Nine Layers of Esports Data: An Industry That Has Never Been Audited

Nine Layers of Esports Data: An Industry That Has Never Been Audited

Four thousand words. Nine analytical layers. Not a single fact.

A deep-dive esports report ran its entire framework: the patch and meta layer, the tournament format layer, the team and player layer, the regional layer, the club finance layer, the rules and governance layer, the risk layer, the public narrative layer, the industry transmission layer. Nine layers. Each with a table. Each with an assessment grid. Each with a "confidence level" field.

In every cell, the same line: insufficient information to assess.

What stopped me was not the emptiness. It was its form. No syntax errors. No forgotten cells. Every field populated in the correct format, fully labelled, ready for publication. A report that analysed nothing, and yet looked entirely valid.

I have seen that exact shape somewhere else. Not in a technical fault. In the way the esports industry handles its own data.

Context: three layers of data production, and none of them speak to each other

Across fourteen years of following this industry, from competing as an amateur and running small tournaments in Vietnam, through a media phase, to a club finance seat in Seoul, I have never encountered a sector with so many tables and so few audits.

Esports produces data on three tiers. The first tier is the publisher: Riot Games, Valve, Tencent, Blizzard publish patches, competition rules, and sometimes raw match data through APIs. The second tier is third-party aggregators, the statistics sites anyone writing about League of Legends, Counter-Strike or Valorant opens in the first tab. The third tier is commercial data suppliers serving broadcasters and the betting market.

Those three tiers do not share a unit system. The first publishes by season and region. The second collects by match and by game, by recording the screen. The third buys data by the second. When an article cites "this champion has a 54 percent win rate", the reader does not know which tier the number came from, what the sample was, which region, which patch. Numbers do not lie. People misread them.

In South Korea, where I live and work, the LCK has operated on a franchised model since 2026: teams pay an entry fee and share revenue through a central mechanism, with no promotion or relegation. In Vietnam, the League of Legends regional competition that once held major-region status was restructured and merged into a shared Asia-Pacific league from 2026.

Those two models generate two different kinds of data. One is centralised, standardised, with someone accountable, with contracts and internal review. The other is distributed, largely collected by the community, with no one accountable, sustained mostly by trust.

The gap is not one of player quality. It is one of measurement infrastructure. And measurement infrastructure decides who is considered talented, who gets paid what, and who is discarded.

Layer one: a meta built on a few dozen games

Patch notes are public. Anyone can download them. Real professional win rates are not. Third-party statistics sites collect match data through APIs and screen capture, then assemble pick-ban tables.

The problem is sample size. One week of competition in one region may produce only a few dozen games for a specific champion. A few dozen games is the sample size of a preliminary survey, not of a conclusion. Yet the entire media narrative that follows, who is leading the meta, who is falling behind, who is misreading the patch, is built on that base.

This industry reads a forty-game sample as if it were a law of physics, and calls it analysis. When teams reach the knockout stage of an international event, where the denominator changes completely because teams only face each other a handful of times, the entire prior set of conclusions collapses at the same moment as the win rate.

More striking: the teams know this. In sessions sitting in club analysis rooms, I watched coaches use public data mainly to cross-check their own instincts, not to make decisions. The actual pick-ban decisions lived in internal scrim data that never leaves the room. Which means the data the public argues about most is the data least used to make decisions.

Layer two: format decides upset probability, and nobody writes about it

Best-of-one, best-of-three or best-of-five is the single largest variable determining whether a strong team exits early. In a single-game format, upset probability is many times higher than in a five-game format, simply because variance never gets a chance to flatten out.

Yet in most coverage, format appears only as an administrative line. The story of "team X is inconsistent" is mostly the story of "team X played a group stage in best-of-one". The story of "team Y cannot close out games" is mostly the story of "team Y had a seven-game break and lost rhythm".

When a region restructures its league, the format changes, the number of international slots changes, and the entire frame of reference for cross-regional comparison is knocked out of alignment. Nobody publishes an impact model. Nobody re-runs historical data to see how upset rates would shift. Format is decided in a closed room and explained to the public in a three-paragraph press release.

