Inside the Esports Analysis Room: The Line Between Data and Fiction

At three in the morning in Seoul, in a documentary editing room, I opened a spreadsheet sent over by a colleague and found it empty. Not empty because of a technical error. Empty because there was nothing to fill in. The message that came with it was a single line: "Just write to the framework; the numbers can be added later."

I remembered the summer of 2026, when I spent twenty days measuring the left elbow angle of a 100m sprinter across six starts. The average deviation was 14.2 degrees, worth 0.048 seconds. On the track, no one let me write "roughly 0.05 seconds" and skip the rest. The clock does not negotiate. An empty cell in a track dataset is a wrong cell, and a wrong cell is a wrong conclusion.

In esports, where everything is measured by metrics, there is an opposite temptation: to fill the empty cell with something that sounds reasonable. When that temptation is repeated often enough, it stops being an individual error. It becomes an industry.

Esports has travelled from chat rooms with a few thousand online viewers to a global industry worth billions of dollars, with finals staged in arenas seating tens of thousands. Vietnam sits in that current, and not on the margins. Arena of Valor (Lien Quan Mobile) is a title where Vietnamese teams regularly rank among the regional leaders. League of Legends has the VCS, with a steady viewership across many seasons. Alongside the growth of the competitive stage is the growth of a layer rarely seen: the analytical layer.

Today's fans no longer stop at watching who wins. They watch metrics, standings, predictions. That creates enormous demand for "data-driven" content. And that demand creates pressure: an article every day, a conclusion for every match, a number behind every conclusion.

In athletics, I used to tell young editors something I still hold to: starting 0.05 seconds late is, sometimes, the way to finish earlier — because that lateness forces a sprinter to recompute the entire rhythm. But in esports, there is no clock standing by as referee for lateness. There is no marker forcing the writer to be honest. That gap creates both freedom and trap.

When I helped build the analytical framework for esports features sent to the Korean market, I worked with a nine-layer structure: patch and meta; tournament system and format; teams and players; regional context; club finance and business; rules and governance; risk profile; public narrative and expectation; and industry transmission. With enough data, those nine layers produce a reusable analysis. With too little data, they produce a trap.

The first layer, and the easiest to overlook, is the patch. In esports, the title is not a static arena. It is a field that shifts with every patch. A small change to a champion's damage can push the win rate of the team using it up by a few percent, and a few percent, at the top, is the distance between the group stage and the final. A serious analysis must answer three questions: which title, which version, and how large the change is — a numeric tweak, a mechanic change, or a full rework.

Without a title, every metric behind it becomes meaningless. A 55% win rate for a champion in Arena of Valor says nothing about a champion of the same name in another title. This is what most esports content online overlooks: it cites very specific numbers without saying which reference system they belong to. A number without a reference system is not evidence. It is decoration.

The second layer is format. A BO1 in the group stage is a different sport from a BO5 in the final. BO1 rewards surprise, an off-meta strategy, a single soaring day. BO5 rewards depth, the ability to read and counter-read an opponent, the calm to last four hours. When someone says "team A is stronger than team B," the next question must be: in which format? Because the upset rate in the group stage is always significantly higher than in the knockout series. An analysis that does not distinguish the two contexts is only retelling the result, not explaining it.

The third layer brings us closest to people: teams and players. Here there is every kind of metric — kills, kill participation, gold per minute, fight win rate. But it is also here that metric abuse is most visible. A player with beautiful numbers is not necessarily the best reader of the game. Metrics tell the outcome of an action; they do not tell the decision that led to it. In a feature I once worked on, a player with a below-average stat line was the one who set the pace for the whole team, because he kept choosing positions that earned no kills but opened space for his teammates. The metric was not wrong. It simply did not answer the question people assigned to it.

Based on my experience watching matches, there is one cross-discipline comparison worth keeping. The 42 goals from set pieces at the 2026 World Cup were not about technique; they were about how a team reads the game. In esports, set pieces have their version: the early invade, the setup before a major objective, the timing of a lane swap. Those plays occupy a small share of a match but decide most of its shape. And they can only be read when we have data on timing, position, and the decision chain behind them — not when we have a single line of score.

The fourth layer is regional context. The same region can hold an entirely different standing in different titles. What is true for one discipline cannot be inferred for another. A country can be a power in the mobile arena yet only competitive in the PC arena. So any statement like "region X is strong," without a title attached, is an empty statement. This matters especially for Vietnamese esports, where results in Arena of Valor and results in League of Legends draw two pictures that cannot be merged.

The fifth layer, club finance, is often dismissed by fans. But money decides who stays, who leaves, and sometimes decides strategy itself. A team dependent on a single sponsor will behave differently in the transfer market from a team with diversified revenue. Financial-reporting pressure can weigh on sporting decisions: signing a player because he sells jerseys, not because he fills a gap in the roster. When fans' emotions are converted into money, the question of "who starts" gains a variable that is not on the field.

The sixth layer is rules and governance: competitive integrity, transfer rules, contracts, protection of minor players. This is the layer where a small mistake can destroy a whole season, and the layer the media usually touches only after a sanction has landed.

The seventh layer is the risk profile — competitive, financial, personnel, rules, public opinion, and systemic risk. The eighth layer is public narrative and expectation: motifs such as "new king," "dynastic succession," "all-domestic roster," "last dance." Motifs help a story spread, but they also easily hide the gap between expectation and reality. The ninth layer is industry transmission: from publishers upstream, through clubs and broadcast platforms midstream, to sponsorship and derivative markets downstream.

But there is a hidden tenth layer that none of the above mentions: the layer of honesty. An analysis can fill all nine layers and still be wrong, if the tenth — the origin of every number — is skipped. Every conclusion in a serious analysis must be traceable to a specific information point. No information point, no conclusion. That is the driest rule, and also the rule that saves the most analyses.

Those nine layers are a machine. And that machine only runs on fuel: real input information. The most dangerous thing in esports analysis is not a lack of data, but an empty input disguised as a complete analysis. Empty data is not neutral — it invites the writer to invent the missing part, and with each invention, a wrong conclusion is dressed in the coat of a grounded one.

Here I want to pause before a common belief: that the more metrics esports has, the more scientific it becomes. The belief sounds very fair, and most of the time it is true. But it has a blind spot.

In football, xG was once hailed as a revolution, then overused to the point that many people deployed it to explain things it was never designed to explain — player form, a coach's decision, a referee's standard. Esports is walking the same road with KDA and composite ratings. When a metric becomes the answer to every question, it stops being a tool and becomes a faith.

Conversely, there is a behaviour this industry dreads: saying "we do not yet have enough data to conclude." Such a bare answer sounds weak in a market that demands a conclusion every day. But an empty input, properly acknowledged, is more honest than a trend drawn from fiction. Methodical skepticism is a skill that requires practice, not a hesitation. And the hardest skill in analysis is telling apart "not enough data" and "enough data to conclude wrongly."

From the track to the pitch, every moment of genius begins with a decision that seems meaningless — including the decision not to write. A good analysis is not the one with the most numbers, but the one that knows which number is still missing. And perhaps the most valuable skill for a practitioner this decade is to stay honest when there are moments when we have nothing to say.