesis. Nutrition & Metabolism 1(1):5.
- The model behind the NIH Body Weight Planner, and the worked example compared hereHall KD, Sacks G, et al. (2011). Quantification of the effect of energy imbalance on bodyweight. The Lancet 378(9793):826–837.
- The planner itselfNational Institute of Diabetes and Digestive and Kidney Diseases (2011). Body Weight Planner. National Institutes of Health.