The Two-Source Rule: Lessons From a Sports Analysis With No Data

In July 2026, in the newsroom, I filed a live piece on the World Cup semifinal between France and Belgium. In that piece, I wrote that France had 61 percent possession. The correct figure was 49 percent. I also called defender Lucas Hernandez Hernan three times. My editor called me in, put the printout on the desk, and asked exactly one question: where did you get this number?

Six years later, I sat in front of a different document. It carried an esports domain label, a skeleton headline, nine neatly numbered analytical sections, tables, a risk section, and recommendations. Skimmed, it looked like a professional report. Read closely, every data cell said: insufficient information. No tournament name. No team name. No player name. No patch version. No date.

A report with nothing to report.

That was the moment I understood something this profession teaches more slowly than any other skill: the greatest value of a sports writer is not in what he dares to assert, but in where he dares to stop when the data is not there.

When the live feed stumbles, I learn to tell the story more slowly.

This is the regular season. No World Cup, no Olympics, no single major event strong enough to pull all attention into one place. The calendar stretches out, a few matches each week, a few numbers each match, and the newsroom is always hungry for content. The pressure of a regular season is not as loud as the pressure of a final. It is more of a low hum: pieces must keep coming, data columns must be full, and empty space is not permitted to exist.

In that environment, an empty analysis is a real temptation. You already have a beautiful frame: nine sections, each with a table, a conclusion, a risk rating. All you have to do is fill it in. Fill in a team name. Fill in a number. Fill in a sentence that sounds reasonable. Nobody can check immediately. And the piece ships on time.

But readers do not remember pieces that shipped on time. They remember pieces that were right.

In 2026, after being called into that office, I spent a full month rewatching matches, logging every minute, every pass, every tackle. I built my own statistics table, and from then on, every number had to pass through two independent sources before it reached a sentence. The two-source rule is not a ritual to make the process look tidy. It is the only fence keeping me from confidently saying something false.

Viewers remember the goal; filmmakers remember the silence before the goal.

There is a question I ask myself before every piece: if someone cross-checks tomorrow, do I dare stand next to my own number?

That question forces me to separate three kinds of data. The first is data verified across two independent sources. The second is provisional data, drawn from a single source, not yet cross-checked. The third is empty data, meaning nothing at all. The gravest mistake in this trade is not writing a wrong number. It is merging all three into one place and calling the result statistics.

Take an example I used in a short documentary series about great teams that were forgotten. Liverpool in 2026-20 finished with 99 points from 38 matches, scoring 85 goals and conceding 33. Those are numbers you can look up and cross-check in at least two sources. But if I stop there, I am only re-reading the league table in a more solemn voice.

The story lives in the second layer of data: their xG ranged from 1.2 to 3.1 per match, and the squad averaged 112 kilometres of running per match.

The 112 kilometres, standing alone, means nothing. It only means something next to pressing time. I recorded that Liverpool under Klopp recovered the ball in an average of 7.2 seconds after losing it, roughly 1.5 seconds faster than the league average. Only then did 112 kilometres become a proposition: they did not run a lot because they liked running, they ran a lot because their system turned distance into pressure, and pressure into goals.

That is exactly where I learned to tell stories with data columns. I do not write that this team played well. I write: they recovered the ball in 7.2 seconds, 1.5 seconds faster than the rest of the league. Readers may not remember 7.2. But they will remember the feeling that this team never let opponents breathe.

Euro 2026 pushed that rule to another level. The Italy side I tracked decoded opponents by controlling the penalty area. In the final against England, Italy recorded 61 touches inside the opponent box, against England's 22. Italy's total passes were 847, at 92 percent accuracy. And there were 25 deliberate slips to stretch the opposing defensive line.

The 847 passes alone tell you nothing. They only tell a story next to the 22. One side had 847 passes and 61 box touches; the other had a deep defensive block and patience. Behind that block stood Giorgio Chiellini and Leonardo Bonucci, with Jorginho setting the rhythm in midfield. Italy did not win because they passed a lot. Italy won because they passed in the exact zone where the match was decided.

