Industry · music
Korean Celebrities Are Not Born Evenly Across the Year. Idols Alone, We Can't Tell
Among 9,249 Korean public figures, February is the most common birth month, June the least — a 10,000-draw test found this is essentially not chance. Narrowed to 1,317 idols, the same test can't tell the pattern from noise.
The table this is written from: /what-moves-english-reading
Among 9,249 Korean public figures with a full birthdate on record, February is the most common month to be born — 30.6 people per day on average, 21% above what an even 12-way split would produce — and June the least, at 14% below. A 10,000-draw randomness test put the odds of a spread this uneven happening by chance at effectively zero.
Narrow that same question to 1,317 K-pop group members with a known birthdate, and the answer changes. January and March are tied for most common there, July the least — a wider swing in raw terms than the broad list showed. But the identical randomness test, run on this smaller list, could not rule out chance: the odds came back at 1, meaning this specific pattern is statistically indistinguishable from noise.
What we counted
| List | People | Most common month | Least common month | Chance this is random |
|---|---|---|---|---|
| Broad (all Korean public figures) | 9,249 | February (1.21x even) | June (0.86x even) | ~0 |
| Narrow (K-pop group members) | 1,317 | January / March (1.19x even) | July (0.68x even) | 1 (cannot rule out) |
Both counts excluded a handful of dates — 1 January and 28 February chief among them — that show an unusually large cluster of births, 1.5 to 1.9 times a typical day. Those look like recording defaults rather than real birthdates, and running each test again without them did not change which list passed and which did not.
Why we report both, not just the one that “worked”
If we quoted only the broad list, a reader would reasonably assume the same held for K-pop idols specifically — it does not, at least not to a degree this test can confirm. If we quoted only the narrow list, we would be reporting a null result and hiding that the wider population shows a real pattern. Which list a claim is about changes whether it is true.
More single-cause and multi-day patterns are tracked on our what-moves-english-reading hub.
These are statistics, not you. This is not investment advice.
Where these numbers come from
Sources
- Wikidata — SPARQL — Korean public figures (P569 date of birth), full birthdate recorded, CC0. · https://query.wikidata.org/
What we checked
- Broad list: 9,249 Korean public figures. February averaged 30.6 people born per day, the highest of any month (1.21x an even split); June averaged 21.8, the lowest (0.86x). A 10,000-draw resample test (seed 20260904) put the odds of this month-to-month spread arising by chance at effectively 0.
- Narrow list: 1,317 K-pop group members (across 263 groups, 1,581 member slots) with a known birthdate. January and March tied for the most (1.19x even), July the least (0.68x) — a wider swing than the broad list. The same resample test, run on this list, put the odds of this specific spread arising by chance at 1 — meaning the test cannot distinguish this pattern from randomness.
- Both lists exclude days with an unusually large cluster of births (1 January, 28 February and a handful of others averaged 1.5-1.9x a typical day) — those look like recording defaults, not real birthdays, so a version of each test excluding them was also run and did not change which list passed and which did not.
What we left out, and why
- A cause for the broad-list pattern. We did not test explanations (school-year cutoffs, agency scouting seasons, or anything else) — we report that the pattern is unlikely to be chance, not why it exists.
- Extrapolating the broad-list finding onto individual idols or groups. The narrower, idol-specific list is the one that actually failed to clear the randomness bar.
The data behind this
Written from the same data
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 /what-moves-english-reading.
What a company sheet contains, and what it cannot tell you →