Zodiac compatibility, tested on 259 real K-pop groups
Astrology sites tell you which signs go together. Nobody checks. We counted all 4,651 member-to-member pairings inside 259 K-pop groups and set each pair of signs against chance. Every one lands between 0.75x and 1.25x.
The table this is written from: /star-sign-pairs
Search for zodiac compatibility and you will find a great many pages telling you which signs belong together. What you will not find is anyone checking.
K-pop is an unusually good place to check. Groups are assembled deliberately, members’ birth dates are recorded, and there are hundreds of groups. If signs paired off in any way at all, it would show up here.
So we counted.
What we counted
Every group in our list with at least two members whose birth date is on record. Within each group, every member paired with every other member, and the two signs tallied.
| Groups | 259 |
| Members counted | 1,513 |
| Member-to-member pairings | 4,651 |
| Members left out, no birth date | 65 |
That is 78 possible sign pairs, from Aries + Aries through Pisces + Pisces.
The result
The most common pairing is Leo + Aquarius, 93 times. Chance alone would give 82.1.
That is the whole finding. Of the 78 pairs, the one furthest above chance is Leo + Scorpio at 1.25×, and the one furthest below is Virgo + Libra at 0.75×. Everything else sits between.
With counts of this size, gaps of that size are what randomness produces. Run 78 comparisons and a few will land two standard deviations out; that is not a signal, it is arithmetic.
| Pair | Together | By chance | Ratio |
|---|---|---|---|
| Leo + Aquarius | 93 | 82.1 | 1.13× |
| Gemini + Aquarius | 91 | 77.3 | 1.18× |
| Libra + Aquarius | 90 | 75 | 1.20× |
| Gemini + Leo | 86 | 72.6 | 1.18× |
| Leo + Scorpio | 85 | 68.1 | 1.25× |
| … | |||
| Pisces + Libra | 59 | 74.5 | 0.79× |
| Libra + Libra | 25 | 32.1 | 0.78× |
| Virgo + Libra | 49 | 65.3 | 0.75× |
The full 78-row table is here.
No two signs find each other, and no two avoid each other.
Why Leo and Aquarius top the table
Because they are two of the three commonest signs among K-pop members — Aquarius 147, Pisces 146, Leo 138 — and not because anything draws them together. This is exactly why the by chance column exists. A ranked list of raw counts is an invitation to read meaning into the top row. Putting the expected figure beside it takes that invitation back.
The one thing that does bend
Birthdays are not spread evenly through the K-pop calendar. July is the thinnest month, at 0.69 times what an even year would give. We measured that separately, and it is a fact about when people are born in Korea, not about star signs — the same skew shows up in populations that have never heard of Leo.
What this is not
It is not a compatibility reading, and it is not a claim that astrology is wrong about anything. It is a count of who stood on a stage with whom, which is the only part of this question a birth date can settle.
One limitation cuts in our favour and is worth naming for that reason: because a member sits in many pairings at once, the pairings are not independent, so any test treating them as independent overstates how surprising a gap is. In other words the real spread is even less remarkable than the table makes it look. A limitation that only ever helps the writer is one to say out loud.
Counted from recorded birth dates. We do not read star signs, and nothing here says one pairing is better than another.
Where these numbers come from
Sources
- Wikidata — Recorded birth dates (P569) for members of the K-pop groups in our own group list, read 2026-09-06 · https://www.wikidata.org/
What we checked
- The expected column is not decoration. A raw count of pairings invites the reader to find meaning in the largest number, so every observed figure is printed beside what chance alone would give
- Members with no recorded birth date are left out rather than filled in. 65 of 1,581 member slots have no date, and the figures describe the 1,513 we could count
- A person who is in two groups is counted in both, because the question is how often two signs stood on the same stage
- Pairs are not independent of one another, since one member sits in many pairings at once. Any test treating them as independent overstates how surprising a gap is, which makes the null result here stronger rather than weaker
- We make no claim about whether two signs get along. We do not read star signs, and no sentence here says one pairing is better than another
The data behind this
Written from the same data
Take this away — 10 cards
The figures in this article, as images. Free to repost with the address on them.
If you work in this business
Everything on this page is measured from the same weekly Netflix lists, per market and per title. We publish the pages free; what we sell is the same measurement cut to one company's catalogue — including the figures behind /star-sign-pairs and /kpop-birthdays and /star-signs.
What a company sheet contains, and what it cannot tell you →