A 100% Promise on the Fifth Pull, a Ledger That Reads 9%: KRAFTON's Four Years, Four Fractures

The number sits in my ledger in red ink: 100. Beside it, in black: 9. Both describe the same event — one from KRAFTON's announcement, the other from the Korea Fair Trade Commission's investigation. In 2026, when the NewJeans collaboration brought the PUBG X NewJeans Loot Pack and the PUBG X NewJeans Premium Bundle to the PUBG: Battlegrounds store, players were told that after four consecutive failures the fifth pull would guarantee the set blueprint at 100 percent. Many players bought more than five Premium Bundles and never received it. PUBG's team later confirmed the Premium Bundle never carried the so-called bad-luck protection, and that certain in-game disclosures were inaccurate. My first model was wrong, which is how I knew the data was honest.

What began as a storefront dispute has become a balance sheet on institutional credibility. Esports taught me that a patch note is a transfer window with faster consequences — rules change, and within hours hundreds of thousands of accounts change their decisions. Loot boxes work the same way, only faster: the credit card decides, and the evidence survives as a screenshot.

I work as a team data consultant. In 2026, hand-tagging 1,344 shots across 132 Malaysia Super League matches, I learned every number needs its provenance, or it is a mood rather than a model. For loot boxes, that provenance is the published probability table. Since South Korea amended its game industry law in 2026, probability disclosure has been mandatory. The story is no longer about ethics alone; it is about regulatory paperwork.

In June 2026, Yonhap reported that the FTC fined KRAFTON 2.5 million won for providing misleading information about item drop rates in PUBG: Battlegrounds. The file places two numbers side by side: a guarantee after four failures, and an actual rate of 9 percent. KRAFTON later refunded roughly 1.1 billion won to about 380,000 buyers and issued in-game compensation worth approximately 9.8 billion won. Because the company corrected the information and paid compensation on its own initiative, the sanction was limited to a fine.

Now the arithmetic. Under the advertised promise, a player's expected cost was five pulls. At a genuine 9 percent, the expected number of independent pulls is 1 divided by 0.09, roughly 11.1. The gap between promise and reality is about 2.2 times — the hidden price that never appeared on any receipt. The pattern was never in the averages; it was hiding in the outliers who refused to behave, such as the player who bought seven times and walked away empty seven times.

A second thread from the NewJeans collaboration surfaced the same year. Some players combined member face skins with unauthorised, offensive outfits and circulated images and videos; the matter grew more sensitive because Haerin and Hyein were minors at the time. KRAFTON and management company ADOR said they would act against uses outside the collaboration's original purpose, and later restricted combining NewJeans face skins with certain outfits. Part of the community asked a fair question: does responsibility rest only with users, or also with a customisation system that never pre-registered any limits? I did not measure the crowd; I measured whether the action arrived before the incident or after it went viral.

An older accounting came into public view in 2026. In its IPO filing, KRAFTON disclosed that it provides technical services to Tencent-operated Peacekeeper Elite in China and receives service fees for it. Previously the company's position was that PUBG Mobile and Peacekeeper Elite were separate products. PUBG Mobile shut down in China in May 2026 over licensing issues, and players were moved to Peacekeeper Elite, a title with altered gameplay, visuals and content. Place the service-fee line beside the project-explanation line in the same document and the earlier argument loses room.

The most recent chapter began at a showmatch and ended in a question of community trust. At PUBG Asia Stars 2026, Korean streamer Soopi of Gen.G alleged that two Vietnamese players — Himass of Anyone's Legend and TanVuu of The Expendables — were using information from opponents' livestreams. Organisers removed both from the rest of the tournament, adjusted scores, increased stream delay, cancelled the third match day and split the prize pool. On 23 September 2026, KRAFTON published its investigation: outside information, particularly other players' livestreams, had been used to form judgments and build tactics; the conduct was classified as stream sniping, both accounts were permanently banned, and both players were barred from any official PUBG Esports event operated or approved by KRAFTON, including PGC, PGS and PNC. No further violations were found in other cases.

Born in Bangladesh and working in Malaysia, I can see that the Vietnamese PUBG community did not stay silent. More than 4.1 million signatures were collected in support of Himass and TanVuu. Độ Mixi, PewPew, Rambo, DjChip and Ngân Sát Thủ reacted publicly, some announcing they would drop the game. GAM x The Expendables demanded a clear explanation, while Anyone's Legend defended the players' right to be heard. The debate now covers not only guilt but the scale of punishment, the consistency of process and whether the same standard applied to everyone.

This is where I question the ledger's architecture rather than its averages. In the probability case, no rival raised doubt; at Asia Stars, the ladder from suspicion to proof sat entirely with one party: stream data, who was watching where, how long the delay ran and what was on screen mid-match all lived on the publisher's own servers. The body that writes the rules also investigates the complaint and issues the sanction; all our verification ends at a file we never see. Such a conflict of interest is unusual in sports administration, but here it is structural. Between a 2.5 million won fine and 4.1 million signatures lies a gap that says public pressure rises fast while institutional accounting does not.

The second gap is documentary, not nominal. Had a document published six months earlier specified exactly which data counts as proof of stream sniping, the punishment could have been even harsher and still credible. The system's weakness is not the severity of sanctions; it is the standard that should be promised before them.

I am registering a time-stamped forecast: within twelve months, at least one Asian regulator will mandate real-time disclosure of genuine loot-box probabilities, and PUBG Esports will publish its stream-sniping evidence standard before the next PGC. If I am wrong, the results will answer for me.

What this model cannot see: the FTC file contains KRAFTON's compensation figures, but whether departed players returned, or how much income a streamer lost night after night, appears in no document. I counted only the numbers whose provenance was written down. The ledger began as 1,344 shots; it went on to measure loot-box probability, and it ended as a question I could not unask — when one company owns the game, the tournament and the data, where is a player's last address for appeal?