The T1 File: A Four-Year Discrepancy in a Single Data Line and the Silent Restructuring

On May 29, a periodic disclosure filing recorded the term of Joe Marsh — CEO of T1 — as extending to March 30, 2029. Previously, that term had been recorded as ending at the close of 2026. A four-year discrepancy sits inside a single administrative field. For most readers, it is a detail to scroll past. For me, it is the only fact in the entire T1 story that can be verified on paper. I have followed professional League of Legends since 2026, when I still recorded every match metric by hand in a paper notebook. That work taught me one thing: the most important facts rarely live in the headline. They live in the appendix, in the footnote column, in a field nobody double-checks. A date shifted by four years inside the filing of a top-tier esports organization is exactly that kind of fact. T1 was established in 2026 as a joint venture between SK Telecom (now SK Square) and Comcast Spectacor. That structure is uncommon in esports. Most Korean organizations are owned by a single conglomerate, or by a small group of aligned investors. T1 chose the two-owner path, and every governance question that followed has its roots in that initial decision. According to public sources, SK Square holds roughly 53.13% of the shares, while Comcast Spectacor holds more than 30%. A second source puts Comcast's figure at approximately 34.3%. The gap between the two figures is not large in monetary terms, but it says something important: ownership data is being released by different parties at different moments, or under different interpretations. On the competitive side, T1 has just come through the most successful stretch in its modern history: two consecutive League of Legends world championships. This is the pivotal variable in the whole story — not a tactical variable, but a valuation variable. Brand value rose sharply after those two titles. Another event heated up the context. Lee Sang-hyeok — Faker — met with Jensen Huang, founder of NVIDIA. Images of the two quickly drew the attention of the international esports community. Around the same period, Huang referenced PC bang culture and Korean esports in the story of NVIDIA's development. The AI industry is growing strongly in South Korea, and the strategic value of large esports brands is drawing more attention. These three events — two world titles, the Faker-Huang meeting, and the date line in the filing — do not sit in isolation. Together they form a context in which any dispute over control has a clear economic rationale. When an asset appreciates, the parties involved begin to look at their own ownership structure with different eyes. This is where I stop and split the data into two layers: the verifiable layer, and the purely inferential layer. The first layer — ownership and board structure. SK Square holds 53.13%. According to Sports Seoul, the board seat ratio was previously 3-2 in favour of the SK side. In April, T1 was reportedly adding Kim Jaerin — whose background is at SK Square — to the board. Daily Esports later recorded the ratio as 4-2. Two numbers, two sources, the same window of time. I cannot confirm which is correct. But I can say this: when two sources in the same market produce two different board structures, the inconsistency itself is data. The second layer — the CEO term. This is the hardest fact available. A filing dated May 29 records Joe Marsh's term as running to March 30, 2029. He remains listed as CEO on T1's official information page, responsible for global operations. Daily Esports read the change as a possible sign of shareholder disagreement. That same report also warns explicitly that this is a hypothesis, not a confirmation. I mark confidence as medium that this is governance-relevant, and low that it signals an adversarial power struggle. The reasoning: a CEO term is an administrative field, and administrative fields can be updated for many reasons — ordinary contract extension, internal adjustment, or an agreement between shareholders. A date line on its own does not tell a story. There is another detail I consider more important than either layer above. In 2026, there was speculation that SK Square might transfer T1 shares to Comcast. That speculation did not materialize as predicted. But the more interesting question is: why did it not happen, and why has the issue resurfaced precisely when T1's brand value is at its peak? The answer lies in a variable the original reporting calls strategic value. The AI industry is growing strongly in South Korea. The strategic value of major esports brands is drawing more attention. When an asset appreciates, holders want to keep more of it, not sell. That means a renegotiation of the joint-venture structure is far more plausible than a hostile takeover. Here is the point I want to underline: a 53.13% stake sits above a simple majority but below a supermajority threshold. SK Square controls ordinary resolutions. Comcast retains veto leverage on any matter requiring a supermajority. That structure is itself a source of tension, without anyone needing to declare war. This is the arithmetic of corporate governance, not an emotional drama. Both major shareholders are reported to have participated in board