Nine Layers of Data Behind a Major Esports Tournament

On the night of 2 November 2026, at the O2 Arena in London, T1 beat Bilibili Gaming 3-2 in the League of Legends World Championship final. According to Esports Charts data, nearly seven million people were watching at peak. After the final whistle, most of what spread across social media was about Faker's nerve, about a 28-year-old who had been playing professionally for more than a decade. None of those lines were wrong. They were simply missing one thing: the ability to be verified. If that night in London were retold through the gold difference at minute 15, the rate of major-objective control, and the number of full teamfights in the last ten minutes, the story would keep every ounce of its emotion, but readers could reopen the record and check it themselves. That is the entire difference between a good commentary and an analysis that can be used again. I write with a spreadsheet. The spreadsheet is an altar, and I give myself to every figure. In 2026, when I was a mid-level editor at a football platform in Shanghai, I refused to write a piece praising Shanghai Shenhua's fighting spirit after a derby they won 2-1 while being pinned back. The visitors took 20 shots for 2.8 xG; the hosts scored twice from 0.9 xG. I published three metrics and said plainly that the win belonged to luck. On that Shanghai derby night, I chose the metrics instead of the whole city. Fans threw stones, analysts read it, and from then on every piece of mine has had to carry at least three different metrics before any judgement is made. In March 2026, I analysed ten of Germany's qualifying matches and found their average PPDA was 11.3, far above the 8.5 to 9.5 band of the leading pressing sides. I wrote that Germany would be eliminated in the group stage. On 27 June 2026, Germany lost 0-2 to South Korea and finished bottom of Group F. They told me I was causing trouble. I had simply read the ending a few months early. In 2026, I collected 250 Bundesliga matches after football restarted behind closed doors. The home win rate fell from 43% to 31%, and average goals per match dropped by 0.4. I published a study titled "A silent stand is a metric" and lost a private syndication contract for insisting the conclusion stand unchanged. Without a crowd, football changes shape. I found that out, and I was rejected for it. In 2026, I used that same model to predict Denmark would beat England in the Euro semi-final, based on 118.7 km covered against 112.3 km, and 18 shots against 11. Denmark lost 1-2 after extra time. I had ignored squad depth and the ability of a substitute to change a game. Since then, every piece of mine carries its own section: "Where could the assumptions be wrong?". Moving into esports, I carried that entire discipline with me and ran into a paradox. Esports produces more data than football, because every item purchase, every click and every second is logged by the system. But the pressure to tell stories is greater, because tournament cycles are short and content has to be out within hours of the final whistle. The more data exists, the more people choose the shortcut. From the Bundesliga to Worlds, I look for the same thing: a fact that can repeat itself. The framework I use before every major tournament has nine layers. It is an order of questions, and every layer is allowed to return an empty result. Every esports analysis starts with the patch. League of Legends ships an update every two weeks, but a major tournament is played on a version locked weeks earlier. That gap creates a familiar paradox: teams at home practise on the new meta, while teams at the event must solve the old problem. At the 2026 World Championship, Invictus Gaming won 3-0 against Fnatic in a tournament shaped by early skirmishing and the ability to close games before minute 25. In hindsight, that was a consequence of patch structure, not of any press statement. If the competitive patch cannot be identified, every cross-tournament comparison loses its value. This layer sits above all the others because it sets the meaning of every metric that follows. The patch sets the meaning of the metrics; the format sets the meaning of the results. Since 2026, the World Championship group stage has been replaced by a Swiss format with sixteen teams and five rounds; pairings are drawn by record, teams on three wins advance and teams on three losses are eliminated. That change lowered the probability of a strong team being knocked out by a single match and raised the value of squad depth. Under the old format, a four-team group with two qualifying spots made the margin for error in the first two games decisive. The same roster can go a long way in one format and collapse in another, so I always state the series length, the tiebreak rules and the schedule density before I say anything about anyone. Structure is only the frame. Inside that frame are people. DRX won the 2026 World Championship after coming through the play-in stage and beating T1 3-2 in the final. That roster included Deft, who entered the tournament at 26 after nine years as a professional and several team changes. An individual form curve does not travel in a straight line; it follows a trajectory tied to role, to the jungler beside him and to how the team allocates resources. The same player, changing teams and coaches, can see his numbers swing thirty percent without any decline in skill. I do not compare KDA across different roles. The metrics of a mid laner and of a support do not share a unit of measurement. Forcing every position onto one scale is the most common error in the player rankings I read, and it tends to appear exactly when readers want a fast conclusion. People belong to a region. China's LPL won in 2026, 2026 and 2026. Korea's LCK won in 2026, 2026, 2026 and 2026. Europe has a single title, from the inaugural season in 2026 when Fnatic lifted the cup, and North America has never won