Core insight: a sports number has value only when placed next to another number, and both must survive the question of source.

Here, data only gives us the door, but the story is the one who turns the key.

That is also why I am wary of xG. xG is a good tool for describing chances, but it has been overused to the point of becoming the answer to every question it was never designed to answer. xG does not explain why a defender steps up instead of dropping. xG does not explain a player's form over the last three weeks. xG says nothing at all about refereeing standards, which in many matches carry more weight than any other variable.

When an analysis has only one xG column left to lean on, that is usually a sign the writer has nothing else in hand.

With esports, the problem is even clearer. I work at the intersection of traditional sport and esports, reporting on esports for the Chinese market, and I learned very early: in esports analysis, the first mandatory step is identifying the specific game title. A strong team in League of Legends is not automatically a strong team in DOTA2. A country with a foundation in CS2 does not automatically have one in Valorant.

Update cadence also varies so much that one framework cannot be applied across the board. Riot patches on a two-week rhythm. Valve patches more rarely, tied to majors. Tencent operates on a seasonal cycle. These three rhythms produce three completely different kinds of analysis. If you do not know which title you are talking about, every claim about meta, about roster, about champion pool depth is meaningless.

The empty report in my hands illustrates that in its most primitive form. It carries an esports label. It has no game title. Without a game title, it has no patch. Without a patch, it has no meta. Without a meta, it has no winners, no losers. Without winners and losers, every remaining section collapses into an identical string of empty cells.

What is notable is that the report did not fabricate. It chose to say insufficient information in each cell, with a clear note that any conclusion about teams, players, patches or finances drawn from that input would be a product of imagination, not analysis.

In this trade, that is a commendable act. But it is also a warning. Because with only a few lines changed, someone can turn that very empty frame into a report that looks highly professional, complete with team names, numbers and judgements.

The distance between an honest report and a fabricated one is sometimes just a few lines of filling in.

The transfer map is not on paper; it is in relationships.

In esports, the value chain begins with the game publisher. The publisher controls patches, schedules and event licences. Then comes the middle layer: clubs, tournament organisers, streaming platforms. Then the lower layer: sponsorship, derivative products, and the process of merging into the mainstream sports flow.

If you cannot identify the publisher, you cannot trace any flow. A change in patch frequency can upend how a team builds its roster. A change in qualification format can change the value of a tournament slot. Those effects do not appear on the scoreboard. They appear in personnel decisions, in contract structures, and in how a team allocates resources to its academy.

In a regular season, this is the kind of story I chase most. There is no final to call, so the sport's pulse lives elsewhere: contract flows, youth development systems, the data infrastructure of national squads.

A region's strength depends on the game title. A region strong in League of Legends is not necessarily strong in DOTA2. So before comparing regions, you must lock the title. Otherwise every regional ranking is a jigsaw puzzle built from pieces of different toy sets.

Once the title is locked, I look at four indicators: international results, talent pool depth, academy output, and ecosystem health. The fourth is the least discussed and the most important over the long run: how many clubs are still operating, slot trading activity, viewership trends, and how the publisher allocates resources by region.

Talent-flow signals work the same way. A wave of imported players can be a sign of a region lacking domestic depth, or it can be a sign of a region with surplus resources. These two readings lead to opposite conclusions. Without data on academies and youth contracts, you cannot tell which is happening.

Then there is club finance. This is where I want to speak most slowly, because it is where misreading is easiest.

In a financial report, an empty cell can carry two completely opposite meanings: there is no problem, or nobody checked. The two cannot be merged.

When I have no figures on sponsorship revenue, on publisher distributions, on salary bills, on equity injections, the only correct conclusion is: not yet screened. It is not that the club is healthy. It is also not that the club is in danger.

