Empty Data and the Discipline of Not Fabricating in Esports Analysis

At 2:14 a.m. on August 13, I reopened a nine-section analysis file on my screen and read it from top to bottom. The game-version section: empty. The tournament-format section: empty. The roster section: empty. Nine sections, and each one closed with the same line — insufficient information to assess.

It took me four hours to produce a document whose only conclusion was that the source material did not exist. In the esports industry, that is the kind of document that goes in the bin. But after six years of tracking data, I think it is more honest than most of the analysis published that same week.

A process that does not begin with prose

Deep professional analysis runs on two layers. The first layer extracts: read the source, pull out discrete information points — a figure, a timestamp, a quote, an event — and label them, along with related entities, time sensitivity, and source quality. Only the second layer analyses: it places those points across nine dimensions, from game version and tournament format to roster, region, club finance, governance compliance, risk profile, and public narrative.

When the extraction layer returns an empty set, the analysis layer has nothing to consume. Without a game title, the correct analytical unit cannot be selected — League of Legends, Dota 2, CS2, Valorant, and Honor of Kings operate under entirely different conventions. Without a version number, there is no way to say which direction the meta is shifting. Without a team name, every roster judgment is just inference dressed in terminology.

What is worth noting is that the empty document was not worthless. It diagnosed a specific fault: the source had been truncated or parsed in the wrong format. A self-contradicting report — instructing the reader to identify entities from the information points above, while providing none — shows that the data pipeline broke at the ingestion stage. The emptiness here has a shape, and that shape points in exactly one direction.

What the industry fills the gap with

Most esports newsrooms do not have the option of stopping. They have publishing schedules, quotas, pageviews. And when the source is thin, they fill with the cheapest thing to produce: conjecture presented as information.

I keep a tracking sheet for every transfer window, labelling each item by evidence tier. Tier one is an official announcement from a team or tournament organiser. Tier two is a contract record in a public database. Tier three is a report by a journalist with a correct-call history. Tier four is an article aggregating another outlet. Tier five is a social media post with no sourcing.

The distribution across that sheet barely moves from season to season: tiers one and two are always smaller than tiers four and five. Most of the transfer traffic fans consume has passed through at least one amplification loop before reaching them. Each loop makes the story look more certain without adding a single piece of evidence.

That is why I write what I write: transfers are a market, and a market has no emotions — only liquidation value and investment value. A rumour has no liquidation value. It has only amplification value.

Evidence from elsewhere

This limit is not unique to esports. On 22 November 2026, I tracked the PPDA figures for Saudi Arabia against Argentina in Qatar. Saudi Arabia pushed its defensive line very high and sprang Argentina into an offside trap ten times in a single match. The final score was 2-1. When the data speaks, the whole stadium falls silent.

But two years later, I was wrong myself. My xG model for Euro 2026 leaned toward France and Kylian Mbappé. Spain won the title with a lower xG figure, and Lamine Yamal reached the final at sixteen years and 362 days old. I rewrote the entire conclusion section that night, and added a permanent section to every report since: the limits of the data.

The limits of the data lie in the fact that it only describes what has already been recorded. It cannot capture the variable of outstanding individual talent, and it does not generate meaning on its own — the analyst assigns the meaning, along with every one of their own biases. Remove that section and a report becomes nothing more than a league table rewritten in a confident voice.

The contrarian angle

The usual reaction to an empty report is to blame the analyst. I think that diagnosis is backwards.

The problem is that the entire system encourages analysts to speak regardless. In football, xG models are publicly audited: every wrong prediction leaves a trace, and an analyst whose model drifts repeatedly loses credibility within a few seasons. Esports has no such mechanism. Nobody publishes an accuracy scoreboard for transfer predictors. Nobody checks roster forecasts against actual outcomes window by window.

The result is that a wrong prediction costs nothing, while saying there is not enough data yet costs a great deal — it is read as weakness. That incentive structure does not produce analysis. It produces confidence.

And here is the counterintuitive point: a full analysis is not necessarily better than an empty report. A full report can be wrong in every section and nobody will notice. An empty report admits its limits on the first line, and limits can be verified.

I do not commentate football. I read football through charts — and in the same way, I read esports.

The signal for the next cycle

There is a change under way, slow but measurable. Professional esports teams are gradually moving from social media announcements to structured contract disclosures: term length, release clauses, effective dates. Every time a team does this, one layer of rumour is removed from the system.

The signal worth watching this transfer window is which teams are willing to publish contract structure. Behind every shot that hits the crossbar lie thousands of data points whispering that nobody has the patience to hear; the same is true behind every contract.

As for that nine-section file, I kept it. It sits in a folder called not enough data, alongside others of its kind. An industry only matures when it starts keeping a record of what it does not yet know.