MSI and Worlds: Six Out of Six and the Trap of a Sample Too Small

On July 13, 2026, in Vancouver, Gen.G defeated T1 in game five of the MSI final. I did not watch that game as a spectator. I sat in a radio studio in Busan, one hand on the microphone, the other on a twelve-column Excel file. The first column held dates. The second held MSI champions. The third held that year's Worlds champions. By game five, I had circled the sixth row in red and typed a single word into the notes cell. "certainly." The studio audience laughed. I did not. In my trade that word is a tool for verification, not a claim to broadcast. MSI was born in 2026 as a gathering of regional champions. Across its first six editions it was famous for one paradox: winning MSI rarely meant winning Worlds. SKT took MSI 2026 and then Worlds 2026, but the SKT of MSI 2026 lost to SSG in that year's Worlds final. RNG won MSI 2026 and then fell to G2 in the Worlds quarterfinals. G2 won MSI 2026 and then lost to FPX in the final. In 2026, MSI was cancelled because of the pandemic. The pandemic did not kill the transfer market, it only stripped away the rules we disguised under financial fair play — and in esports it also stripped away a whole tournament that could be erased from the calendar by a single health notice. After that, the MSI format was expanded. The problem Riot had to solve was not entertainment but representation: a mid-season event only has value if it reflects the real strength of each region. From MSI 2026 through MSI 2026, five consecutive titles went to the LPL or the LCK. Over the same window, the six most recent Worlds titles did the same. The same names kept repeating: JDG, BLG, Gen.G, T1. This is where the story started being sold to the public. The community called it "six out of six" and concluded that MSI had become a reliable preview of Worlds. I spent four nights rebuilding that number from scratch. My Excel sheet is full of formulas, but the answer always sits outside the cells. The result: depending on how you count, the ratio swings from two out of six to six out of six. If "success after MSI" means winning Worlds, the ratio is two out of six. If it means reaching the Worlds semifinals, it rises to four out of six. If it means "the Worlds champion and the MSI champion both come from the LPL or the LCK," the ratio is six out of six — but at that point the number is no longer about MSI. It is about the gap between two regions and the rest of the world. The interesting part is not the ratio. It is the seeding structure. An expanded MSI puts more teams into the group stage, stretches the series count, and turns the bracket into a double-elimination playoff. Statistically, that change makes it harder for strong teams to exit early, and when strong teams stop exiting early, the correlation between MSI placement and Worlds placement rises on its own. No miracle required. Just less variance. I tested this another way. I took the four semifinalists from each of the last five MSI events and compared them with the four Worlds semifinalists of the same year. The overlap grows over time, not because teams started playing alike, but because regional quality converged onto two ecosystems. Chovy of Gen.G, Knight of JDG, Faker of T1, Bin of BLG — four names for four teams, and for three straight years no team outside that group reached a Worlds final. The paradox sits here. If the expanded MSI makes results more predictable, then winning MSI becomes less valuable as a forecasting signal, not more. Once all four MSI semifinalists are Worlds contenders, one of them winning Worlds stops being an MSI signal. It becomes a consequence of the fact that only about six teams in the world can win the title at all. Six out of six sounds impressive, but the sample is six, and a sample of six cannot justify calling anything a rule. Three years of results, by basic statistical standards, is an unbroken lucky streak. I understand why the story sells. It hands viewers a prediction tool that spares them the group stage. Fans only need to watch MSI to know who wins Worlds, and that turns summer into a dress rehearsal for autumn. But a reliable report needs three signatures: the assistant coach, the agent, and the person in the kitchen. In esports statistics, the third signature is a large sample. We do not have it yet. People ask what I look at before a deal closes. I look at motive, not price. With MSI and Worlds, Riot's motive is clear: make the mid-season event part of the road to Worlds so every summer game carries commercial weight. That is why the format was expanded. The matching results were not part of the plan, but they serve the plan very well. MSI 2026 will be the real test. If the MSI 2026 champion also wins Worlds 2026, the record becomes seven out of seven and the narrative gets even harder to resist. If a team from outside the LPL and LCK wins MSI 2026, or if the MSI 2026 champion collapses in the Worlds knockout stage, the streak breaks and we will finally see what is structure and what is coincidence. What I want to track is not who lifts the trophy. I want to track where the second- and third-place MSI teams finish. If runner-up and third-place finishes at MSI start predicting Worlds outcomes with similar accuracy, that will be the real evidence that the new format carries predictive power. If only the champion slot keeps matching, then we are measuring the concentration of power, not the quality of a tournament.