Layer three: no salary table exists, so no value table exists either

Esports has no effectively functioning players' union in most regions. No salary disclosure mechanism. No injury list. No body that tracks the average career length of a professional player.

The result is that every player valuation is built on rumour, on a handful of leaked contracts, and on feeling. The world looks at the star; I look at the value table. But in esports, the value table has never been drawn up.

Injury is the clearest example. Wrist injuries and tendonitis are an occupational risk specific to this industry, because training intensity is measured in thousands of clicks per day. The case of Lee Sang-hyeok, known as Faker, having to sit out matches due to a wrist problem in 2026 is one of the very few publicly disclosed injuries in the history of professional League of Legends. It was disclosed because that figure was too large to hide.

For the rest of the industry, injury is private data. A player can be declining thirty percent because of a wrist, and the market will call it a form slump. There is no way to tell the difference.

On a two a.m. call with a club's board, I once had to argue, using GPS and aerial-duel data, that an unheralded winger was worth more than the market price assigned to him. We signed him. But the notable part is this: if the board had not trusted the model, they could have refused with a single sentence, and there was no public data source with which to argue back.

Layer four: the Vietnam-Korea power map through a data lens

Vietnam has one of Asia's strongest talent export pipelines in League of Legends. Names such as Đỗ Duy Khánh and Lê Quang Duy have proven that at international events for years. But a strong talent pipeline and a strong data system are two different things.

In South Korea, a seventeen-year-old prospect entering an academy system is measured from day one: games played, practice indices, reaction times, progression curves. He exists in a database before he exists on a scoreboard. In Vietnam, a seventeen-year-old prospect usually exists first in one observer's eye, and is converted into data later, if anyone bothers.

This difference determines price. A player measured from childhood has a long record and can be valued along a development curve. A player who has only highlight reels and reputation is valued by sentiment, usually at a discount, because scouts have nothing to prove their case with.

Vietnam's region merging into a shared Asia-Pacific league from 2026 changed both sides. Vietnam gained access to more centralised media and data infrastructure. The larger regional teams gained access to a cheaper talent market. Talent flows and data flows move in the same direction, and data flows usually move a step ahead.

Layer five: half a financial picture is drawn, half is hidden

This is the layer I work in daily, and the least transparent in the entire industry.

Esports is remarkably generous on the revenue side. International prize pools are fully disclosed, sometimes exceeding sixty million US dollars for a single Esports World Cup in Riyadh. Sponsorship values at tournament level are announced with a handshake photo. The revenue of a few of the largest clubs is released in annual press statements.

On the cost side, near silence. No club publishes independently audited financial statements. Starting-roster salary structures are not disclosed. Parent-company investments, internal loans, and unpaid player obligations appear in no public document.

The salary-to-revenue ratio across most professional esports organisations is commonly cited above eighty percent. In European football, that threshold is a red alert. In esports, it is normal.

A financial picture with only one half drawn is not a picture. It is an invitation to guess. And because nobody audits, every club valuation model, every brand-value figure, every growth forecast stands on sand.

Layer six: the lawmaker, the ticket seller and the referee are the same party

In esports, the publisher owns the game, runs the tournament, sells in-game items, and arbitrates disputes. No independent arbitration body sits above that structure.

The regional restructures of recent years were all announced in the same mould: one notice, one chart, one timeline, and a list of affected parties who were not consulted.

There is nothing legally wrong with that. But in data terms it creates a specific hole: when governance decisions are made by a party with a direct commercial interest, the figures used to justify those decisions are never fully published. Nobody knows whether a region was shrunk for competitive performance, for viewership, or for an internal cost calculation.

In finance we call that governance risk. In esports, it is the default condition.

Layer seven: the biggest risk is the risk you cannot see

A complete risk profile in professional sport covers competitive, financial, personnel, regulatory, reputational and systemic risk.

Esports has all six. It can only measure three. Competitive risk is measured by match statistics. Reputational risk is measured by viewership and engagement. Regulatory risk is measured by sanctions already issued.

