Korean pop culture, in numbers

Stars

Jimin's birthday is 13 October, and 25 other Korean stars share it — day 137 of 366

Fans are organising around 13 October. We hold full birth dates for 9,249 Korean entertainers, so we counted every calendar day. That date holds 26 people, ranking 137th of 366. Two of the busiest single days are artefacts we can name.

The table this is written from: /kpop-birthdays

#bts#kpop#jimin#birthdays#wikidata#method#data-gaps

Fans have started organising around 13 October, the birthday of BTS’s Jimin, weeks ahead of the date. We hold full birth dates for 9,249 Korean entertainers, so the question of whether that date is special is one we can answer by counting rather than guessing.

It is not special. 26 people in our panel share it — Jimin and 25 others. The average calendar day holds 25.27. Ranked against all 366 days, 13 October comes 137th.

That is a finding worth publishing precisely because it is dull. A date matters to the people who care about it, and no amount of counting changes that; what counting can do is stop anyone claiming the calendar itself is unusual there.

The months are not even, though

Month People per day Against average
January 28.94 1.14×
February 30.56 1.21×
March 27.71 1.10×
April 24.70 0.98×
May 24.84 0.98×
June 21.80 0.86×
July 23.16 0.92×
August 24.39 0.96×
September 24.87 0.98×
October 23.97 0.95×
November 24.30 0.96×
December 24.94 0.99×

February holds 40% more birthdays per day than June. Months are compared per day here, not by total, because February has the fewest days and a raw count would make it look empty.

Two of the busiest days are artefacts, and we can name them

These are the busiest single days in the panel:

Day People Against average
1 January 48 1.90×
28 February 48 1.90×
11 February 41 1.62×
20 March 41 1.62×
27 February 40 1.58×
3 March 38 1.50×

1 January is not a real cluster. When only a birth year is known, editors sometimes record 1 January, so that date collects people whose actual birthday is unknown. 28 February sits at exactly the same height, which is the sort of coincidence that deserves suspicion rather than a paragraph of explanation.

So we ran the test that matters: delete every day above 1.5 times the average — five days, 218 people — and re-count. February stays first at 29.08 per day, June stays last at 21.80. The threshold chose which days to drop, not us, and the ranking does not depend on the suspicious ones. That is the only reason we are willing to print it.

Our own day page says 18, and that is not a contradiction

We already publish a page for 13 October. It says 18, not 26.

Both numbers are right, because the panels are different and the difference is the point:

Page Panel 13 October
Our day pages 7,015 people who have an English Wikipedia article 18
This count 9,249 people with a birth date recorded at all 26

The day pages rank each date’s names by how much they are read, so they can only list people who have an article to read. This calendar count does not rank anybody, so it can include people who have a birth date and nothing else.

We are printing both rather than quietly aligning them, because a reader who finds 18 on one of our pages and 26 on another is entitled to be told which is which — and because a house that silently harmonises its own numbers cannot be checked from outside.

Our own two counts disagree, and that is the honest headline

We published a birthday page earlier built on 1,317 group members. It found March fullest and July thinnest. This panel of 9,249 entertainers finds February and June.

Same method, same source, different panel, different answer. Neither is wrong; they are answers to different questions — one about idol group members, one about Korean entertainers broadly. What it means practically is that anyone quoting “the month with the most K-pop birthdays” should say which panel they mean, because the two do not agree.

We could have quietly retired the older number. Publishing both is more useful, and it is also the more accurate description of what we know.

What this is not

Not a prediction and not a claim about any person. A birth date on Wikidata is what editors recorded.

Not a complete panel. People with only a birth year recorded are excluded rather than placed on 1 January, which is the mistake this article is partly about.

Not an explanation. Birth seasonality, school-entry cutoffs and who ends up with a Wikidata entry at all would each produce a February peak. We cannot separate them and do not try.

Month-by-month tables, including the version with clustered days removed, are on K-pop idol birthdays by month.

This is a count of public records. It is a statistic, not a statement about any individual.

Where these numbers come from

Sources

  • Wikidata — SPARQL — date of birth (P569) for Korean entertainers, taken only where a full year-month-day is recorded. 9,249 people qualified. · https://query.wikidata.org/sparql
  • Star News (스타뉴스) — Front-page headlines archived by us on 9 September 2026 at 05:30 KST, including a post about fan voting ahead of 13 October. Used to date the occasion, never as a source of a number.

What we checked

  • Months are compared per day, not by raw total, because February has fewer days and would otherwise look empty. February is credited 28.2425 days to account for leap years across the panel.
  • The February result survives deletion of every suspicious day. Removing the five days that hold more than 1.5 times the daily average — 218 people — leaves February at 29.08 per day and June at 21.80, the same ranking. The threshold picks the days, not us, and we ran the test because a result that depends on 1 January is not a result.
  • Fixing this article's panel exposed a defect in our own earlier count and we corrected it rather than working around it. People were being de-duplicated by name, which merged different people who share one: 288 names in this panel carry more than one birth date. The panel is now keyed on Wikidata identifiers, and our group-member page moved from 1,295 people to 1,317 as a result.
  • We checked our own earlier count and it disagrees. Our page built on 1,317 group members found March fullest and July thinnest; this wider panel of 9,249 finds February and June. Both are printed, because the disagreement is information about how much either can be trusted.
  • 13 October was looked up after the headline, and its rank was computed against all 366 days rather than against a shortlist. Had it been unusual, that is what this article would say.
  • Our own day page for the same date says 18, and we print that beside our 26 rather than harmonising them. The day pages cover 7,015 people who have an English Wikipedia article, because they rank each date's names by reads; this calendar covers 9,249 people with a birth date recorded at all.

What we left out, and why

  • People whose Wikidata entry records only a birth year, or a year and month. A partial date cannot be placed on a calendar day, and filling it in would have manufactured the clustering this article warns about.
  • Any claim about why February leads. Birth seasonality, school-year effects and the composition of who gets a Wikidata entry would all produce this shape, and we cannot separate them.

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 /kpop-birthdays.

What a company sheet contains, and what it cannot tell you →

← All 215 articlesFront page