The Empty Data Table in the Press Room: Discipline of the Beat Keeper

In June, in the press room of a major tournament, I sat beside an editor twenty years my junior. On the screen in front of him was the live data table pushed through by the official provider: a column of player names, a column of minutes, a column of passes, a column of expected goals. Four matches had finished. The expected-goals column was completely empty. Three of the four secondary columns were empty too. He did not see it. He typed into his draft: the team controlled the game with 62 percent of possession. I waited until he finished the sentence, then asked where the 62 percent came from. He pointed at the empty cell.

That afternoon I opened the ruled notebook I have carried for thirty-three years in this trade. One page per match. Each page divided into four columns in different ink colours: what I saw with my own eyes, what I heard from someone who was present, what I read from a traceable source, and one column left blank. The blank column is the most important one. It is filled only when I am certain what I am missing, and sometimes the only thing written in it is two words: not known. A notebook page with a blank column is an honest page. A news report with no room for that honesty is a report lying in the voice of data.

In Moscow I learned that a match can end, but its echo does not. I wrote that line into the spine of my notebook in the summer of 2026, and it still holds, except the echo has changed channels: from the stands into the hard drive.

Context: when football is packaged as a stream

Over roughly a quarter of a century, professional football has gone through a shift far quieter than the way the media describes it. That shift is not in tactics, not in transfer fees, but in this: every action on the pitch can now be assigned a quantified value, and that value is sold in real time.

The mechanism is simple enough. A data provider signs an exclusive contract with a competition organiser. At each match, their team sits in the stands, logging every pass, every duel, every shot, then pushes the data to a server within seconds. From that server the data flows down two branches. The first serves television, journalism, match-tracking apps and analytics platforms. The second serves the betting market, where every second of delay converts into money.

I have watched live data operate across several competitions, from the Spanish national league to national-team matches in Asia. Based on my experience of following matches, the quality of a live feed depends on three things few people notice. The first is how many data loggers are present in the stadium. The second is the quality of the network connection at that stadium. The third is the cross-checking process between the loggers on site and the reviewers at the operations centre.

Those three variables never appear on screen. The reader only sees the final result: a straight row of figures, no notes, no margin of error, no question marks. The interface has flattened the entire production process into a single truth.

What matters is that the core metrics of modern football are, in essence, conditional measurements. Expected goals, xG for short, assigns a probability to each shot based on location, angle, the body part used, the type of pass leading to the shot and the pressure applied by defenders. A shot from eleven metres, straight on, from a laid-on pass, carries a very different xG from a shot from eighteen metres through three bodies.

PPDA, passes allowed per defensive action, measures pressing intensity. The lower the figure, the harder the press. A side with a PPDA of 6.5 and a side with a PPDA of 14.0 can both win 1-0, but those two wins belong to two different sports.

Both xG and PPDA are good tools. The problem is that they are only good when there is input data, and they become dangerous when the input is empty and the user is not told.

The crux is here: an empty data cell on an interface is not displayed as an empty cell. It is displayed as a zero. And in most algorithms, zero is a valid value, while emptiness is undefined.

I once watched the analytics department of a Spanish second-division club build an opponent report from the live data of the opponent's last three matches. The second of those matches was played in heavy rain, the positional tracking system ran intermittently, and almost every action in the right half of the pitch went unrecorded. The report concluded the opponent attacked heavily down the left. That conclusion was true about the data and false about the football.

The core: dissecting a null result

I tell that story not to dismiss data. I tell it because in thirty-three years in this trade I have learned that the ability to read a gap is far harder than the ability to read a metric.

Back to the empty table in the June press room. It had four columns. The player-identification column was intact, because names are entered by hand and do not depend on the tracking system. The minutes column was intact, because it derives from the match clock. The remaining two columns, expected goals and passes, were empty, because they depend on the real-time event-logging module, and that module received no signal.

In other words, the system did not fail at every layer. It failed at one specific layer, and that layer dragged two dependent columns down with it. A disciplined reader spots the tell immediately: when two interdependent metrics are empty while an independent one is full, the fault lies in the pipeline, not in the match.

But a disciplined reader is not the default of this era. The default of this era is to fill the gap with a sentence that sounds reasonable.

In a deep analysis, nine dimensions need scrutiny. I name them in the language of my trade rather than the language of academia. One: tactics and technique, meaning the shape, the style, the way the system operates. Two: club finance and the transfer market. Three: results and the public-opinion cycle. Four: the league landscape and the team's position. Five: rules and compliance. Six: management and the dressing room. Seven: the risk profile. Eight: media narrative and expectation. Nine: transmission through the football industry.

With an empty table such as the one I saw, not one of those nine dimensions can be analysed responsibly. Not because the analyst is weak. Because the founding principle of the trade is that every conclusion must be anchored to a traceable information point. No information point, no conclusion.

Three years ago I sat down with three colleagues in a Seville hotel to review an investigation into a club's injury cycle. We had twelve information points. Each had its own source, its own date, its own reliability grade. By the ninth point we discovered that the minutes-played data we had used to calculate workload came from an aggregator site with no update date. We dropped that point. The piece lost a two-hundred-word section and kept everything else.

At La Masia every training session looks the same, but that boy changed every day. I wrote that line in my notebook in the 2026 season, when I spent nine months following a seventeen-year-old midfielder at the La Masia academy, a boy who made twelve appearances for the B team that season. News sites raced to publish sensational headlines; I checked his match data against the precedents of five young talents in the same position over the previous ten years. When the long-form series ran, a young coach at the club wrote to confirm every figure was accurate. For the first time I understood that following one specific thread creates a silent kind of power: trust.

