What Cycle Lab Does With a Zodiac Degree
Pick a degree, choose what counts as an event, and see two centuries of sky measured against it, with a baseline so the number cannot flatter you.
Astrology makes a lot of claims about particular degrees. An eclipse fell on your Sun. Three total eclipses in a row landed in the same corner of Leo. Venus keeps stationing on the same handful of points. Some of those claims are true and interesting, some are true and unremarkable, and some are what a small sample looks like when you stare at it hopefully. Telling them apart is a matter of counting, and counting two hundred years of eclipses and stations by hand is exactly the job nobody does before posting. I did not do it either, until I had built something to do it for me.
Cycle Lab does the counting. You give it one point in the tropical zodiac, an orb, a span of years and the event classes you care about. It hands back every catalogued event inside that orb, with the date, the kind of event, the exact degree it fell on, and the signed distance from your point, so you can see which side of it each one landed. The list is plain and sortable and it is complete. There is no editorial pass picking out the interesting ones, because the interesting ones are the ones you would have picked, and that is the problem this tool exists to remove.
The second thing it hands back matters more. The identical filter is run again at all 360 whole degrees, and your count is shown beside the median, the range and your percentile. That comparison is what turns an impressive-sounding number into one you can judge. 63 events in two centuries means nothing on its own. Once you know the typical degree collects 51 and the busiest collects 67, the fair description is a little above average, which is a much less exciting sentence and the correct one.
Every setting is also written into the address bar as you change it, in place, with no page load and no request. That makes the link the evidence. A screenshot shows a number and asks you to trust whatever settings produced it. A permalink carries the degree, the classes, the orb and the span, so the person you send it to runs the same computation and sees the same baseline, and if they want to widen the orb and watch the claim dissolve, they can do that too.
- A degree, an orb, a span of years and the event classes you care about
- Every matching event with its exact degree and signed offset
- The same search run at all 360 whole degrees, as a baseline
- A URL that reproduces the whole query for anyone you send it to
How the 1900–2100 Event Catalogue Is Computed
Every event in the catalogue is derived from the ephemeris; none of it is copied from a published table.
Positions come from astronomy-engine, the MIT-licensed ephemeris this whole site runs on, and each event is stored as an apparent geocentric ecliptic longitude for the equinox of date. That is the tropical position an astrologer reads: where the body appears from Earth, measured against the moving frame the signs are defined in. It is deliberately not the heliocentric longitude, which is what a naive call to an ephemeris library gives you and which would put every event in the wrong place by an amount that drifts slowly over the two centuries.
Eclipses are found by walking the library's global eclipse search forward from 1900 until it passes 2100, recording the Sun's longitude at the peak of each solar eclipse and the Moon's at the peak of each lunar one. Stations are found by following each planet's unwrapped longitude on a step short enough that no retrograde loop can hide between two samples, marking every turning point, then narrowing in on the instant the motion reverses. Both solar and lunar eclipses are kept at every magnitude, down to the faint penumbral ones, along with both the retrograde and the direct station of each planet.
That gives 4,572 events between 1 January 1900 and 31 December 2100: every solar and lunar eclipse, plus both stations of Mercury, Venus, Mars, Jupiter, Saturn, Uranus, Neptune and Pluto. Each one is packed as three integers, an event type, an hour count from 1900 and a longitude in hundredths of a degree, and the whole file compresses to roughly 27 KB, which is smaller than most photographs. I floor the hour rather than round it, because rounding pushed the roughly 2% of events that fall after 23:30 UTC onto the following day, and the hit list prints dates.
That size is the whole reason the privacy claim holds. The catalogue is one static file. Your browser fetches it once, and every query after that, the 360-degree baseline included, is answered locally with no request back to any server. Change the orb, change the degree, tick another class: nothing goes out. The degree you type never leaves your machine, and you do not have to take my word for it. Open the network panel in your browser, change a setting, and watch nothing happen.
- Every solar eclipse, whether total, annular or partial
- Every lunar eclipse, whether total, partial or penumbral
- Both stations of all eight planets, Mercury through Pluto
- 1 January 1900 to 31 December 2100, with dates in UTC
The same positional model, the same tropical frame and the same checks against published values apply to every calculator on the site. The methodology page sets out how the numbers are produced and where I compare them with reference tables.
Why Every Result Comes With a Baseline
A hit count on its own can be made to sound remarkable for any degree in the zodiac, and I know because I have done it.
Here is the trap the baseline panel exists to close. Choose a degree. Widen the orb until the count feels satisfying. Write the sentence. It reads like a finding every time, and the number went up because you opened the window wider; the sky at that point did nothing at all. The only defence is knowing what the same window produces at every other degree, which is why the baseline is computed with the identical filter on every query, rather than quoted from a table I worked out once and stopped checking.
This is not hypothetical. Two of my own claims died to exactly this check. The first was a line that eclipses clump around particular degrees; a commenter ran a null model, and they do not. The second was more seductive. Three total solar eclipses within a couple of degrees of 20° Leo looked like a pattern worth an essay, until a scan of all 360 degrees turned up fifteen other triples just as tight. Neither claim survived meeting its own baseline, and neither would have been caught by staring harder at the hit list. 20° Leo is the degree the tool opens on for exactly that reason.
So the readout leads with the comparison: your count, the median count, the full range across all 360 degrees, and where you sit in it. The verdict words are deliberately dull. A degree is toward the bottom of the range, on the quiet side, ordinary, on the busy side, or toward the top of the range, and that top band only starts at the 97.5th percentile, because on 360 samples anything looser is what nine degrees in ten look like on some setting. It is never called significant, and never called evidence of anything. If that reads as deflating, it is the right amount of deflating. A tool that tells everyone their degree is special is telling them nothing.
Reading a Hit List Without Fooling Yourself
Three habits that keep a two-century scan from turning into a horoscope.
Look at the orb before you look at the count. With a two-degree orb over 1900 to 2100 a typical degree collects about fifty events, for no deeper reason than that there are four and a half thousand events and only 360 degrees to share them between. Double the orb and you roughly double the count, for your degree and for every other one. If a number only became interesting after you widened the window, the window is what changed, and the fair thing is to report the setting alongside the number. The permalink does that for you whether you meant it to or not.
Look at the classes too. Mercury alone contributes more than twelve hundred stations across the span, so leaving it ticked while you test an eclipse claim drowns the eclipses in Mercury. The event classes are kept separate precisely so a question about eclipses can be asked about eclipses and nothing else. Narrow the filter to what your claim is actually about, then read the baseline for that narrower filter, which the tool recomputes on the spot rather than reusing from the wider one. This is also why the page opens with eclipses only and every station class one tick away.
Last, keep straight what a station or an eclipse at your degree is and is not. It is a real, checkable fact about where a body stood on a date, computed here to a hundredth of a degree, and you can take it to any other ephemeris and get the same answer. It is not a measurement of anything about you, and this page makes no claim that it is. What Cycle Lab can settle is whether the pattern someone described is really in the data. What it cannot settle is what the pattern means, and I have tried to build it so that it never pretends otherwise.
If you came for your own chart rather than a degree in the abstract, the natal retrograde calculator works from a birth date and puts the same kind of population baseline beside every placement it finds.