5 Nub Theory Myths Everyone Believes (Debunked)

The claim: "Nub theory is 99% accurate" or even "100% accurate" at 12 weeks. You will see this all over forums and prediction websites.

The reality: Those figures are real, but they come from a single small study of 172 pregnancies performed by expert sonographers at the Harris Birthright Centre under ideal conditions (Efrat et al., 1999). They represent a best case, not the typical result.

Larger, more representative studies tell a more honest story:

• Manzanares et al. (2016), n=672: sex assignable in 90.5%, correct in 87.5% of those (PubMed).
• First-trimester review, n=2,314: 90.1% success overall; >95% once CRL exceeds 65 mm (PubMed).
• Chelli et al. (2009), n=312: ~85.7% overall accuracy.

The honest, combined picture is about 85–90% at 12–13 weeks, rising above 95% as the baby grows. That is still genuinely good for such an early estimate — but it is not 99%. For the full breakdown, see our accuracy guide.

The claim: A squarer, blockier skull means a boy; a rounder, smaller skull means a girl. You can read this on any prenatal scan.

The reality: There is no peer-reviewed research supporting skull theory, and there is a clear biological reason why it fails. Human cranial sexual dimorphism does not appear until puberty — the skull features that differ between adult men and women develop under the hormonal changes of adolescence. They are simply not present before birth, so there is nothing to read on a prenatal ultrasound (Healthline).

On top of that, fetal skull shape on a 2D ultrasound changes dramatically with the baby's angle and position, so any "pattern" people spot is interpretation, not anatomy. For the full comparison, see our nub vs Ramzi vs skull breakdown.

The claim: "You can predict your baby's sex from 6 weeks using nub theory." Some websites promise this, confusing nub theory with Ramzi theory.

The reality: The genital tubercle is bipotential — identical in both sexes — until about week 7. It only begins to respond to sex hormones from about week 9, and it is not readable on an ultrasound until roughly week 12, with development largely complete by week 14 (StatPearls embryology).

At 6 weeks there is literally nothing to differentiate. The 70% accuracy figure at 11 weeks is already marginal; at 6 weeks, any "reading" is pure guesswork. This myth likely comes from conflating nub theory with Ramzi theory, which does claim to work from about 6 weeks by looking at placenta position — but Ramzi has never been replicated and performs near chance. See our guide to the best week for nub theory.

The claim: "If the nub is lying flat and parallel, it is definitely a girl."

The reality: A flat nub usually suggests a girl — but "usually" is not "always," and the timing matters enormously. At 11–12 weeks, a boy's nub may not have angled upward yet, so it can look flat like a girl's. This is the most common cause of false girl predictions at 11–12 weeks.

Efrat 2006 found that female accuracy at 12 weeks was only 91.5%, compared with 99–100% for males. The gap closes by 13 weeks, when the boy nub has tilted upward. So a flat nub at 11–12 weeks is a clue, not a certainty — and this is exactly why accuracy improves so sharply with each extra week. For the visual cues, see our boy vs girl nub guide.

The claim: "Nub theory is just as good as a blood test for finding out the sex."

The reality: NIPT (non-invasive prenatal testing), which analyses cell-free fetal DNA in the mother's blood, is approximately 96–99% accurate for sex determination from about 10 weeks (Devaney et al., JAMA 2011). It is a medically validated screening test.

Nub theory, at roughly 85–90% on a clear scan, is a for-fun early estimate — not a diagnostic test. It is excellent for its timing and the most evidence-backed of the early ultrasound methods, but it does not match NIPT on accuracy. The two are complementary, not equivalent: many parents use nub theory for an instant glimpse from the 12-week scan, then confirm with NIPT or the 20-week scan. See our NIPT vs nub theory comparison.

Strip away the myths and the honest picture is this: nub theory is the only early ultrasound gender prediction method with real peer-reviewed evidence behind it. It is roughly 85–90% accurate at 12–13 weeks with a clear image, rising above 95% as the baby grows. It is not 99%, it does not work at 6 weeks, the skull is irrelevant, a flat nub is not always a girl, and it is not as accurate as a blood test.

But within those honest limits, it is genuinely useful — and that is exactly why NubHub pairs an AI model with a human double-check to give you a more consistent reading. Results are back in about 2 hours, the cost is under £40, and every reading is backed by a money-back guarantee. Read more about how our AI works, or start with our complete guide to nub theory.