But the more memorable thing from those nine months was not the figures that were right. It was the figures I chose not to use.

My notebook carries one entry in plain terms: the boy had six matches in which his completed-pass count exceeded his season average. I left it out. The reason was concrete: five of those six matches came against opponents whose morale had already collapsed, and he was sent on in the seventieth minute with the game long settled. Had I put those six matches on the front page, I would have sold a false story using a true number.

A match can end, but its echo does not. And the echo of a wrong metric lasts longer than the echo of a defeat.

The contrarian angle: the fashionable belief that more data means better analysis

A belief is in fashion across football, spreading from club analytics departments into newsrooms. It says, roughly: if you have more data, you will make better decisions.

I do not dispute the first half. I dispute the second, because it skips an intermediate variable: the ability to recognise when data does not exist. In practice, a system reports that it has no data if and only if its designer defined that state. Most systems do not define it. For them, an empty cell and a zero value are the same thing.

The consequence is one very specific kind of error, which I call false-confidence error. The user receives a densely populated table, believes they have been equipped, and makes a decision with more confidence than someone holding real data.

This error is more serious than having no data, because it does not incriminate itself. A person who knows they lack data goes looking. A person who believes they have enough stops.

A second fashionable belief travels with the first: that live data helps fans understand football better. Technically, live data serves three customer groups in a clear economic order of priority: bookmakers, rights-holding broadcasters, and fans. That order is not an accusation; it is a fact about cash flow. A live data stream is worth most during the match itself, because that is the only window in which it converts into betting stakes.

I once sat in an operations room in Barcelona and heard an engineer explain how his system distributed latency. He said it plainly: the customers who pay most receive data earliest, with gaps measured in thousandths of a second. Fans watching television receive it later. That is commercial architecture, not a conspiracy.

But precisely because of that, when a newsroom recycles live data without naming the source, without naming the timestamp, without naming the latency, that article is inadvertently advertising a product it does not control.

I do not hunt for the moment; I wait for the moment to stand up on its own. In the data trade, the moment that stands up is usually the moment an empty cell appears and nobody in the room will admit it is empty.

A historical precedent is worth recalling here. Before xG existed, English scouts judged strikers by an unnamed metric: the number of times the player was in the right position for where the ball would arrive. That metric demanded an observer sit long enough, across enough matches, and log the occasions when the ball never came. In other words, they measured by absence. My trade does the same thing today, only the environment has changed.

In Russia in 2026, in the round-of-16 match between Japan and Belgium, I witnessed a gap that could not be measured. Japan led 2-0 through goals from Genki Haraguchi on 48 minutes and Takashi Inui on 52. Belgium replied through Jan Vertonghen on 69 and Marouane Fellaini on 74, then won 3-2 through Nacer Chadli on 90 plus 4. The final goal came from a counter measured at roughly nine seconds, starting from goalkeeper Thibaut Courtois collecting the ball after a Japanese corner. I had believed data was supreme. At Rostov-on-Don that day, the tempo of the match and the emotion broke every statistical forecast, and what I recorded was not a metric but the image of head coach Akira Nishino picking up a tactical sheet off the grass.

One hundred days without fans, and I could hear the coach shouting more clearly than the ball rolling. In the summer of 2026, when competitions returned inside empty stadiums, I did not leave the second-division club I had been following for three years. Across that hundred days I phoned twenty-seven players, logging diaries of sessions in living rooms and matches on rooftops. The result was a long-form work called Football Without Fans. I did not gloss over grit. I stuck to rules and asked concrete questions: who lost a contract, who was depressed, who was forced into early retirement.

It was in that period that I recognised something which became a working principle. When the stands are empty, television data is still full, but data about people is empty. Nobody measures how many times a player opened his living-room door at three in the morning to run on the spot.

What to watch next

Back to the June press room. After I asked my question, the young editor cut that sentence from his draft. He did not replace it with another. He left the space blank and wrote a note in the margin: awaiting source verification.

That was the right action, and it was so small nobody noticed.

In Vietnamese football, this story is still current. Most reporting on the national league recycles basic metrics: passes, shots, possession share. Higher-tier positional tracking appears more and more often but is not yet widespread, and each time it appears it carries an old risk: the writer trusting the figure more than their own eyes.

The memory of the 2026 AFC U23 Championship final in Changzhou, when Vietnam's U23 side lost 1-2 to Uzbekistan in heavy snow, is one I keep. No metric measured what happened in extra time that day. You could only stand still and see.

I am not proposing we abandon data. I am proposing one simple rule, applicable immediately in any newsroom: every metric appearing in a piece must carry three things — a source, a publication date, and a note on the metric's limits. Anything short of all three does not make the piece.

Tomorrow, when another data table appears in some press room, I want the person in front of the screen to look at the empty cells before the rows of figures. That is the only way this trade keeps what it has gradually lost over twenty years: evidence-based slowness.

Three facts to anchor on

On 2 July 2026, in Rostov-on-Don, Japan led Belgium 2-0 and lost 3-2 in the World Cup round of 16.

Belgium's winner came on 90 plus 4 through Nacer Chadli, in a counter of roughly nine seconds that began from a Japanese corner.

Expected goals assigns a probability to each shot based on location, angle, body part and defensive pressure; PPDA measures passes allowed per defensive action.