Nine Layers of Esports Data: When an Analyst Must Say 'Unassessable'

One late October evening in Gangnam, Seoul. I opened the nine-tab spreadsheet I have used for every major esports event over the past five years. Tab one: patch and meta. Tab two: tournament format. Tab three: roster and players. The remaining six tabs ranged from club finance, regional mapping, rules and governance, risk profile, media narrative, all the way to industry-wide transmission. That night, all nine tabs were blank. No game title. No patch. No team. No player. No transaction. No timestamp to anchor an analysis.

The greatest temptation in this profession is filling gaps. When the sheet is blank, an inexperienced writer will type a few generic lines to populate the columns — "the situation is stable", "the roster has depth", "the meta is shifting" — and close the laptop. I once did exactly that. In 2026, after Korea beat Germany 2-0 in Kazan, I wrote an article showing the home side's xG was only 1.12 against the opponent's 2.31, possession under 40 percent, and that the win came from fifteen minutes of late pressing. The piece exploded. Some called me a traitor to a historic victory. Traffic jumped from two hundred to twenty thousand in three days, but I cried because I was misunderstood. The Seoul night of 2026 taught me that the truth can be lonely, but never wrong.

That lesson followed me for seven years. It is the reason tonight, in front of a blank sheet, I did not fill anything in. Before trusting a number, ask where it was born. And when there is no number at all, the most honest answer is one nobody wants to hear: unassessable.

Esports is not one sport. It is a cluster of overlapping ecosystems, each run by a different publisher, operating on its own rhythm, paying players through its own mechanism. Riot Games ships League of Legends patches on a fortnightly cycle, regular as a clock. Valve is the opposite: Dota 2 and Counter-Strike live on large, infrequent updates heavy enough to overturn an entire season overnight. Tencent runs Honor of Kings on an Eastern seasonal calendar, where the schedule is bound tightly to local holidays and culture.

Because of this, every esports analysis must begin with a single question: which game title? Without a title, there is no patch to read. Without a patch, there is no meta. Without a meta, there is no way to know whether a team is rising or dying. Identifying the title is not a soft step that can be skipped. It is a blocking precondition, like knowing whether you are discussing football or basketball before talking tactics.

In my trade, there is a trap more dangerous than a data error: confusing different titles. A team that dominates Southeast Asia in one MOBA may be a group-stage side in a tactical shooter. The same country, the same region, yet regional standing flips entirely depending on the title. That is why conclusions about a region cannot be borrowed across titles. It is why my spreadsheet has nine tabs, but the first tab must always carry the game title before anything else is opened.

Once the title is clear, nine layers of analysis unfold like nine geological strata, each with its own data origin and its own measurement limits.

The first layer is patch and meta. This is where data is born, not where it is consumed. In League of Legends, a fortnightly patch can lift a champion's win rate by several percentage points, enough to reshape the entire pick-and-ban rate. I track both presence and win rate, because a champion picked often but winning rarely is a sign of being solved, not of strength. In Counter-Strike, what matters is not weapon statistics but round economy and how the patch adjusts money rewards. In Dota 2, a large update can erase a playstyle that lasted an entire year. A good analyst must distinguish a mechanical stat tweak from a change driven by the community discovering a new exploit. The two look identical on a chart but have very different lifespans.

The second layer is tournament format. This is the most underrated layer. A single-elimination bracket has a far higher upset rate than a double-elimination one, simply because the strong team gets one chance. A BO1 differs from a BO3 which differs from a BO5, and the difference is not team quality but probability. The Swiss system pairs teams with identical records, creating an environment where a weaker side still has a path. When I model upset probability, I always start from the format, never from team names. Format is the skeleton; team names are only the flesh.

The third layer is roster and players. Here, metrics must be read in the context of the title. In tactical shooters, people look at opening-kill success and kill-death differential, but those two only mean something when you know who the in-game leader is. A strong IGL may not have pretty individual stats yet is the reason the team wins. In MOBAs, CS per minute and damage per minute are only trustworthy when you know which position the team is playing around. I always separate two questions: is this player strong, and is this team using him correctly. The two answers can contradict each other, and that contradiction is where I find mispriced transfer value.

The fourth layer is regional mapping. Regional strength shifts by title and by season. A region can dominate the group stage and then collapse in the knockout rounds for several years running. I record international results, talent-pool depth, academy output, and the overall health of the local ecosystem. Those four rarely move in the same direction. A region can have a deep talent pool but a weak ecosystem, and the result is that talent bleeds out to other regions. Import flow is a slow signal; it only becomes visible two to three seasons later, when people look back and ask why a generation was lost.

The fifth layer is club finance. This is the layer esports media usually avoids, because it is dry and easy to get wrong. Yet this is precisely where the most severe signals appear first: unpaid wages, a slot put up for sale, a sponsor withdrawing. I once tracked a transfer window where xG per 90 showed a young striker was being played out of position; the club then loaned him to a second-division side, and I was among the first to report it. The training data I used came from a contact I had met at a community seminar. Since then I have understood that transfer news comes not from intuition but from networks.

