What Is Nub Theory? The Complete Guide to Early Gender Prediction

The “nub” is the genital tubercle — a small bump of tissue that every fetus develops early in pregnancy. It is bipotential, meaning it can become either a penis or a clitoris depending on the baby’s sex.

Up to around week 7, male and female fetuses look identical. From about week 9 the tubercle begins to respond to hormones, and by roughly week 12 the external genitalia become distinct enough to start reading on a clear scan, with development largely complete by week 14 (StatPearls embryology). That maturation timeline is exactly why nub theory needs a 12-week-plus image — before then there is simply nothing reliable to measure.

Nub theory reads the angle of the genital tubercle relative to the baby’s lower body on a mid-sagittal (side-profile) view. This is sometimes called the “sagittal sign” or, more playfully, the “angle of the dangle.”

The thresholds come from the foundational study by Efrat and colleagues at the Harris Birthright Centre (Efrat et al., 1999). Measured against a horizontal line through the lower (lumbosacral) skin surface:

• Boy: the nub is angled more than 30° upward.
• Girl: the nub is parallel or less than 10° (lying flat along the lower body).
• Indeterminate: angles between 10° and 30° can’t be reliably assigned.

Most websites simplify this to “boy if >30°, girl if parallel,” which is broadly faithful — but the honest version keeps that middle indeterminate zone, and that is part of why no one should treat a single reading as certain.

The sweet spot is 12 to 13 weeks. The genital tubercle differentiates enough to read from about week 12, and accuracy climbs steeply with each extra day of growth.

• Before 11 weeks: too early — the nub looks the same for both sexes.
• 11 weeks: readable but unreliable (around 70% accurate).
• 12–13 weeks: the ideal window for nub theory.
• 14+ weeks: still works, and the genitals themselves become easier to see directly.

Crown-rump length (the baby’s measured size) matters too: prediction becomes most reliable once it passes roughly 65 mm.

With a clear 12–13 week image, real-world accuracy is roughly 85–90%, rising above 95% as the baby grows. The famous “98–100%” figures are real but come from a small, expert-operator study — treat them as a best case, not the norm.

Accuracy depends heavily on gestational age, image quality, and the baby’s position — and on the experience of whoever reads the nub. Untrained guessing at home is far less reliable than a trained reading. For the full breakdown, see our guide to nub theory accuracy.

Beyond the angle, the shape of the nub gives clues. A boy nub is often more defined and shows “stacking” (layered lines as the scrotum develops); a girl nub tends to look flatter and lies closer to parallel.

It is easy to mistake the umbilical cord for the nub, or to misjudge the angle when the baby is curled up. Our boy vs girl nub guide walks through labelled examples.

The biggest source of error in nub theory is inconsistent human reading — two people can look at the same scan and disagree. NubHub pairs an AI model with a human double-check to make the reading more consistent.

Peer-reviewed research supports the approach: deep-learning models have classified fetal sex from ultrasound images with accuracy in the 89–98% range on research datasets (Expert Systems with Applications, 2024). Our AI checks angle, shape, stacking, and image clarity together, then a real person double-checks the result before it reaches you — typically within 2 hours. Read more about how our AI works.

Nub theory is the most evidence-backed of the early ultrasound methods, but it is not the most accurate way to learn your baby’s sex overall.

If you want the earliest reliable answer, NIPT wins; if you want a fun early glimpse from the scan you already have, nub theory is the method with real research behind it. Compare them in detail in our methods comparison.