Chinese Gender Predictor

The Chinese gender predictor, also known as the Chinese gender chart, birth chart, or Chinese gender calendar, is a lookup table: 28 rows for the mother's age, twelve columns for the month the baby was conceived, and a boy or girl verdict waiting in every one of its 336 cells. So far, so simple. The catch, and honestly half the fun, is that both numbers are counted the Chinese way. The age is your nominal age, 虚岁 (xūsuì, roughly "shoo-sway"), by which you were already one year old on the day you were born. The month is a lunar month, which drifts against the Western calendar and every few years quietly doubles itself. Read the chart with your Western age and a wall-calendar month and you are, quite confidently, reading somebody else's cell. Most people online are.

Its Chinese name, 清宫图, means "Qing palace chart," a nod to the legend that it once picked princes for emperors. Whether it really did is a story we take up below, and it is worth the scroll.

Here is the full traditional grid, transcribed from the version most widely reproduced in Chinese-language sources and cross-checked against independent copies. Run a finger down your row and you'll notice the pattern refuses to be a pattern: age 21 is eleven girls and a single boy, age 45 flips nearly every month. And we'll let you in on something no other site will show you: the little dots mark cells where circulating copies of the chart disagree with each other. There are 43 of them, about one cell in eight, mostly in the older rows. Somewhere along the photocopy trail, the chart started quarreling with itself, and we think you deserve to see where.

Your row is your nominal age at conception. In the traditional count, a baby arrives in the world already one year old, and everyone's age ticks over together at Chinese New Year rather than on birthdays. Which means the chart may well consider you a year or two older than your driver's license does. Take it as a compliment to your accumulated wisdom. The formula: lunar year of conception, minus lunar year of your birth, plus one. The trap is the New Year itself: someone born on February 1, 1983 was born before that year's Spring Festival, so their lunar birth year is 1982, and every row on the chart shifts with it. This one detail is why two sites will happily hand the same mother two different answers. (Curious what the traditional count makes of your own age? Our Chinese age calculator works it out for any birthday.)

Your column is the lunar month of conception. If you know the conception date, we convert it. If you know your due date, conception is roughly 266 days earlier. From the first day of your last period, add about 14 days. Every conversion we make is shown with your result, so you can check our work, which is not a courtesy the original eunuchs extended.

Leap months and knife-edge dates. Every few years the lunar calendar inserts a thirteenth month, and the chart has no column for it. Tradition has a rule: the first half of a leap month reads as the month it repeats, the second half as the next month. And when a conception date sits within a day of a lunar boundary or of Chinese New Year, a single day would flip the answer. Other calculators silently pick a side. Ours tells you the result could go either way, because on a knife edge that is the only honest answer.

The story goes like this. Seven centuries ago, court astrologers devised a chart for the imperial household, tuned to the lunar calendar and the mother's years, so that an empress might time the conception of a prince. It was kept in the inner palace, guarded by eunuchs, its use forbidden to commoners. When the Qing dynasty fell, the chart vanished in the looting and resurfaced abroad: carried to England, some say. To Austria, say others. Locked in a Taiwanese university vault, says a third telling. It depends entirely on who is telling it, which is your first clue.

We went looking for the palace original, because that is the kind of thing we do here. Arithmetic objects first: the Qing dynasty ran 276 years, 1644 to 1912, which makes a 700-year-old Qing heirloom quite a trick. But the arithmetic is the least of it. The story arrives with two specific, checkable details attached — a royal tomb near Beijing, and a research institute in Beijing where the original is said to be kept — and both of them fall apart the moment you check.

Start with the tomb. Seven centuries back from today lands in the 1320s, and the dynasty ruling from Beijing in the 1320s was the Yuan, whose capital, Dadu, sat under the modern city. The Yuan is, as it happens, the one Chinese dynasty that left no imperial tombs at all. The Yuan Shi, the official dynastic history compiled in 1370, records that the emperors were carried north and buried at a place called 起辇谷 (Qǐniǎngǔ) — thirteen of the fifteen of them. Mongol imperial burial was deliberately unmarked: no mound, no tree, no stele, the ground levelled afterwards so that nothing would show. Qiniangu has never been located. Most scholars place it in the Khentii mountains of Mongolia, more than a thousand kilometers from Beijing, and generations of expeditions and satellite surveys have failed to find it. Not one Yuan emperor's grave has ever been excavated. There is no dig for the chart to have come out of.

