Korean pop culture, in numbers

The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first

When the same Korean title is written up on four Southeast Asian Wikipedias, which one gets there first — and how long do the others take?

Of the 59 Korean titles we hold, 25 have an article on all four editions. The Indonesian Wikipedia wrote first, or joint-first, for 24 of them. The Vietnamese Wikipedia — the largest of the four on every measure we checked — did so for 8.

Who writes first

A title counts for an edition if that edition's article is the oldest, or ties for oldest. 7 of the 25 titles are ties, which is why the numbers below add up to more than 25.

WikipediaFirst or joint-firstArticlesActive editors
Indonesian24 of 25790,7844,690
Vietnamese8 of 251,304,0014,726
Thai1 of 25186,4342,963
Malay1 of 25440,8402,018

The obvious explanation does not hold. A bigger encyclopaedia with more editors would be expected to cover anything sooner, but the Vietnamese Wikipedia leads on articles, on active editors and on total edits, and it is not the one that writes about Korean titles first. We can rule that explanation out. We did not measure what replaces it, and we are not going to guess.

How long the rest take

Counting from the first edition to write a title to the last, the median gap is 8 months, and it runs from 0 to 83 months.

Removing any single title from the 25 leaves this median at 8 months. That rules out one way it could be fragile — it does not show that it is stable. The check is a jackknife, which is known to understate how much a median moves, so a swing it finds is real while a swing it misses may be one it cannot see. The method and its limits are here.

TitleFirstthenthenthenthenSpread
Signal (South Korean TV series)2016-02IndonesiaVietnam +2Malaysia +9Thailand +8383 months
Itaewon Class2019-10IndonesiaVietnam +4Malaysia +17Thailand +3131 months
The King: Eternal Monarch2019-09IndonesiaVietnam +5Malaysia +8Thailand +3131 months
Hotel del Luna2019-06IndonesiaVietnam +1Malaysia +4Thailand +2929 months
Big Mouth (South Korean TV series)2022-01IndonesiaVietnamMalaysia +6Thailand +2222 months
All of Us Are Dead2020-06IndonesiaMalaysia +19Vietnam +20Thailand +2020 months
Gyeongseong Creature2022-09IndonesiaThailand +14Vietnam +16Malaysia +1717 months
Hospital Playlist2020-02IndonesiaVietnam +1Thailand +2Malaysia +1616 months
Reply 19882015-12IndonesiaVietnamMalaysia +4Thailand +1616 months
Reborn Rich2022-01VietnamMalaysia +9Indonesia +10Thailand +1111 months
Little Women (South Korean TV series)2021-12IndonesiaMalaysia +7Vietnam +8Thailand +99 months
Crash Landing on You2019-07IndonesiaVietnam +4Thailand +4Malaysia +88 months
Twenty-Five Twenty-One2021-11IndonesiaVietnam +3Thailand +3Malaysia +88 months
Goblin (TV series)2016-09IndonesiaVietnam +3Malaysia +3Thailand +88 months
Lovely Runner2023-10IndonesiaThailand +4Malaysia +6Vietnam +88 months
Parasite (2019 film)2019-05IndonesiaVietnamThailand +3Malaysia +77 months
Alchemy of Souls2022-03IndonesiaMalaysia +4Vietnam +5Thailand +77 months
Start-Up (South Korean TV series)2020-09IndonesiaVietnam +1Malaysia +4Thailand +77 months
Train to Busan2016-08IndonesiaVietnamThailand +6Malaysia +66 months
Queen of Tears2023-11IndonesiaThailand +3Vietnam +4Malaysia +55 months
Vincenzo (TV series)2021-02IndonesiaVietnamThailand +1Malaysia +44 months
Extraordinary Attorney Woo2022-05IndonesiaVietnam +2Thailand +2Malaysia +22 months
Marry My Husband2023-11IndonesiaVietnam +2Thailand +2Malaysia +22 months
A Killer Paradox2024-01IndonesiaVietnamThailand +1Malaysia +11 month
Squid Game2021-09IndonesiaVietnamThailandMalaysia0 months

Read a row across, not down. The order of the country names is the order those editions wrote the title, and +n is months after the first of them.

