IVF Gender Selection Success Rates

Last reviewed 2026-10-27

Two different figures get called the success rate. The accuracy of identifying an embryo's sex is close to 100% and is the number clinics advertise. Your chance of a live birth of the sex you chose is much lower, because it is the product of three things: how many euploid embryos you make, the coin flip on each one, and the implantation rate of a single frozen transfer. Only the second figure is about you.

Figure one: test accuracy

When a trophectoderm biopsy is performed correctly and sequenced in a competent laboratory, the sex chromosomes are identified essentially without error. Residual failure modes are sample-handling and labelling problems, not limits of the method, which is why laboratory accreditation and chain of custody are the right things to ask about rather than the accuracy percentage. Clinics quote 99.9% and that claim is fair — about this question.

Figure two: the chance of the outcome you came for

Three things multiply together, and each one can be the one that fails.

How many embryos reach day 5 and test euploid
This is the dominant term, and it falls sharply with the age of the egg provider. The large PGT-A datasets — the reference being Franasiak and colleagues' review of more than 15,000 consecutive trophectoderm biopsies — show euploid rates declining steadily through the thirties and collapsing in the forties. Ask your clinic for their own figure for your age band rather than a published average; a programme that tracks it will have it.
The coin flip on each euploid embryo
Each is about as likely to be male as female. One euploid embryo: about 50%. Three: 87.5%. Five: about 97%. The full table is on the arithmetic page, and it is the single most useful piece of arithmetic in this treatment.
Implantation of the one embryo you transfer
A euploid frozen blastocyst transfer has a meaningfully better chance than an untested one, but it is not a certainty, and it is per transfer. Ask the clinic for their live birth rate per euploid single frozen transfer, by age band — not their pregnancy rate, and not per cycle started.

Multiply them and you have your real number. A 41-year-old who produces one euploid embryo in a cycle is not looking at a 99.9% chance of a daughter; she is looking at roughly a coin flip on sex, and then a transfer that may or may not implant. The same treatment for a 31-year-old with six euploid embryos is a different proposition entirely. Age is not a footnote in this treatment. It is the treatment.

The four numbers to ask every clinic for

Ask for this	Not this	Why the difference matters
Euploid blastocysts per cycle started, for my age band	“Our success rate is 70%”	Without this figure, no probability on this site can be applied to you.
Live birth rate per euploid single frozen transfer, by age	Pregnancy rate	Pregnancy includes losses. You are not buying a positive test.
Per cycle started	Per transfer, or per embryo	Per-transfer rates silently exclude everyone who never reached transfer — the patients most at risk of this treatment failing.
Your own programme's figures, for the last full year	National averages or the laboratory's claims	National data describes other people's clinics. You are choosing one.

Does sex selection lower the success rate?

Not per embryo. A euploid male embryo and a euploid female embryo have no meaningfully different prospects. What selection does is reduce how many of your embryos are eligible for transfer, which lowers the cumulative chance of a live birth from a given cycle and raises the number of cycles some patients need. That is a real cost, and it is the honest answer to “does this hurt my odds”: not the odds of the embryo, the odds of the cycle.

Want your own number worked out?

Send your age, previous IVF history, and AMH or antral follicle count if you have them. We will show you the arithmetic for your situation and what to ask the clinic for. If an introduction we make leads to treatment, we may be paid a referral fee.

Work out my number

What is the success rate of IVF gender selection?

Identifying the sex of a tested embryo is accurate to close to 100%. The chance of a live birth of the chosen sex is much lower and depends on how many euploid embryos the cycle produces, a roughly 50/50 split on each one, and the implantation rate of a single frozen transfer.

How accurate is gender selection with PGT-A?

Essentially always correct when the biopsy and sequencing are done properly. The rare errors come from sample handling and labelling rather than from the method.

Does gender selection reduce my chance of having a baby?

Not for an individual embryo. It reduces how many embryos you can transfer from, which lowers the cumulative chance from one cycle and means some patients need more cycles.

How many embryos do I need for a good chance of the sex I want?

With three euploid embryos the chance that at least one is the sex you want is about 87.5%; with five it is about 97%. With one it is about 50%.

Sources

Franasiak JM et al., The nature of aneuploidy with increasing age of the female partner: a review of 15,169 consecutive trophectoderm biopsies, Fertility and Sterility, 2014
Society for Assisted Reproductive Technology (SART) — US clinic-level outcome reporting, the model for the per-cycle-started figures this page asks you to request — link
Human Fertilisation and Embryology Authority (UK) — national fertility treatment outcome data — link

Related