The Empty Heat Map: When Football Analysis Lives on Unsourced Conclusions

The Empty Heat Map: When Football Analysis Lives on Unsourced Conclusions

10:47 p.m., 14 March 2026, nineteenth floor of a building in Shenzhen. My second monitor shows a V.League 1 match dashboard: the player-name column empty, the pass column empty, the heat-map panel filled with one flat grey block, like a sheet of paper nobody has printed on. The data feed had dropped in the 30th minute of the first half, and nobody in the newsroom knew.

An intern messaged me on the internal chat: "Just write it by eye for now, we'll add the numbers tomorrow." I read it three times, then switched the screen off.

The pitch never lies — it was only me who once misheard a name. But an empty dataset lies brilliantly, because it says nothing at all while the reader fills in the rest. A ready-made headline, a grey heat map, a phrase like "a source close to the club", and a conclusion is born that nobody has to answer for. My job for eight years has been to spot those gaps before they get papered over with prose.

Today I want to write about the gap itself.

A country reading football through somebody else's data

Vietnamese fans in 2026 consume more football data than at any point in history. On television, every domestic league match carries a graphics layer showing possession and passes, and in big games a heat map after the final whistle. On social media, analysis accounts live off screenshots of metrics that nobody in Vietnam had access to a decade ago.

The supply side has not grown at the same pace. Most Vietnamese-language data content comes from three sources: international data providers, community accounts that translate them, and articles written from gut feeling with a few metrics attached for the look of objectivity. That third group is not lying. They are doing what the trade still rewards: delivering on time.

The 2026 World Cup expands to 48 teams, opens on 11 June 2026 at Estadio Azteca in Mexico City and closes on 19 July 2026 at MetLife Stadium. FIFA has scheduled 104 matches in 39 days. For the Vietnamese market this will be the biggest content cycle ever produced, because the match count rises 50% against the old 64-match format. Vietnamese newsrooms' verification capacity does not rise 50% just because FIFA changed the format.

What I have observed across years of working between two markets is this: the gap between the volume of data displayed and the volume of data verified keeps widening. A possession figure that is 3% wrong costs nobody a job. A player identity that is wrong does. Inside the space between those two error thresholds, an entire content industry operates comfortably.

Based on my experience following matches, I split Vietnamese football-data producers into four groups. Group one has raw data access and recalculates everything. Group two has raw data but only extracts, never recalculates. Group three uses other people's numbers through an intermediary. Group four writes first and looks for numbers afterwards. Groups one and two account for a small share of daily output. Group four accounts for the largest share, and gets shared the most, because it writes fast and writes with certainty.

I used to belong to group four.

The source ladder erodes when it crosses into Vietnamese

In this trade, information is tiered. Tier one is official club statements, governing-body rulings and match records. Tier two is journalists with a verifiable record — people who have been right many times and who correct themselves publicly. Tier three is media aggregating other people's reporting. Tier four is aggregator accounts and social-media rumour.

The problem is that the tier does not travel with the information when the information is translated. A tier-three line becomes, in Vietnamese, "a source close to the club". A tier-four line becomes "European media report". Nobody adds a single technically false word. But the reader receives a higher level of credibility than the claim actually holds, and then makes emotional decisions on that fake credibility.

Transfers are not about buying a player — they buy the story people want to believe.

I learned this in the summer 2026 window. While colleagues poured attention into Kylian Mbappe staying at Paris Saint-Germain, I followed a small detail: Erling Haaland's representatives had retained a law firm based in Manchester to handle his image rights. I called a source close to the Borussia Dortmund coaching staff, who confirmed the 60 million euro release clause had been triggered. Manchester City announced the Haaland agreement on 10 May 2026; the deal was completed on 13 June 2026. My story ran 48 hours before the official announcement.

That story did not win because I was smarter. It won because I called the right number, and because I accepted the phrase "according to a source close to the situation" instead of asserting certainty. That phrase is not decoration. It is a legal guardrail and an admission that I know which tier I am standing on.

In Vietnam that guardrail is often dropped. Domestic transfer news drops it most easily, because verifiable sources are scarce and because V.League 1 contract values are rarely published. With no official figure, a writer has two honest options: say the data does not exist, or say it is unconfirmed. The third option — quoting a plausible number — is the most popular, and the fastest way to destroy trust, because readers remember the number and forget the footnote at the bottom.

I once wrote a transfer piece using a fee I knew was an estimate. Three weeks later that fee was being cited by other outlets as confirmed fact, with my name attached. That was the lesson in how an estimated number turns itself into truth simply by being repeated enough.

Anatomy of an empty dataset

Back to that dashboard in Shenzhen.

In the analytics systems I have worked with, an empty dataset carries three traits, and each is dangerous in its own way.

The first is that field labels survive while values vanish. The panel still lists Possession, Key Passes, Successful Pressures. Only the numbers are missing. To a skimming reader, it looks structurally like a serious document. This is the most dangerous shape of all, because the form of precision remains after the content has gone. An article written from that panel will be full of terminology and empty of a single fact.