The other three have no data. Financial risk has no audit report. Personnel risk has no injury list and no burnout data. Systemic risk, meaning risk arising from the structure of the industry itself, is not measured because nobody stands outside the structure to measure it.

Burnout is the textbook case. A professional team's practice schedule can run twelve to fourteen hours a day for months. No industry-wide index tracks it. When a player retires at twenty-two, the public calls it lost motivation. Nobody calls it an occupational injury, because there is no data to call it that.

Layer eight: the heat cycle and the narrative trap

Every esports star passes through four media phases: discovered, hyped, peaked, backlash.

The second phase is the most dangerous and the least measured. When coverage converges on a name, every match statistic for that person is read favourably. When the cycle reverses, the same statistics are read in the opposite direction, and the community has its own word for such cases.

What is striking is that both readings lack a data foundation. A player's index depends on the patch, the format, the opponent, game length, and whether his team is winning or losing. No single standard metric corrects for all of those variables.

In the Vietnamese market, where the fan community is exceptionally large and exceptionally fast, this cycle completes within weeks. A player can be called the best in the region after one match and washed up after the next. That speed is not a sign of understanding; it is a sign that no sufficiently reliable yardstick exists to hold the argument still.

Layer nine: where the money enters and where it leaves

An esports transmission chain has four nodes: the publisher upstream, clubs and streaming platforms midstream, sponsorship downstream, and mainstream integration at the output.

Money enters mainly at the first and third nodes. Sovereign funds, telecoms groups, retail chains and technology platforms pour money into tournaments and teams in exchange for brand presence. Money leaves mainly at the second node: player salaries, operating costs, travel, development.

The conversion efficiency between the two ends is unmeasured. How much real value a ten-million-dollar sponsorship creates, what share of it converts into direct media revenue, and what share remains inside a brand-awareness slide. I once took part in exactly such an assessment for a sponsorship campaign at an Olympic Games, and the conclusion was that most of the supposed value did not exist, because the primary channel for the young audience had shifted to short-video and streaming platforms, where most viral athlete moments came from competitors with no official sponsorship relationship.

The counter-intuitive angle: this industry does not lack money, it lacks auditing

Public debate about esports always revolves around money: are prize pools big enough, are investors withdrawing, is sponsorship sustainable. That framing assumes the problem is the scale of the cash flow.

I think that diagnosis is wrong. A sixty-million-dollar prize pool does not make this industry one cent more transparent. The larger the sums, the more dangerous the audit gap becomes, because more people make decisions based on numbers nobody verifies.

And here is the point I want to make as someone who works in Seoul but was born in Vietnam: precisely because of early-stage scarcity, some smaller markets are forced to be honest about their unit economics. When you cannot afford a ten-man bench, you are forced to know exactly how much a player returns. Poverty, in this case, is a form of data discipline.

Conversely, wealth in large markets creates room for inflated numbers. A well-funded team can pay against potential, and potential is an unverifiable category. When the flush period ends, those contracts become evidence that nobody was auditing.

What worries me most is not a wrong number. It is a wrong analysis that reads fluently. A report that is empty and beautifully formatted, with confidence labels fully populated, is the perfect example of the most dangerous thing in this industry: a conclusion based on nothing that still looks valid enough to be cited. When data speaks, the whole world suddenly listens. When data is silent and someone speaks on its behalf, that is when this industry is in real trouble.

In my own career I once found a deal the market had missed, purely because the data proving it sat in a source nobody bothered to open. I found a diamond in a pile of noise. But for every diamond found, there are hundreds of decisions made on a fabricated number that nobody checked.

An open ending: the question is not whether there is data

Nine analytical layers. All nine returned empty. And the frightening part is that nobody in this industry felt anything was unusual.

Esports will not solve its problems by producing more data. It will solve them only when someone agrees to publish an empty field, a cell that says we do not know, a report that admits it could not measure what it claimed to measure.

In an industry where everyone is trying to look informed, the most disruptive act available may be a blank table. The only question is who will be the first with the nerve to publish it.