87% of Korean chart runs have no gap in them at all

Squid Game has been in India’s top 10 for 43 weeks spread across four years. Crash Landing on You took 96 weeks of calendar to accumulate 72 weeks in Japan. Those are the cases that get written about, and they gave us a question we had not answered: is a scattered chart life normal?

It is not. It is rare, and here is the count.

Nearly nine in ten are one unbroken block

A cell is one title, with its season label, on one chart, in one country. There are 11,181 Korean cells in the five years of files. 7,730 of them last two weeks or more.

Read the three medians together. The typical Korean chart life is three weeks long, spanning three weeks, with an unbroken run of three — the same number three times, which is what it looks like when a title arrives, stays, and leaves once.

Only 631 cells — 8.2% of the 7,730 — are scattered enough that their span is at least twice the weeks they were actually present. Those are the ones that make good headlines, and they are the exception.

The other finding is the one-week cell

We left one-week cells out of the table above on purpose: their span and their streak are both one week, so counting them would make “unbroken” true by definition rather than by measurement.

But their number is worth having. There are 3,451 of them — 30.9% of all Korean cells. Nearly a third of the times a Korean title reaches a country’s top 10, it is there for one week and never comes back to that chart.

That is not a failure. A country’s top 10 has ten places and 52 weeks a year, so 520 places a year in each of 93 markets; getting one of them for one week is still getting one.

Why we had to measure this before writing about the long ones

The number Netflix publishes on each row — weeks in top 10 — is cumulative rather than consecutive. It adds one on every consecutive week without exception, and after a gap it keeps counting 98% of the time instead of restarting.

So a large counter can mean either of two very different things, and the field alone cannot tell you which:

• a long unbroken stay — which is what 87% of Korean runs are, and
• a scattered return over years — which is what the famous examples are.

Every run length on this site is computed from the weeks themselves for exactly this reason. The counts above, the medians and the row totals behind them are on the weeks-counter page; what a departure from a chart looks like is on the exit page.