Why two sites give you different birth charts

You put the same birth data into two calculators and got two different rising signs. One of them is wrong, or both are right about different questions. There are four things that cause this, and you can check three of them yourself in about a minute.

1. The birth time, which is almost always the answer

The rising sign moves about one degree every four minutes. An hour of error moves it roughly fifteen degrees, which is half a sign, and a sign boundary is never far away. So if one site was given 14:00 and the other 14:30 — or if one of them silently assumed noon because you left the field blank — you have your answer, and the rest of this page is academic.

Check what time each site actually used. A calculator that does not show you is asking to be trusted about the single most sensitive input in the calculation.

2. The timezone the site guessed

Your birth time was recorded on a local clock. To place the planets, a calculator has to turn it into a universal instant, and that means knowing what offset that clock was keeping on that day in that town. After about 1970 this is a solved problem. Before that, it is worth reading what the timezone database nearly every site relies on says about itself, because it is blunter than any astrology site will be:

The pre-1970 entries in this database cover only a tiny sliver of how clocks actually behaved; the vast majority of the necessary information was lost or never recorded. […] Most of the pre-1970 data entries come from unreliable sources, often astrology books that lack citations and whose compilers evidently invented entries when the true facts were unknown, without reporting which entries were known and which were invented. […] In short, many, perhaps most, of the tz database’s pre-1970 and future timestamps are either wrong or misleading.

Read that twice. The timezone data the whole industry relies on for historical charts was, in places, reconstructed from astrology books whose authors guessed — and those guesses now come back to astrology software as authoritative-looking offsets. Two sites can therefore disagree by an hour without either being careless: they resolved an ambiguous record differently, and neither of them can prove it was right.

The United States and Canada before 1970 are the thinnest part of the record, which is unfortunate, because that is a large share of the people asking this question.

This site shows you the offset it used, says how far it can be trusted, and lets you override it. If a chart elsewhere disagrees by exactly one hour, compare the two offsets before anything else.

3. The house system

There are a dozen or so ways to divide a chart into houses, and they do not agree. The same planet can sit in the eighth house under Placidus and the ninth under whole sign, because the two systems answer different questions about what a house is. Neither is a mistake.

What a house system never changes is which sign a planet is in. So if two charts disagree about your Sun’s sign, the house system is not the reason — go back to point one. If they agree about every sign and disagree about the house numbers, this is almost certainly it. The seven systems here, and what each one divides, are worth two minutes if you have never chosen one deliberately.

4. The obliquity convention, which is probably not your problem

This one is included because it is real, it is obscure, and it is the sort of thing that gets blamed for differences it did not cause.

The Earth’s axial tilt wobbles slightly — nutation, about nine arcseconds — so there are two defensible values to compute houses from: the mean tilt, or the true tilt of that moment. Swiss Ephemeris, which most of the industry validates against, uses the true one. A calculator using the mean value lands within about a thirtieth of an arcminute of the same answer in temperate latitudes, which is a fifth of a second of birth time and invisible to you.

It only becomes visible near the Arctic and Antarctic circles, where the ecliptic lies almost flat against the horizon and a small difference in tilt is amplified into most of an arcminute. This site got that wrong at first and measured it: 0.966 arcminutes of rising degree at polar latitudes, down to 0.033 once the convention was corrected. If you were not born above about 60 degrees of latitude, this is not what happened to your chart.

What is not the reason

The planetary positions themselves. Every serious calculator draws on the same underlying planetary theory, and the differences between implementations are far below anything astrology distinguishes. This site’s positions agree with NASA’s published ephemeris to within 0.287 arcminutes across 1900 to 2050, and its angles and house cusps agree with Swiss Ephemeris to within 0.033 arcminutes. For scale, one arcminute of rising degree is four seconds of birth time. Nobody’s birth certificate is that good.

How to find out which one it was

1. Compare the birth time each site used, including whether either of them substituted noon.
2. Compare the UTC offset. A one-hour disagreement is a daylight-saving disagreement almost every time.
3. Compare the house system. If the signs match and the house numbers do not, you have found it.
4. If the two rising degrees differ by almost exactly 180 degrees, one calculator has a bug rather than a convention — the two points where the ecliptic crosses the horizon are opposite each other, and picking the wrong one is an easy mistake to make above the Arctic Circle. This site made it, and fixed it.

Calculate your chart here if you want a fifth opinion — with the offset, the confidence and the house system all stated on the page, so you can tell which of the four applies.