Weight Percentile Calculator

Adults 20+ use the NHANES 2017-2018 anthropometric reference data published by the CDC and the National Center for Health Statistics. The means and standard deviations are 90.6 kg / 22.9 kg for US males and 77.8 kg / 21.7 kg for US females. Children ages 2 to 19 use the CDC 2000 weight-for-age LMS parameters (NCHS, May 2000): the Box-Cox power L, median M, and generalized coefficient of variation S at each whole-year age, which preserve the right-skew of pediatric weight instead of forcing a normal fit. The full NHANES report is at cdc.gov/nchs/data/series/sr_03/sr03-046-508.pdf and the CDC growth charts are at cdc.gov/growthcharts.

It computes a z-score (z = (weight in kg minus the cohort mean) divided by the cohort SD) and runs that through the standard normal cumulative distribution function. The CDF is implemented with the Abramowitz and Stegun 7.1.26 rational approximation to erf, which is accurate to within 1.5e-7 across the entire real line. That precision is well below the noise floor of a population reference, so the percentile you see is limited by the cohort data, not the math. The output is clamped to a 0.01 to 99.99 range so extreme z-scores render cleanly.

For children, the percentile compares weight with an age-and-sex reference. For adults, this calculator uses a broad sex-specific NHANES reference and does not make a same-age comparison. Weight alone says nothing about body composition or distribution. Pair it with a height-normalized measure and discuss sustained changes with a clinician.

Population-level weight data thins out at the extremes. NHANES publishes adult tables through roughly age 80 and beyond with diminishing sample sizes; we cap at 100 because actuarially almost no published reference data exists past that age. At the lower end, the CDC weight-for-age growth chart starts at age 2; infants 0 to 24 months are tracked against the WHO growth standards, which use a different reference cohort and a different shape (more right-skew). Adding WHO infant references would broaden the calculator's scope without changing what it does, so we keep the floor at age 2.

Bands map directly to percentile ranges. Very low is under the 5th percentile, low is 5 to 25, below-average is 25 to 40, average is 40 to 60, above-average is 60 to 75, high is 75 to 95, and very high is above the 95th. About 1 in 20 people fall in each tail (very low or very high) under a normal distribution; those tails warrant a closer look from a clinician because they sit outside the typical 90 percent of the population. The middle 20 points around the median (40 to 60) is the band where roughly 1 in 5 people land and is what most people mean by 'average for my age and sex.'