The same logic applies to contract structure. A long-term contract signed with a player past his peak is a risk signal. But to recognise that signal, I must know the age, the length and the compensation. Without those three, I have no right to attach a label.

And contagion risk from a parent company works the same way. If a club is backed by a real-estate group or a streaming platform, the health of the parent becomes part of the health of the club. Without knowing the owner, you cannot assess.

The same principle, this time applied to governance and compliance: silence does not mean innocence.

Each game title has a different governing body, and they handle violations differently. An integrity allegation, a contract dispute, an issue involving age rules or streaming regulation, all must sit inside the frame of reference of the relevant publisher.

If I cannot identify which frame of reference governs, I cannot draw a conclusion in either direction. And more importantly: an empty cell in a compliance section must never be read as no violations found.

This is the most common error I see in reports that look very professional. Tables are complete, every section is present, every conclusion is tidy. But read closely, and every cell is saying the same thing: we have no data. And instead of writing exactly that, people write: no risk.

That is a substitution. And it is more dangerous than a wrong number, because it cannot be detected by looking anything up.

There is a reflex it took me years to remove: the reflex to fill the void.

Sports writers are paid to have opinions. When data is empty, the natural reflex is to talk more to compensate. Add adjectives. Add comparisons. Add a judgement that sounds profound. The result is a long, fluent piece with nothing verifiable in it.

But there is a paradox: readers today have more cross-checking tools than at any point in history. A wrong number can be caught in minutes. A wrong name can be screenshotted and spread. The void is no longer a safe place to hide. It is the most exposed place there is.

The counterintuitive part is here: in a regular season, when there is no major tournament to anchor to, credibility does not come from the volume of pieces, but from the number of times a writer dares to say he does not yet have enough data to conclude.

I used to think that sentence was a sign of weakness. Now I think it is a sign of professionalism. A good editor will understand immediately: if the reporter says there is not enough data, the next step is to find a second source, not to write something wild to meet the deadline.

In the Chinese market, where I live and work, information gaps take many shapes. There are periods when certain regions face media restrictions. There are events outside the reach of mainstream coverage. My way of handling it is neither to complain nor to avoid. I shift the angle: read the tactics at the edge of the frame, cross-check historical head-to-head records, and measure the reaction of the local fan community.

When a restricted zone gets covered, the match starts being seen through different eyes.

And when I do that, I am forced to be honest about the certainty level of each claim. An observation from the edge of the frame is a lower-confidence observation than an official record. If I do not say so clearly, I am deceiving readers with a confident tone.

There is another point I want to state plainly, because it concerns how the media operates. The media loves underdogs. A weak team toppling a strong one is a story with traffic, emotion and shareability. But only by following a weak team all year do you understand the price of that miracle: training sessions short of players, cheap long trips, short-term contracts, players worrying about rent while playing.

The number on the scoreboard tells only the visible part. The submerged part lies where nobody measures.

In a year without football, I found the true pulse of this sport.

In 2026, when every competition was postponed, I was 26 and fell into crisis because there was no match to write about. I did not wait. I went looking for what operates quietly when the cameras are off: contract flows, youth development systems, the data infrastructure of national squads. Those things generate no headlines, but they generate the foundation for every headline that follows.

Since then, I carry one habit: before attaching a label to anything, ask why. In 2026 and 2026, as the Euros and the Club World Cup followed each other, I wrote a series on eight tactical models, labelling Manchester City as absolute control, and was called out for overlooking their ability to use Erling Haaland for fast counters. I labelled too early. I revised, moved towards an open framing, and accepted that a team can have many shapes depending on the moment.

That lesson and the lesson of the empty report are one. When you do not know the game title, do not analyse the meta. When you do not have a second source, do not assert the number. When there is not enough data, say there is not enough data.

One slip in front of the camera, a lifetime of rewriting the script.

So the question I leave for myself, and for those who do this work as I do: if someone opens your piece tomorrow and asks where this number came from, do you have enough confidence to answer with two names instead of a silence?