meetings and to have shared CEO candidate lists. That is a significant fact. It shows the matter is receiving attention. It also shows the parties are still at the same table. Two people sharing a candidate list are not two people preparing for court. Both SK and T1 issued responses saying they had no content they could confirm. Technically, this is the standard corporate answer. It neither confirms nor denies. I do not read it negatively, and I do not read it positively either. In data analysis, a neutral response must be treated as a neutral response. One structural point deserves attention: T1 is a multi-title organization. League of Legends is the brand pillar, but the organization runs several other rosters. Each title has its own cycle, its own cost base, and its own degree of brand dependence. Any instability at the governance layer trickles down into investment decisions at the title layer — more slowly, but more certainly. On the brand-value side, Faker plays a special role. Throughout this story, he appears as a commercial asset and an outward-facing icon, not as a competitive subject. The meeting with Jensen Huang is the trigger of the media narrative. Two consecutive world titles are the valuation foundation. Both are assets tied to one individual and one short achievement run. A decade in esports has taught me that brand value can rise very fast, but its durability depends on whether the organization can separate the brand from a single person. At the industry layer, the T1 story is an example of a larger trend: esports brands are increasingly seen as strategically valuable in the AI era. NVIDIA referencing PC bang culture and Korean esports within its own development story is one marker of that trend. But I keep the rule of separating a macro trend from a micro event. A real trend does not automatically confirm a specific transaction. The counterintuitive angle sits here: the story that spreads most widely is the story with the least evidence. The NVIDIA-T1 connection — via the Faker-Huang meeting — carries enormous media value. The images spread fast. The international community paid attention. But a direct link between Huang's visits and T1's share decisions has never been confirmed. Any conclusion that NVIDIA is involved in T1's ownership structure is unsupported. This is the kind of error I encounter often in analysis: two events occur close in time, and the human mind automatically connects them with a causal line. Correlation is not causation. But in esports, where the speed of transmission outruns the speed of verification, correlation usually wins. The same applies to the power struggle frame. It is the most attention-grabbing phrasing, but also the least substantiated. The original reporting states plainly: there is not enough basis to affirm that an open power struggle has appeared. When the source says there is not enough basis, readers remember the headline instead. And there is a small paradox worth noting: the two board-ratio figures — 3-2 and 4-2 — differ, and the two figures for Comcast's stake — more than 30% and roughly 34.3% — also differ. If this were a clearly defined power struggle, the parties leaking information would have to be more consistent. This inconsistency suggests an ongoing negotiation phase, where the parties deliberately preserve room to manoeuvre. Every conclusion in this piece carries uncertainty. The sample size here is the number of official filings — and it is very small. I set confidence intervals as follows: the likelihood that the board structure is being reshaped, medium to high. The likelihood of an open hostile takeover, low. The likelihood that the NVIDIA-T1 link has a real transactional basis, low. The largest structural risk is not shareholder conflict. It is valuation dependence on one individual and one short achievement run. Two consecutive world titles and Faker's image are the two pillars holding up T1's brand value. Neither is a permanent asset. Variance is not the enemy — it is the mirror that reflects the arrogance of prediction. Signals to watch over the next one to two quarters do not live in rumours. They live in three administrative fields: updates to the Korean corporate registry, a board-seat ratio that appears consistently across at least two sources, and any officially published share transfer filing. If Joe Marsh is replaced or a successor is formally named, that is confirmation of governance change. If the board ratio stabilizes at 4-2 across multiple sources, that is a sign SK Square is consolidating influence. If any official announcement arrives regarding cooperation between NVIDIA and T1, that is when the most widely shared story becomes verifiable. Data does not know how to lie, but it learns how to hide the most important thing. In T1's case, what it is hiding is not in the headline about a power struggle. It is in a date line inside the May 29 filing, and in the gap between two numbers whose names the parties have not yet agreed on. A decade in esports has taught me that: during the pandemic, I built an empire out of numbers nobody was watching. It still stands today. But it only stands because I never called a correlation a cause. Esports does not move more slowly than football — it simply runs on a different clock.