How can you predict your adult height (or your child's future height)?

Predicting adult height is one of the most common questions parents and curious adults ask, and the honest answer is that no method is perfectly accurate. The best-established techniques can narrow a child's likely adult height to within roughly 5 centimeters (about 2 inches) of the final result, but every prediction carries a margin of error that widens or narrows depending on the child's age, the method used, and the quality of the data fed into it. This guide walks through the main methods in use as of 2026, explains how each works, compares their accuracy, and flags the mistakes people most often make when trying to forecast how tall someone will grow.

The Direct Answer: What Actually Works

Also worth reading: How accurate is a bone age X-ray for predicting a child's adult height? · How to calculate your final adult height? · growth hormone therapy height outcomes?

There are four practical ways to predict adult height, ranked here from least to most accurate. First, the mid-parental height formula uses only the biological parents' heights and requires no measurements of the child at all. Second, doubling methods — such as doubling a boy's height at age 2 or a girl's height at about 18 months to 2 years — offer a quick folk estimate with wide error bars. Third, growth-chart tracking plots a child's height against standardized percentile curves maintained by organizations like the CDC and WHO; because children tend to track along a stable percentile from toddlerhood onward, the chart gives a reasonable projection of where they will land at maturity. Fourth, bone-age assessment using an X-ray of the left hand and wrist, interpreted with systems such as Greulich-Pyle or newer skeletal-age models published in journals like Nature, is the gold standard for clinical accuracy because it measures biological maturation directly rather than inferring it from chronological age.

In practice, clinicians combine these: a pediatrician will plot the child's height on a growth chart, calculate mid-parental height as a sanity check, and order a bone-age X-ray only when there is a specific concern — extreme shortness, extreme tallness, or growth that has fallen off its established curve. For an ordinary healthy child with no red flags, growth-chart tracking plus the parental formula is usually sufficient, and it costs nothing beyond a tape measure and a well-child visit.

Why Height Prediction Is Possible at All

The reason any of this works is that human growth follows a predictable sequence. Children grow rapidly in infancy, slow through middle childhood at a fairly steady 5 to 6 centimeters per year, accelerate during puberty with a growth spurt that can add 8 to 12 centimeters per year at peak velocity, and then stop when the growth plates (epiphyseal plates) at the ends of the long bones fuse. In girls, this fusion typically completes between ages 14 and 16; in boys, between ages 16 and 18, though some boys add small amounts of height into their early twenties. Once the plates close, no further linear growth is possible, which is why adult height is effectively fixed by late adolescence.

Genetics supplies the bulk of the explanation for where a person lands. Twin and family studies consistently attribute roughly 80 percent of variation in adult height to inherited factors, with the remaining 20 percent shaped by nutrition, chronic illness, sleep, physical activity, and hormonal status. Because the genetic contribution is so dominant, parental heights are genuinely informative — but not deterministically so. Siblings sharing both parents routinely differ by 10 centimeters or more in adult stature, which illustrates why formulas built on parental height alone can only produce a range, not a point estimate. The statistical concept underlying all prediction methods is regression toward the mean combined with tracking: a child who is tall for her age at age 3 is statistically likely to be tall for her age at 18, even though the exact percentile can drift by several points over childhood.

Method One: The Mid-Parental Height Formula

The mid-parental height (sometimes called the Tanner) method is the simplest calculation available. For a boy, take the mother's height plus the father's height, add 13 centimeters (about 5 inches), and divide by two. For a girl, take the sum of the parents' heights, subtract 13 centimeters, and divide by two. The result is the child's predicted adult height, and the accepted practice is to present it plus or minus 8.5 to 10 centimeters (roughly 3.5 to 4 inches), because that is the observed spread among children of identically sized parents.

Worked example: a father who is 180 cm and a mother who is 165 cm have a combined height of 345 cm. Their son's predicted height is (345 + 13) / 2 = 179 cm, with a realistic range of about 170 to 188 cm. Their daughter's predicted height is (345 − 13) / 2 = 166 cm, with a range of about 157 to 175 cm. An imperial-units variant adds 5 inches for boys and subtracts 5 inches for girls before halving the average of the parents' heights.

The strengths of this method are obvious: it takes thirty seconds, requires no equipment, and works from birth onward. Its weaknesses deserve equal attention. It assumes the parents' measured heights are accurate (self-reported heights are often inflated by 1 to 2 cm), it ignores whether either parent had an unusual growth pattern themselves, and it cannot account for a child with a medical condition affecting growth. Treat it as a rough corridor rather than a forecast.

Method Two: The Doubling Rule and Growth-Chart Tracking

The doubling rule holds that a boy's height at age 2, doubled, approximates his adult height, while a girl's height at around 18 months to 2 years, doubled, does the same for her. A 2-year-old boy measuring 87 cm would be projected to reach about 174 cm. Pediatric sources including Parents magazine describe this as a popular approximation, and it is indeed directionally useful — but its error margin is larger than the mid-parental formula, often exceeding 7 to 10 centimeters, because a single measurement at one moment captures nothing about growth tempo or pubertal timing.

Growth-chart tracking is more defensible. Standard CDC and WHO charts plot height-for-age from birth to 20 years across percentile bands, with deciles marking the bottom and top 10 percent of the population. The key empirical finding, noted in reference literature on growth charts, is that children generally maintain their percentile position over time: a child tracking the 60th percentile at age 4 will very often be near the 60th percentile as an adult. To use the method, plot the child's current height and age on the chart, read off the percentile, then follow that percentile line to the adult axis at age 20. Repeating the measurement every 6 to 12 months and confirming the child stays on their curve substantially improves confidence. A child whose percentile drops persistently — say from the 50th to the 15th over two years — warrants medical evaluation regardless of absolute height, since crossing percentile lines is a stronger warning sign than being short.

