Nine Layers of Esports Analysis: When the Data Is Empty, the Only Correct Report Is an Empty One

11 p.m. in Busan. On my screen sat a nine-layer analysis of an esports article. Layer one, patch and meta: insufficient information. Layer two, tournament system and format: insufficient information. Layer three, teams and players: insufficient information. And so on down to layer nine, industry transmission. The only populated field was the domain label: esports.

An outsider would call this report worthless. I stayed another twenty minutes. A record that is empty but structurally correct always tells me more than a record that is wrong but full of words.

I work as a transfer market administrator specialising in esports, after twelve years inside the industry, moving from player to tournament organiser to media. My job is to build the frame before reading any conclusion. That frame has nine layers: patch and meta; tournament format; teams and players; regional landscape; club finance; rules and governance; risk profile; public narrative; and industry transmission.

Those nine layers do not exist to fatten a report. Each layer answers a question the layer below cannot answer in its place. One patch line can wipe out an entire champion pool and rewrite the ban-pick phase. A best-of-one format and a best-of-five format produce two very different upset probabilities, even when the same two teams walk onto the server. A team that is stable, a team in transition and a team rebuilding lead to three different readings of the same win.

The entity layer is the lowest layer, and it is the load-bearing one. Without a game title, a team name, a player name, a tournament name, the other eight layers are empty frames with decoration.

That is exactly what I was looking at. No game title. No patch version. No team. No player. No tournament. No publisher. No region. No jurisdiction.

When the entity layer is empty, every conclusion above it is literature, not analysis. I wrote that line on a sticky note and stuck it to the edge of my monitor, because it is a mistake I have made, and the mistake I see most often in transfer reports.

The patch layer dies first. Patch cadence differs between titles: some update every two weeks, others let one large build overturn an entire champion pool for months. Without the title and the version number, I cannot say who benefits, who loses, or how long the meta honeymoon window lasts.

The format layer dies second. Best-of-one raises upset probability, best-of-five compresses it. A Swiss format accelerates meta iteration. Without a format, every judgment about stability or luck is meaningless.

The finance layer dies third. I carry an industry prior: salary-to-revenue ratios at many esports organisations routinely exceed 80 percent, well above the healthy band in traditional sport. A prior only has value when I can name an organisation. An anonymous club has no ratio to compare against.

The governance layer dies in the most dangerous way. Rules here form a hierarchy: publisher rules, league rules, third-party organiser rules, national policy. If I cannot establish which system applies, no compliance judgment stands. Silence carries no evidentiary weight. An empty record does not prove a violation, and it does not prove the absence of one.

I learned that principle through an expensive lesson. In 2026, as a first-year university student in Busan, I collected K League 2 data myself and found a divergence: league leaders Asan Mugunghwa generated only 1.02 xG per match, while Busan IPark, below them in the table, generated 1.48. Asan were winning on six penalties in six matches. I wrote that they would fall. They finished fourth and lost in the play-offs. The post drew 2,000 views, an enormous figure for a student blog.

Do not trust the table, ask xG. The table tells the past, the data tells the future.

In 2026 I proposed signing a midfielder for eight million euros. My data showed he ranked in the top ten in La Liga for chances created per 90 minutes, at 2.8, above Isco. The board rejected the move, arguing he could not defend. Six months later that player helped his club survive relegation, while my club finished eighth. I wrote a fifteen-page internal report, admitting a process failure rather than blaming any individual.

A transfer fee is the number one person is willing to pay. True value is the number the data does not need to negotiate.

But that story taught me a second thing: had I only held one metric at the time, I would have been wrong. What I lacked was the context layer, namely format, schedule and the quality of teammates.

Moving into esports, the biggest temptation is metric import. I was once attacked for daring to question PPDA. FIFA confirmed it. Their report, published three weeks later, confirmed what I had written: a low PPDA is not a certificate for a sustainable pressing system, and that system can break after minute 60 once you split the data into fifteen-minute blocks.

But football PPDA does not translate directly into esports. A title has a meta, a patch cycle, a ban-pick phase. A player with a high first-blood rate may simply be playing the one patch that forces every team into early skirmishes. That rate is a consequence of the ruleset, not a skill signal. Read it wrong and you pay with a roster slot.

In the summer of 2026, when leagues had to play behind closed doors, I tracked 214 matches in the Bundesliga and K League 1 from May to August. The home win rate in the Bundesliga fell from 43.2 percent to 37.8 percent, and average goals rose from 2.79 to 3.12. Many people call that a natural experiment. I call it an opportunity to measure luck.

The lesson applies directly here: inside an empty record, the most dangerous mistake is substituting base rates for evidence. An analyst under deadline pressure will happily fill a blank cell with a plausible story, this team is rebuilding, that player has a wrist injury, this organisation is about to collapse under wages. Those sentences are not wrong in style. They are wrong in sourcing.

There is a second, subtler trap. When every risk cell reads not assessable, a reader skims and registers no risk at all. A risk that has not been rated is not yet a risk of zero. The gap between those two phrases is wide enough for a team to lose a slot, a player to lose a contract, an organisation to lose a licence.

Since that night I have imposed a hard rule on myself: empty input means empty output, and that empty pass must be logged and attributed, never quietly deleted. An empty record is the cheapest and earliest signal the system can send me, reporting that source data never arrived, or arrived empty because of a login wall or a fetch error. Fixing one fetch is small work. Distributing a conclusion built from nothing is large work.

In the next analysis cycle, before I read any claim about an esports team, I will ask one question ahead of all others: has the entity layer been populated? If it has not, the rest is decoration. Data does not care who you are, only whether you read it correctly.