a world championship. That concentration is not luck. It is tied to infrastructure: academy systems, controlled practice hours and a domestic market large enough to sustain several teams at once. When I analyse a team, my first question is always where they come from, because the region sets both the quality of their scrim partners and the standard they must clear every week. A third-place LCK team can be stronger than the champion of a smaller region, and that never shows up in any international ranking. Regions run on money. In 2026, North American franchise slots were reported to carry a fee of around ten million US dollars each. Through 2026 and 2026, several major organisations left the system, sold their slots and cut payrolls. In parallel, the Esports World Cup, held for the first time in Riyadh in 2026, announced a prize pool of more than 60 million US dollars, the highest ever recorded for an esports event. Those two currents run in opposite directions, and any judgement about a team's strength has to sit on that team's own financial base. A team that pays wages on time keeps its roster across seasons. A team that owes wages loses people, usually in the week before the roster lock. Money operates inside a rulebook. Esports betting is eroding competitive integrity faster than traditional sport, simply because the regulatory framework has not kept pace with how fast the market grew. Bodies such as ESIC exist, but their power largely stops at recommendations, internal investigations and bans inside a voluntarily participating tournament circuit. In football, a betting-related sanction can rest on both international convention and national law. In esports, most cases are still settled indoors. This is the highest-risk and thinnest-data layer in the entire framework. A match with an abnormal move in the odds can be evidence, or it can simply be a group of people betting badly. Without a long enough public database, every conclusion at this layer has to travel with a probability attached. A rulebook only matters when risk is named. In 2026, Faker was absent for about a month with a wrist injury and T1 struggled badly during that stretch. Injuries in esports are not like injuries in football: they concentrate in the wrist, shoulder, back and eyesight, parts that cannot be rotated the way a centre-back or a winger can. Add the competition calendar, long-haul travel and visa problems. Before every major tournament, I build a dedicated risk column for each team: who cannot be replaced, who is overloaded, who the team depends on as a single individual. That column has generally predicted outcomes better than any metric table, because it measures endurance rather than peak ability. Risk is measured; public expectation usually is not. The Chinese community has a word for someone who is hyped up and then collapses: cjb. It was born precisely from the mismatch between media expectation and actual results. A team can keep the same roster and the same numbers, yet walk into a tournament under three times the pressure simply because it is treated as a title favourite. I read market expectation as a variable, not as evidence. When a team is rated above its true level for three months straight, that gap becomes a signal. When it is rated below its true level, the opportunity usually sits right there. Public expectation is the surface layer. The deeper layer is industry transmission, running from the publisher through clubs and tournament organisers, through streaming platforms, to sponsors and derivative markets. In 2026, esports was included as a medal sport at the Asian Games in Hangzhou, and South Korea took gold in League of Legends after beating Chinese Taipei 2-0. That event pulled esports into the same regulatory paperwork as traditional sports, with the full set of obligations around testing, contracts and participant age. For a writer, this layer decides which content survives the tournament. A patch change can be forgotten in two months. A change in broadcast rights, or a new testing standard, can shape the next decade. Nine layers, and one possibility I have to state plainly right here: all of them can return an empty result. In analytical work, the greatest temptation is not being wrong. The greatest temptation is guessing loudly when the data does not exist, and then giving it a respectable name: expert instinct, market feel, or years of experience. Correlation is not causation, and in esports the distance between the two is wider than in football. A major tournament holds only a few dozen matches; a player contests only a few dozen games in a season. With small samples, any metric ranking is easily distorted by a handful of outliers. I have seen rankings published after three matches, and they travel faster than any correction ever does. My "Where could the assumptions be wrong?" section for this cycle has four lines. I assume the competitive patch is locked exactly as announced and that no hotfix lands mid-tournament; that public data on schedules and rosters is complete, when in reality teams hide their hands until the last minute; that expectation pressure does not affect in-game decisions, when evidence from football behind closed doors shows the playing environment changes human behaviour; and that one tournament is a large enough sample to draw conclusions from, when in reality it is only large enough to propose a hypothesis. Every crowd is wrong. The only thing that is not wrong is probability. The signals I will track when the next round begins come in two parts: how many teams keep their starting roster across two consecutive seasons, which is the most stable indicator of real strength, and how franchise slots and payrolls shift toward regions with the biggest prize pools. If those two lines cross, the regional order will change before any international ranking catches up. When that moment arrives, I want my spreadsheet to have the date already written down.