Method Three: Bone Age X-Rays and Skeletal Maturity Models

For clinical-grade prediction, the reference standard remains assessment of skeletal maturity. A single radiograph of the left hand and wrist is compared against standardized atlases — most famously Greulich-Pyle — to assign a "bone age" that may differ from chronological age. A 13-year-old with a bone age of 11 has more remaining growth potential than his calendar age suggests; a 13-year-old with a bone age of 15 has less. Combining bone age with current height and chronological age allows calculation of predicted adult height using tables such as the Bayley-Pinneau method or, more recently, percentage-of-adult-height models. Research published in Nature has refined these skeletal-age prediction models, expressing a child's current height as a percentage of expected mature height based on skeletal maturation, which improves precision particularly during puberty when growth tempo varies widely between individuals.

Accuracy figures reported for bone-age-based prediction generally fall within ±3 to 5 cm for children assessed in middle childhood, degrading somewhat during the pubertal growth spurt when velocity peaks. The procedure itself is quick and low-radiation — the effective dose of a hand X-ray is a tiny fraction of a chest X-ray — but it requires interpretation by a trained reader, and inter-observer variability of several months of skeletal age is common. Insurance coverage varies; out-of-pocket cost in the United States typically runs from about $100 to $400 for the imaging and reading, though many plans cover it when ordered for a documented growth concern.

Comparing the Methods Side by Side

No single table row tells the whole story, and the methods are not rivals so much as layers. A sensible approach stacks them: start with the parental formula for a ballpark figure, confirm the child tracks a stable percentile on standard charts, and escalate to bone-age imaging only if the numbers disagree sharply or if growth deviates from expectations. AI-assisted tools now marketed to consumers essentially automate this stacking, combining parental inputs with serial measurements to output a probability distribution rather than a single number — a genuine improvement in framing, though the underlying physics of growth limits how much any algorithm can tighten the interval.

Common Mistakes People Make When Predicting Height

The most frequent error is treating a prediction as a promise. Every method above produces a central estimate surrounded by uncertainty, and presenting a child with "you will be exactly 178 cm" sets up disappointment or false reassurance. The second mistake is relying on self-reported parental heights, which skew high; measure both parents with a stadiometer or against a wall mark if precision matters. Third, people often ignore pubertal timing. Two 12-year-old boys of identical height may differ enormously in adult outcome if one is pre-pubertal and the other is mid-growth-spurt — this is precisely the situation where bone-age assessment earns its keep, and where simple formulas fail worst.

A fourth mistake is overreacting to a single off-curve measurement. Measurement technique alone introduces 0.5 to 1 cm of error, and normal children can dip briefly during illness. What matters is trend across multiple visits spaced at least six months apart. Fifth, some families attribute short stature to "late bloomers" in the family without verification; constitutional delay is real, but assuming it without checking bone age risks missing treatable conditions such as growth hormone deficiency, hypothyroidism, celiac disease, or effects of chronic conditions. Finally, there is the medication question: research summarized by PsyPost links childhood ADHD stimulant medication to slight reductions in adult height and weight — on the order of 1 to 2 cm in some cohorts — which is worth discussing with a prescriber but should not prompt abrupt discontinuation, since untreated ADHD carries its own substantial risks.

When to Act and When to Simply Wait

Most children need no intervention at all — just routine measurement at well-child visits and plotting on a standard chart. Action thresholds worth knowing: referral to a pediatric endocrinologist is commonly recommended when a child's height falls below the 3rd percentile or above the 97th, when height crosses two or more major percentile lines downward, when growth velocity falls under about 4 to 4.5 cm per year in mid-childhood, or when predicted adult height diverges markedly from the mid-parental target range. Girls showing no breast development by age 13, and boys showing no testicular enlargement by age 14, meet criteria for evaluation of delayed puberty, which overlaps heavily with growth concerns.

Timing matters therapeutically as well as diagnostically. Growth hormone therapy, when indicated, works far better when started early; the same dose produces more gained centimeters in a 7-year-old than a 12-year-old, and options become sharply limited once growth plates fuse. Conversely, for families considering interventions to reduce adult height in exceptionally tall adolescents — now rare — timing before plate closure was historically essential. The practical takeaway is that prediction and action belong together: the earlier an abnormal trajectory is identified, the larger the menu of safe options, which is why annual plotted measurements beat sporadic curiosity-driven checks.

Costs, Tools, and Practical Steps in 2026

For a family starting from zero, the cost of a reasonable prediction is essentially zero. Measure the child barefoot against a flat wall with a rigid book or stadiometer, record height and date, repeat every 6 to 12 months, plot results on a free CDC or WHO chart (both are downloadable), and compute the mid-parental formula as a cross-check. Numerous free online calculators implement both methods; paid apps and AI services charging $5 to $30 per month add convenience, trend alerts, and probabilistic ranges, but they run on the same inputs and cannot exceed the accuracy ceiling imposed by biology. If a concern emerges — off-curve tracking, extreme percentiles, or discordance between methods — the next step is a pediatric visit, potentially followed by a bone-age X-ray costing roughly $100 to $400 out of pocket in the US, or less with insurance. Outside the US, public health systems in most developed countries cover growth evaluation when referred by a general practitioner.

One final calibration note: adult height is fixed, but health metrics that depend on it are not. Guidance from NICE advises adults to keep waist size below half their height to reduce cardiometabolic risk, and BMI classifications use height as a denominator — so once your adult height is settled, it becomes a fixed input for lifelong health monitoring rather than a question to be answered. That reframing is useful: predict height once, carefully, and then spend the rest of your attention on the variables you can actually change.