We checked our own dates before using them

A first-revision date says when an article has existed under its current name, not when the title was first written about. Move or merge an article and the earlier revisions go with it, leaving a date that is far too recent.

So we tested every date against the months that article was actually read. If a page was read before its own first revision, the date has been erased. Of 153 dates we could test this way, 0 failed. 83 could not be tested, because those articles have no month with any reading at all.

⚠ The Korean Wikipedia is not on this page, and that is why. Its article for Squid Game reports a first revision of 2025-01 for a title from 2021, and we hold no Korean pageview data to check the rest of its dates against. An edition we cannot test is not an edition we will publish.

Why 25 titles and not 59

Of 59 titles, 25 have an article on all four editions with a first-revision date we could read. A title missing from one edition is left out entirely, because counting it would make that edition look slow when it is absent.

What this page cannot tell you

Not why. We can rule out one explanation — the edition that writes first is not the largest — but we did not measure what does explain it.

Not readers. This counts when an article was written, not when anyone read it. An audience can arrive long before an article does, or never.

Not the Korean Wikipedia. Its first-revision dates are unreliable for this and we have no pageview data to check them against, so it is not on this page at all.

First revisions — Wikipedia API. Reads — Wikimedia Pageviews API, human traffic only, monthly. Edition sizes — Wikimedia siteinfo, read 2026-08-15.

Card set — free to share

Five cards, 1080×1350. Every figure on them is in the table on this page. Take them.

The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first — card 1 of 5The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first — card 2 of 5The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first — card 3 of 5The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first — card 4 of 5The same Korean title reaches four Wikipedias in a fixed order, and the biggest one is not first — card 5 of 5

14 seconds, with sound

Fourteen seconds, read aloud: of twenty five Korean titles written up on all four Southeast Asian Wikipedias, the Indonesian one wrote first for twenty four. The Vietnamese edition, the biggest of the four, wrote first for eight. Size does not explain the order. This counts when an article was written, not when anyone read it. Free to repost with the address on it.

All our short films — each one built from a table on this site

What we wrote from this data

How this was measured

We report the median rather than the mean. A handful of very large values would pull a mean away from where most of the sample actually sits, and in reads-per-title data those large values are the norm, not an error. Stability of a median is checked with a jackknife — remove one observation, recompute, repeat for each (Quenouille 1949, Tukey 1958). The Korean figure is read against a control drawn the same way, because a movement that appears in the control as well is not a fact about Korea. The order itself is measured by the median place each edition takes across the titles, not only by how often it is first, because counting firsts alone cannot tell a queue from noise.

What this cannot tell you

A median tells you where the middle sits and nothing about the shape around it, so it should be read next to the range or the full distribution, not alone. The jackknife understates how much a median varies, because removing one observation barely moves a median, and it is not consistent for the median (Miller 1974). A swing it finds is therefore real; a swing it misses is not evidence of stability. Establishing stability would need a delete-d jackknife or a bootstrap (Efron 1979), which we have not built. A control rules an explanation out; it does not identify the one that replaces it. We say what the control removes and stop there. A first-revision date records when an article has existed under its current name, so moving or merging an article erases the earlier one; we test every date against the months that article was read and drop any edition we cannot test that way. Twenty-five titles is a small sample, and every one of them reached a Netflix chart, so this says nothing about Korean titles that did not.

The order, edition by edition

EditionMedian placeTimes last
ID10
VI21
TH312
MS312

ID was never last. The commonest order was id>vi>th>ms, which happened 10 times out of 6 distinct orders seen. One order being commonest is not the same as it being the rule, so we give the count beside it.