Methodology
How each calculator was chosen, implemented, checked and reviewed — including what it cannot tell you.
How a calculator gets onto this site
Every tool follows the same path: pick the question, find the method with the best evidence behind it, implement it in plain JavaScript, check it against worked examples, and write the page so that someone who does not trust it can see the whole calculation. If a method cannot be implemented honestly — because it needs an input users cannot supply, or because the published equation is not reproducible — the tool does not ship.
Where the formulas come from
Sources are ranked. Clinical guidelines and position stands from professional bodies come first; systematic reviews and meta-analyses next; then the primary studies those reviews rest on; then reference values from public databases such as USDA FoodData Central. Commercial websites are not sources, and neither are summaries that do not name the study they came from.
Where a guideline gives a single figure — 400 mg of caffeine a day, 7–9 hours of sleep — the calculator uses it but shows the surrounding range and the population it applies to. Where a formula was derived in a specific group, such as the weight-based caffeine limits for children, the page says so.
Units and rounding
Calculators work in metric internally and convert at the edges, so switching between metric and US units never changes the underlying calculation. Height converts between centimetres and inches, weight between kilograms and pounds, volumes between litres and fluid ounces. Displayed values are rounded for readability, and tables state the rounding where it could confuse — percentages that do not quite add to 100, for example.
Testing
Each calculator is checked against independently worked examples: a 200 mg dose with a five-hour half-life must leave 100 mg after five hours; Mifflin-St Jeor for a 35-year-old man of 178 cm and 78 kg must return 1,723 kcal; 10,000 steps at a 0.73 m stride must come out at 7.3 km. The site is then rendered at mobile and desktop widths to check that nothing overflows, every internal link is crawled, and each page is checked for the content blocks a reader needs: the calculator, the formula, an interpretation table, sources and the last-updated date.
Review cycle
Pages are reviewed at least twice a year, and immediately when a guideline changes. The current review was completed in November 2026. Reviews re-check the source, the arithmetic and the layout; the date at the foot of each page is the date of that review, not the date the page was first written.
Limitations
- Estimates, not measurements. Prediction equations carry 10–15% error at best. A TDEE estimate can be 300 kcal out; a stride-length estimate can be 10% out.
- Population averages. Most equations were derived in specific groups, often North American or European adults. They travel reasonably well but not perfectly.
- Self-reported inputs. Sleep onset, activity level and portion sizes are all easy to overestimate. The calculators cannot correct for that.
- Health conditions. Pregnancy, kidney disease, thyroid disorders, sleep apnoea, eating disorders and many medications change the answer, and none of them are input fields here.
Reporting a problem
If a result looks wrong, the most useful message includes the page, the inputs you used, the number you expected and the number you got. Send it through the contact page; verified errors are corrected and noted on the page.