Then the institute. The most-copied English version says the original is held, or displayed, at the "Beijing Institute of Science." No such body exists. The Chinese Academy of Sciences, which runs the country's research institutes, has nothing by that name on its roster; the nearest real names are the Beijing Institute of Technology, a defense-engineering university, and the Beijing Institute of Genomics, which sequences DNA. Qing palace paper does have a real address, and it is not a secret: the First Historical Archives of China, a few minutes' walk from the Forbidden City, holds more than ten million Ming and Qing documents across 77 fonds, with the rest of the imperial collection in Taipei. Both are catalogued. Neither lists this chart. In two decades of retelling, nobody has produced an accession number, a catalogue entry, a photograph, or the name of a curator who has seen it.

A third telling relocates the discovery to an underground tomb at Xi'an, which only makes things worse: Xi'an stopped being an imperial capital in 907, four centuries before the chart is supposed to have been buried, and sits eleven hundred kilometers from the institute that is supposed to be holding it.

So who wrote it first? We tried to find patient zero, and this is the honest result: every version cites another version. The scholarly literature is no better — a 2023 review in Actual Gynecology and Obstetrics repeats the emperor's-tomb line and footnotes it to secondary web pages, and even the 2.8-million-birth Swedish study that tested the chart took its grid from a website rather than a manuscript. Not one source we could find names a person, a year, a newspaper, or a shelf mark. The tomb and the institute do not look like a garbled memory of a real find. There is no find underneath them.

What the record does support is more modest, and we will hold it to the same standard we just applied to everyone else. The earliest copies anyone can point to surfaced in Taiwanese and Hong Kong newspapers in the 1970s, and the chart spread from there as a photocopy classic, passed from clinic waiting rooms to kitchen tables to, eventually, the entire internet. We have not laid hands on that newspaper either, and we are not going to pretend otherwise. But it is chronologically possible, it matches how the chart actually traveled, and it does not require a tomb nobody can find inside an institute that does not exist. A beloved piece of modern folklore in an antique costume. We think that is the better story anyway, and it has the advantage of being checkable.

You had to ask. The largest test ever run leaves little wiggle room: epidemiologist Eduardo Villamor and colleagues applied the chart to 2.8 million Swedish births and found it right about 50 percent of the time, at every maternal age and in every conception month (Villamor et al., Paediatric and Perinatal Epidemiology, 2010; the University of Michigan's press release was titled, wonderfully, "Don't paint the nursery just yet"). A coin does the same job and fits in your pocket.

You will see sites claim 70, 90, even 93 percent accuracy. None of them cite a study, because there is none. Some quietly run a little formula instead of the actual chart and hope nobody checks; we checked, and the two disagree on dozens of cells. Add the 43 cells on which published copies of the chart contradict each other, and "93% accurate" dissolves into marketing.

Does that make the chart worthless? Not remotely. Nobody consults it the way they consult a lab report. It is a tradition, a conversation starter, a fifty-fifty thrill for the group chat, and one of the few pieces of imperial theater you can play with on your phone. Enjoy it exactly that way.

And here is the part that is not a coin flip: the lunar math behind your reading is real, and so is what it says about the baby. The zodiac animal and element of the birth year are the genuine article, the same reading a grandmother in Chengdu would give, and they anchor a tradition with actual substance: a Chinese name chosen to fit the child's year and element. The chart is the game. Those are the keepsakes.

Medicine has this one covered. Cell-free DNA screening (NIPT) can report fetal sex from a maternal blood draw at about 10 weeks (ACOG, MedlinePlus/NIH). The routine anatomy ultrasound at 18 to 22 weeks is the standard look (ACOG on ultrasound exams). Talk to your midwife or doctor about which is right for you; nothing on this page is medical advice.

One boundary we hold: this page is for curiosity and celebration, whatever the answer. It is not a tool for choosing, and the chart could not help you choose anyway.