How Korean Music Charts Keep Rewriting Their Rules for K-Pop Fans: A Short History of the Streaming Party
When a K-pop group releases a song, a message goes out from a fan account: play it at these times, on these services. Thousands of people press play together. This is a streaming party, and in Korean it is called 총공 (chonggong), short for "all-out attack".
The party itself is easy to describe. What is more interesting is what happened on the other side. Since 2020, the Korean services and charts that count those plays have changed their rules several times, and each change shows what the chart believed fans could do to it. This article follows those changes in order, because they explain how a Korean streaming number is actually made, and how much weight one deserves.
Four words you will keep meeting
- 총공 (chonggong): short for 총공격, "all-out attack". A streaming party.
- 스밍 (seuming): Korean fan shorthand for streaming. A "seuming list" or "seuming guide" is the schedule and instructions an organiser posts.
- 사재기 (sajaegi): literally "hoarding". Buying or playing in bulk to push a song up a chart, usually discussed as cheating when a company or its agents do it.
- 차트 리포트 (chart report): the page Melon attaches to a ranked song showing its listener numbers over recent windows.
What a streaming party is
Fan-run accounts publish a schedule so that fans play the same song at the same time and can track progress toward a shared goal. Some organisers run giveaways to keep people taking part. English-language coverage gives the fans' reasoning plainly: the number of plays feeds music charts, televised competitions and awards, and artists with the biggest numbers are seen as more successful and get more attention from their companies.
That is the same logic that makes album sales matter, applied to a free action. It costs a fan time rather than an album purchase. It also means the people running the charts have a reason to pay attention to it.
July 2020: the real-time chart is closed
On 6 July 2020, Melon, one of Korea's main streaming services, closed its real-time chart. That chart reflected only the previous hour of listening. Korean reporting at the time described it as easy to distort, both through bulk-buying schemes and through fan streaming parties.
The replacement, called 24Hits, changed four things.
- The window. It reflected the last 24 hours instead of the last hour, updated every hour.
- The unit. It counted one play per person per song within that window.
- The weighting. Usage was counted as 40% streaming and 60% downloads.
- The display. The rank number beside each title, the arrows showing movement, and the live graph were removed, and chart songs were played in shuffled order by default rather than in rank order.
The second change is the one that matters most. Under a one-person-one-count rule, a fan who plays a song fifty times counts once. The only way to move the number is to bring in more different people, not to loop harder. The rest of the changes read as removing targets: no one-hour spike to aim at, no visible rank to watch while trying to move it.
August 2021: the one-hour window comes back, with receipts
A year later, on 9 August 2021, Melon replaced 24Hits with TOP100, the chart it uses today. It combines the last 24 hours and the last hour in equal parts. Between 1 and 7 in the morning, when few people are listening, it uses the 24-hour figure alone.
So the one-hour window returned, which means that timing matters again, within limits. Melon said the new design was built to follow trends quickly while blocking attempts to push rankings up abnormally. It also added something new: every song on TOP100 has a chart report showing how many listeners it had in the last 5 minutes, the last hour and the last 24 hours. A separate HOT100 chart covers recently released songs and reflects the last hour of use.
The report publishes listener counts, which points the same way as the 2020 rule. The reports I found do not restate the one-person-one-count rule for the current chart, so I will not claim it, but the figure on display is people rather than plays. The page pictured below, captured on 7 October 2026, is stamped to the hour (17:00), which fits a chart that recalculates every hour.
October 2023: silent streams
Some fans found a way to keep a party running without listening. They played the song with the volume at zero and carried on with something else.
In August 2023, Circle Chart, which publishes one of Korea's main official rankings, said it would exclude muted streams after finding that muted plays made up more than 7% of average weekly domestic streaming. At certain domestic services alone, that came to about 100 million plays in a single week. It said such use could not reasonably be called listening. Melon applied the same principle from midnight on 12 October that year, excluding plays set to the lowest volume.
Two things to take from this. The number is large, and it shows that non-listening plays were a measurable share of all plays. It does not show that all of them came from fans, and it should not be read that way. And the fact that platforms needed a rule against silent plays suggests that, at least on some charts, volume of plays was worth gaming.