The second is that the category label is assigned by default. When a system returns the tag "football" for empty content, that tag almost certainly comes from a caller parameter rather than from the content itself. In journalism the same bug appears as choosing the section first and writing to fit it. I once took an assignment to write a "tactical analysis" of a match I had not watched, purely because the section needed copy. I wrote it. It was the worst piece of my career, and nobody caught it.

The third is that time-sensitivity is left unassessed. In football analysis, knowing whether a fact expires matters as much as knowing whether it is true. Injury news expires in hours. Club accounts expire in years. An article that fails to separate the two will either panic readers for nothing or let them miss a real change.

Those three traits combine into what I call the unsourced conclusion. It has the shape of analysis, the tone of analysis, and no footing at all. In a major-tournament cycle, as matches multiply and the gap between them shrinks, unsourced conclusions breed fastest, because they are the only content that can be produced without waiting for data to arrive.

I do not write to be loved; I write to make people stop. But stopping because of a shock line is not the same as stopping because of a fact. The first drives traffic. The second drives credibility. The trade pays for the first immediately and for the second very slowly.

The heat map is the new fortune-telling

Of all the data tools in use in Vietnam, the heat map is the most abused and the most misunderstood.

A heat map is fundamentally a compression. It takes thousands of ball touches across 90 minutes, blurs them, and colours them by density. Every compression loses information, and what is lost here is context: whether a player touched the ball in that spot because of his tactical job or because the opponent forced him there.

A striker instructed to drift to the left wing and drag centre-backs out of position will show a heat map with a cold zone right in front of goal. A reader who studies the map without watching the footage concludes he is harmless inside the box. That conclusion can be entirely wrong, and it is wrong systematically, because every decoy striker gets judged the same way.

Expected goals models, known as xG, carry the same limits. The model estimates the probability that a shot becomes a goal from position, angle, shot type and a handful of other variables. It does not know the goalkeeper was badly positioned, that a defender screened the view, or that the player was carrying an injury. Expected goals against, xGA, is the defensive mirror. Passes allowed per defensive action, PPDA, measures pressing intensity, and by convention a lower value means more aggressive pressing.

Used properly, those three metrics separate process quality from result luck. Used as labels, they become shields. I once read a Vietnamese analysis declaring a team's pressing terrible on the basis of a high PPDA, when the footage showed that team deliberately ceding the ball and pressing only in a chosen zone. A high PPDA in that case is evidence of a plan, not of laziness.

When the stands are empty, I hear the breathing of the match — and I find my own voice. I wrote that in 2026, and it holds in another sense: with no external noise, the writer is forced to listen to what is inside. A heat map is external noise, rendered in colour.

Importing the financial framework without importing the data

The second major error in Vietnamese football analysis is importing European financial frameworks without importing the accompanying data.

In the Premier League, the Profit and Sustainability Rules cap a club's losses at 105 million pounds across three seasons. Everton were docked 10 points in a decision announced on 17 November 2026; that was reduced to 6 points on appeal on 26 February 2026. Nottingham Forest were docked 4 points in a decision dated 18 March 2026. Against Manchester City, the Premier League's independent commission brought 115 charges in February 2026, with the hearing beginning in September 2026.

In La Liga the mechanism is entirely different: the league imposes a wage-spend ceiling tied to each club's verified revenue capacity, and that limit governs which players a club can register. The two systems are not interchangeable. Applying one regime's benchmark to the other's clubs is a basic technical error, and I see it regularly in Vietnamese copy.

The deeper problem is this: to claim a Vietnamese club is overspending, you need its revenue. Full financial statements for most V.League 1 clubs are not published in a form that lets an outsider calculate the wages-to-revenue ratio. By international convention that ratio above 70% is the high-risk threshold, and a top-wage-to-average-wage ratio above four times signals dressing-room imbalance.

With no denominator, every conclusion about club finance is a guess delivered in the voice of an expert. I once watched a Vietnamese account compute a V.League club's "losses" from estimated transfer fees and an estimated wage bill, then conclude the club would face financial sanction. No mechanism in Vietnam operates that way. The post was shared thousands of times.

The phantom number nine does not exist on the pitch, but it lifts the trophy. A phantom balance sheet does not lift anything except engagement.

The phantom number nine and the lesson of being contradicted

In 2026 I wrote a deliberately provocative piece on France winning the World Cup in Russia. My argument was that Didier Deschamps did not need a real centre-forward, that Olivier Giroud was a mobile decoy, and that France's counter-attacking block ran around a phantom number nine. A group of young coaches pushed back hard online. After France beat Croatia 4-2 in the final on 15 July 2026, some international analysts began citing the idea.

What made the piece survive was not the tone. It was one specific fact: Olivier Giroud finished that tournament without a single shot on target. A starting striker across nearly every match for the world champions, and not one effort on goal in the entire competition. That statistic turned a shock argument into a testable hypothesis, and shock is only permitted after the testing.

That is the line I try to hold: the headline may provoke; the body must pay the debt in data.

But I have to name the downside. An argument that works becomes a template. After 2026 I was regularly asked to "find the false nine" in every team. Some teams genuinely played that way. Many did not, and I forced them into the frame. When an angle succeeds, it imposes an intellectual penalty: you start watching every match through it. So I set myself a rule — every month, pick one match and analyse it through a lens entirely different from the one currently working for me.

