How it is calculated
The tool uses Tanner's target height, published in 1970 and still the standard reference in paediatrics. It is an average of both parents' height, corrected for sex:
- For a boy: (father's height + mother's height + 13) ÷ 2
- For a girl: (father's height + mother's height − 13) ÷ 2
Where those 13 centimetres come from
They are not arbitrary: 13 cm is the average height difference between adult men and women. When predicting a boy, they are added to the mother to bring her height onto the male scale; when predicting a girl, they are subtracted from the father for the symmetrical reason.
An example with numbers: a father of 178 cm and a mother of 165 cm. For a son, (178 + 165 + 13) ÷ 2 = 178 cm. For a daughter, (178 + 165 − 13) ÷ 2 = 165 cm.
What the ±8.5 cm margin means
The result is not a figure, it is a range. Tanner paired his formula with an interval of ±8.5 centimetres reflecting the variation between siblings in the same family: two children of the same parents do not end up the same height, and that spread is already accounted for here.
Put differently: if target height comes out at 178 cm, the expected adult height lies between 169.5 and 186.5 cm. The chance of ending up outside that range is about 3% — low, but not zero.
That margin is why a paediatrician never says "your child will be 178 cm". They say "he is inside his family lane" or "he is drifting out of it", which is a completely different and far more useful piece of information.
Where the formula fails
Worth knowing before taking the number too seriously. A review published in BMC Pediatrics in 2025 examined the accuracy of Tanner's formula and found a systematic bias: it tends to underestimate adult height in children of very short parents. The practical consequence is that those children are credited with less growth potential than they actually have.
The authors recommend replacing it with the Hermanussen and Cole method (2003), which corrects the bias by accounting for the real correlation between parents' and children's height and for the fact that people tend to partner with someone of similar stature. That formula is not implemented here, and that is deliberate: its full formulation lives in the original paper, and coding it from memory would mean inventing numbers. If your case is parents at the extremes of the distribution, treat this page's result with that reservation.
There are also factors no parents-only formula can see: childhood nutrition, chronic illness, the timing of puberty — an early puberty shortens the growing window — and the secular trend, whereby each generation has been somewhat taller than the last across much of the world.
Other ways to estimate it
Alternative methods exist, each with its drawbacks:
- Doubling height at age 2 (boys) or 18 months (girls). A rule of thumb, very popular and fairly imprecise.
- The Khamis-Roche method, which combines the child's current height and weight with the parents'. More accurate than Tanner, but it needs age-specific coefficient tables that are not in the public domain.
- Bone age, from an X-ray of the left hand. The most reliable method, which is why a paediatric endocrinologist uses it when there is a real suspicion of a growth disorder. Not something done out of curiosity.
When this stops being a curiosity
This page is informed entertainment. What is worth raising with a paediatrician is a growth pattern, not a prediction: a child who changes percentile sharply, who has been flat for a while, or who drifts far from their family lane without explanation. That is seen by following their curve over time at check-ups, not with a formula.