June 2026: the chart leaves Korea
The newest change counts something different. On 1 June 2026, Melon launched a Global-K Chart together with Tencent Music Entertainment of China and Japan's LINE Music. The criteria were published on 28 May. The chart works from the previous day's 24 hours and adds each platform's streaming figures to fan activity on that platform. On Melon, that means listening and downloads together with actions such as following an artist as a fan and pressing like.
The launch figures came from Melon itself. In its first month it said the chart's mobile web page drew about 1.02 million visitors from 160 countries, led by Indonesia, Japan, the Philippines, the United States, Australia and Thailand.
That is a change of kind. The older charts try to count listening and defend it from fans who inflate it. This one counts fandom as part of the score, and does it across borders. Because it takes each platform's own figures, listeners on Tencent Music in China and LINE Music in Japan are part of the count, not only people using Melon. The page pictured below, dated 6 October 2026, shows how the combination works: the group in first place overall was ranked fourth on Melon, fifth on Tencent Music and ninth on LINE Music that day. It topped the combined chart without topping any of the three platforms, and the page lists those three ranks side by side under a heading that reads "ranking by country".
Where overseas streams actually go
A reader outside Korea will reasonably ask whether their streams count in Korea at all.
The answer depends on the chart. Melon's own TOP100 reflects use on Melon. Circle Chart's digital chart, by contrast, draws on data from Melon, Bugs, Genie, Flo, Vibe, Spotify, Apple Music and YouTube, and applies a weighting by service. So a play on Spotify or YouTube can reach one of Korea's main official charts. How much each service counts for I could not find, and the weights are the part that decides how much an overseas play is worth.
Which chart counts what
The word "chart" hides a lot. Here is what the sources above say about four of them, as of October 2026.
| Chart | What it counts | Window |
|---|---|---|
| Melon TOP100 | Use on Melon: streams and downloads, with listener counts published in a chart report | Last 24 hours and last 1 hour in equal parts; 24 hours alone from 1 to 7 a.m. |
| Melon HOT100 | Use of recently released songs on Melon | Last 1 hour |
| Circle Digital Chart | Streams, downloads and background-music use from Melon, Bugs, Genie, Flo, Vibe, Spotify, Apple Music and YouTube, weighted by service; muted streams excluded | Not stated in the sources I could check |
| Melon Global-K Chart | Streaming plus fan activity (following, likes) across Melon, Tencent Music and LINE Music | Previous day's 24 hours |
A song can rise on one and stall on another, and that is not a contradiction. They are different measurements.
What this history does and does not tell you
No platform publishes how much a streaming party moves a ranking, and I found no study that measures it. Anyone giving you a figure for it is guessing.
What the history does show is quieter. Each change was aimed at a specific tactic: the one-hour spike, the repeat play, the silent stream. Platforms do not keep rewriting their charts over something they think has no effect. That is the best evidence available that the effort counts for something, and it is also a sign that the rules will keep moving.
It also means a chart number is a product of its rules. Melon's main chart has been redefined twice since July 2020. A number-one in 2019 and a number-one in 2026 were produced by different counts, so comparing them is comparing measurements, not songs. That is also why each rule above carries its date. A description of how a chart counts is only true on the day it was checked, and these were checked on 2 October 2026.
If you want to take part
Four habits follow from the history.
Check which chart the party is aimed at. Organisers target particular services, and the counting rules differ between them. The organising account's current instructions are the only reliable guide, because the rules above changed within the last few years.
Do not mute. It is the tactic that platforms have specifically ruled out. Muted plays are excluded from both Circle Chart and Melon.
Read the chart report if you can read Korean. Melon publishes listener numbers for the last 5 minutes, hour and 24 hours for every song on TOP100. It is the nearest thing to an outside view of what a party is doing, and it is public.
Keep it in proportion. A party asks mostly for your time, where a fansign draw asks for albums. Its effect is not published, so treat it as taking part in something shared rather than as a lever you can measure.