Esports betting: the patch that never arrived

There is one field where the data gap does not merely produce weak writing but real harm: esports betting.

Betting in esports erodes competitive integrity faster than in traditional sport, because the regulatory framework lags. In traditional sport, betting-market monitoring was built over decades, with links between league regulators, bookmakers and investigators. In esports, most competitions below the top tier have no equivalent monitoring, no reporting obligation and no clear sanction.

The weak point is small tournaments, where player salaries are far below what a fixed result can yield. A qualifier nobody broadcasts can still carry a market. Nobody is watching footage to spot an odd piece of play in the 30th minute.

For Vietnamese audiences this is a direct issue, because the domestic esports community is large and tightly linked to cross-border betting platforms. My position is that regulators should require any competition with a betting market to publish raw data on anomalous plays, the way football federations publish lists of monitored matches. With no public data there is no verification. With no verification, fixing rumours replace facts, and clean players suffer the same public verdict as cheats.

The 23rd minute and the limits of reacting fast

On 18 December 2026, during the World Cup final in Lusail, I wrote and published a piece immediately after Lionel Messi scored the opening penalty in the 23rd minute. My headline was provocative: if Argentina win, the media will be wrong to call this the greatest final ever. I argued the goal came from individual errors in the French defence.

The match ended 3-3 after 120 minutes and Argentina won the shootout 4-2. My piece was mocked heavily. When I rewatched the footage and checked the data, I found I had missed the most important detail: Messi scored twice in that match and registered the highest number of shots on target of anyone on the pitch. I corrected the article publicly.

My mistake was not holding a different view. It was speed. I wrote while the match was still running and turned an observation about one passage of play into a conclusion about an entire final. That is exactly the error I described earlier, except that this time I was the one committing it.

Since then I separate two things: opinions meant to invite debate, and claims that need a basis. The first may open with an uncomfortable line. The second must open with a fact. Confusing the two is the fastest way to lose both.

The empty stands

On 29 February 2026, Watford beat Liverpool 3-0 at Vicarage Road, ending Liverpool's 44-match unbeaten run in the Premier League. I sat at home in front of a screen, with no crowd noise and no stadium atmosphere, and wrote a piece predicting that football without spectators would make teams sloppy.

The Premier League returned on 17 June 2026 behind closed doors. Many matches ran at a higher tempo than usual, partly because in-stadium media pressure eased and teams played with more tactical freedom. I predicted wrongly, and I was wrong for a very specific reason: I took my own feeling in a quiet room and assigned it to twenty-two players on a pitch.

Stadium emotion is a data source, but it is data about the writer, not about the match. Since then I have added a step to my process: after drafting a subjective judgement, I check it against xG, pressing volume and physical output before publishing. That step does not make the writing better. It makes it less wrong, and that is the target.

Where I might be wrong

Everything above has one large weakness, and I want to state it plainly.

The first weakness is cost. A process demanding every claim be tiered and sourced slows output sharply. In a World Cup of 104 matches in 39 days, speed is not a convenience; it is a newsroom's condition for existing. If every piece waits for verification, Vietnamese readers will read the outlets that do not wait. I have no complete answer to that, and anyone who says they do is selling something.

The second weakness is infrastructure. In leagues with dense data, holding to verification is a matter of discipline. In leagues with thin data, holding to it means barely writing at all. There are V.League 1 matches where all I have is footage and my own eyes. Telling readers that everything on the pitch needs a number to be confirmed is a useless position for most fans, because they do not watch through numbers. They watch with their eyes, and their eyes are better than I assume.

The third weakness is the heat map itself. Calling it fortune-telling is too strong. A heat map spreads no superstition. It is a tool misused by people who do not know what it compresses. The tool is blameless. The users are not, and I have been one of them.

The fourth weakness is the moral hazard of writing about your own professional errors. When a writer becomes known for self-correction, he may unconsciously write more carelessly, knowing the correction will later be read as a virtue. I noticed this while preparing a long correction piece. If you are preparing a long correction piece, there is a chance you already knew you were wrong before publishing and published anyway. That is not transparency. That is calculation.

What I will verify over the next 39 days

The name I once mispronounced was the most expensive lesson journalism ever gave me. It taught me that the distance between hearing something and understanding it is the distance between two different professions.

From 11 June 2026, when the opening match kicks off at Estadio Azteca, I will set myself a measurable test. Over the first thirty days of the tournament I will count Vietnamese-language analytical articles that state their data source and retrieval date, against the total number of analytical articles published in the same window. My prediction: that share will stay below one third.

If I am right, Vietnam's football problem is not a shortage of data. It is that a content industry learned how to present data before it learned how to trust data.

The silence after the whistle is the passage I most enjoy writing. When the 104 whistles of this World Cup have sounded and faded, what remains on my screen will not be a heat map. It will be a blank page and one question worth answering: across those 39 days, how many lines did I write that I dare stand behind, and how many times did I stay silent exactly when I should have.