A birth time can be worked out backwards — not from a document, but from what has already happened. First a description of your character, your reactions and your looks narrows down which ascendant signs are plausible. Then, for each of those signs, we calculate at what hour it was actually rising over your city on that day. The result: a day that was twenty-four hours long shrinks to a few.
The honest limits of this method, before you start.
The method has two parts, and they are very different in how reliable they are. It matters that you know which is which.
The ascendant — the sign that was rising over the eastern horizon at the moment of birth — does not describe your character. It describes the way you make contact with the world: first impression, posture, how you react to something unfamiliar, even build. Which is why a description of those things can rank the twelve signs: which are likely, and which are almost certainly not it.
This is the weakest link in the method, and we are not hiding it. Seven questions do not define a person. They only say which end to start searching from.
The second part does not depend on your answers at all. Knowing the date and the city, we can calculate at what hour each sign was rising over the horizon that day. That is pure astronomy, accurate to a fraction of an arcsecond.
The result is where the two parts meet. Not "your ascendant is Leo", but: "if Leo, you were born between 14:12 and 16:31". The second can be checked; the first can only be assumed.
There is one more step after the result, and it is the interesting one. The chart turns at roughly one degree every four minutes — so two guesses at a birth time put the angles of the chart in two different places. But where the planets stood on the date of an event does not depend on your birth time at all: it is simply what it was.
So the question can be turned around: which birth time makes the big events of your life land in the years when a slow planet was sitting on the matching point of the chart? A marriage is checked against the partnership axis, a change of job against the home-and-career axis, money against the money axis. The time with the most matches wins.
This step narrows the result from a few hours to about half an hour — but only with at least four dates. With fewer it produces an interval that proves nothing, and the page says so plainly instead of looking confident.
The most common advice online is "just put 12:00". That is exactly what most calculators quietly do when the time field is empty.
We measured what that costs, on 1106 real charts with a known birth time. Replace the real time with noon in the same data, and the ascendant lands in a different sign in 93 cases out of 100.
So a "noon chart" is not an approximate chart. It is somebody else's. And nothing warns the person that anything is wrong: it all looks convincing.
In practice the time turns up more often than people expect. Four sources are worth going through first — faster and more accurate than any hypothesis.
Working out an unknown birth time backwards: not from a document, but from what has already happened. First the plausible ascendant signs are narrowed down, then for each one we calculate at what hour it was really rising in that place on that day.
It is a hypothesis, not a document. The questionnaire gives a range of a few hours, not a minute. If you know roughly which part of the day it was, the range narrows about fourfold. The optional check against life events narrows it further, to about half an hour — but only if you have at least four exact dates. An exact time is always worth looking for first, in the birth record or a hospital file.
About five minutes: the date and city of birth, a few questions about what you already know, and seven questions about you. A provisional result appears after the first few answers, and progress is saved in your browser — you can close the page and come back later.
There is no account, no password and nothing to confirm. We ask for an email address at the first step so the finished result can be sent to you; the result itself appears on the page straight away. We never ask for a phone number. Only the time windows go into the email — the date, the city and your answers stay in your browser and are sent nowhere.
Only partly — which is exactly why appearance is one question out of seven here, and not the whole basis. A description like "Taurus rising" fits a great many people, so it is not something to draw a conclusion from on your own.
Events with exact dates test a birth time independently of how you rate yourself: where the planets stood on the date of an event does not depend on your birth time, while the angles of your chart do. Four dates are the minimum; from five the result becomes reliable. With fewer, the check still produces an interval, but it proves nothing — and we say so outright.
A rough month is enough. Slow planets take weeks, sometimes months, to cross a single degree, so being a few weeks out barely changes the answer: in testing, shifting every date by two months moved the result by minutes. Being out by a year, though, ruins it — the year is worth checking.