The sixth layer is rules and governance. Esports has a structural weakness: the publisher is both the rule-maker and a commercial stakeholder. No independent arbitration body sits above the publisher. As a result, compliance analysis is only as good as its source documents. At this layer, I check competitive integrity, transfer and registration rules, contract compliance, and protections for underage players. A match-fixing allegation cannot be assessed on feeling alone; it needs documents, dates, and involved parties.

The seventh layer is the risk profile. This is the layer I value most and the one most easily misunderstood. Risk breaks into several categories: competitive, financial, personnel, rules, public opinion, and systemic. An unratable risk profile does not mean a low risk profile. The distance between those two is enormous: low risk means there is evidence of an absence of risk; unratable means there is no evidence at all. In betting, this confusion is the cause of countless wrong decisions.

The eighth layer is media narrative and expectation. Every season has its own heat cycle: budding, accelerating, climax, then backlash. A narrative only sustains when it is supported by a foundation of real strength and a sufficient sample size. When media heat far exceeds the data foundation, a backlash is only a matter of time. I always compare market expectation against objective assessment and record the gap. That gap is where mispriced betting lines take shape.

The ninth layer is industry-wide transmission. From publishers upstream, through clubs and streaming platforms midstream, to sponsors and derivative markets downstream. This layer is the most title-sensitive, because patch cadence, revenue-share mechanics, and governance structures differ fundamentally between publishers. Running this layer without a confirmed title guarantees category errors.

The most frightening thing in this trade is not a wrong number, but a correct number read in the wrong context.

I have seen it many times. A clean metric is pulled out of the context of format and turned into evidence for a conclusion already decided in advance. That is how social media and sensationalist media operate, and it is also how we fool ourselves. Data does not shout, it whispers — and I have learned to lean in and listen. But a whisper can also be misheard if the listener has already decided what he wants to hear.

The Seoul night of 2026 remains the founding lesson. Back then I believed that if only the number was right, the truth would win on its own. I was wrong. Truth needs to be framed with empathy, with context, with an admission of the measurement's limits. An article about Ronaldo cost me three sleepless nights. The lesson was not to write less, but to write more correctly: always state the subject's strengths before presenting numbers, always close with an open question inviting rebuttal, and always note that data can change sooner than you think.

At the risk layer, the key principle is this: correlation is not causation. A team winning many matches when a player is on the field does not prove that player is the cause. The team might simply have an easy schedule. The patch might favour the whole team's style. The opponents might be in internal crisis. To separate causation from correlation, I need a control: compare the team's results with and without that player, in the same patch, against an equivalent group of opponents. Without a control, every conclusion is only a hypothesis.

And this is where the blank sheet that night became a valuable lesson. When the data source contains no title, no patch, no team, no player, then filling nine tabs with generic commentary is not analysis. It is performance. An honest analyst must mark "insufficient information, cannot assess" at each layer instead of substituting speculation.

There is a notable detail about how the blank sheet appeared. The analysis frame rendered intact, but the content cells were all empty. The signature of this failure differs from the signature of a source that inherently contains no content. The first case is a data-pipeline failure; the second is a meaningless source. Distinguishing the two allows automatic retries of blocked sources while correctly discarding genuinely empty ones. That is a technical lesson, but it is also a professional-ethics lesson.

I am not stopping you from betting — I only want you to understand what you are betting on. When a data pipeline breaks, the output still looks complete in form but is empty in substance. That is exactly what is dangerous in betting: an analysis sheet that looks professional but contains nothing trustworthy. The only way to protect yourself is to check the origin of every number before trusting it.

Looking to the next cycle, there are a few signals I will be tracking. First, field-completion rate per source — if one source domain keeps returning blank sheets, that is a sign of a technical barrier, not of a poor source. Second, the frequency of genuine timeliness assessments, because an undated analysis is an unverifiable analysis. Third, the gap between market expectation and objective assessment in upcoming tournaments, because that is where the value sits.

Without an audience, I hear the breathing of the match. And when there is no match at all, I hear my own breathing — the breathing of a man trying not to invent a story just to fill the void. The transfer market is a magic trick: look closely and you see the strings. A blank sheet is a poor magic trick, where the performer has nothing in his hands and is left with only one honest option: to say he has nothing.

We love esports for what data cannot reach — the moment a player steels himself in a deciding game, the roar of the crowd when a comeback lands. And we live on what data can reach — patch cadence, tournament format, player flow, a club's financial health. The analyst does not choose one side. The analyst stands at the boundary and keeps both sides honest.

That night, instead of publishing an empty analysis sheet, I shut the laptop and sent a short note: the upstream process had failed, and the title, patch, source, and date must be recovered before anything is assessed. The next morning, a colleague replied that this new rule should apply across the whole organisation. Refusing to analyse when data is missing is a skill. It is far harder than looking fluent through safe commentary.

The next cycle will begin with a very simple question that many overlook: which game title is this. Whoever answers it first holds the advantage before any number is spoken. Whoever skips it, however beautiful the analysis sheet, is still reading data in the dark.