The Empty Risk Matrix: When Football Analysis Deceives Itself

An evening in the video room

May 2026. The first Bundesliga rounds after the pandemic shutdown were played in empty stands. I sat alone in my video room in Hamburg, rewinding four consecutive matches of a mid-table team I had been tracking. On my notepad, the "risks" section was blank. No positional errors, no signs of fitness overload, no noticeable friction. A file so clean it felt slippery. Three weeks later, that team lost four in a row and conceded eleven goals.

The problem was not the team. The problem was the piece of paper.

With empty stands, tactics reveal themselves as if under a microscope — but a microscope only shows what has already been placed on the slide. If I forget to put the specimen there, it will reflect my own emptiness back at me, and I will call that emptiness "fine". From the HSV video room, I see the Bundesliga as a chessboard, and for the first time I realised I had misread the entire board because one square had been left blank.

The mechanics of a file with no data

In analytical work there is a failure more damaging than a wrong conclusion: drawing no conclusion at all while still presenting a full set of tables. An empty risk matrix does not say "this team is safe". It says "this analyst has not yet found anything". Those two statements are worlds apart, yet on paper they look identical — the same boxes, the same dashes.

In football we meet this mechanism constantly. A new centre-back arrives, the first three matches end in clean sheets, the spreadsheet reads "87% tackle success", and the whole coaching staff exhales. But if those three matches came against weak attacks sitting deep in a low block, then 87% measures nothing except that the player has never been tested. The data still looks clean. In reality, nobody has checked anything.

When the Bundesliga returned for the 2026-20 season, I analysed 89 matches without crowds. Home advantage disappeared, pressing intensity fell by 8.3%, yet pass accuracy rose by 3.2% — simply because players could hear each other without the roar of a crowd on top. That was a finding you can measure. But what kept me awake were the things I could not measure: the sound of players shouting at each other in an empty stadium, the way they turned to look at one another when nobody was cheering, the long silence after a conceded goal.

We have turned football into a collection of photographable metrics, and then forgotten that most of the truth of a match lies outside the frame.

When the numbers fall silent

For years, xG has been praised as the magic mirror that reflects a match's true nature. It is useful, I do not deny that. But xG cannot explain why a defender left his position in the 78th minute, why a midfielder passed backwards instead of threading a ball through, why a coach made his substitution exactly when his team was on top. The decisions that turn a match live inside human heads, not inside a probability distribution.

Worse still, once a metric becomes a target, people start playing to produce that metric. Mid-table Bundesliga sides now use physicality to turn football into athletics: most distance covered, most pressing, most duels. Gegenpressing was once a breakaway idea; it has now been decoded into a fitness drill that everyone knows by heart. When every team presses the same way, pressing stops being a tactic and becomes the minimum wage for getting on the pitch.

And in that treadmill, the data table grows emptier in an even more dangerous way. Because if everyone runs, everyone presses, then the metric can no longer tell you who is who. To find the difference, you have to watch how a team escapes a press — the rhythm of moving the ball through three lines in two seconds — not count the kilometres they ran.

The blind spot: a blank table is not good news

This is where I want to linger, because most readers get it wrong.

A sports report with an empty "risks" section is always read as a reassuring signal. No injuries recorded means the squad is healthy. No disputes recorded means the dressing room is calm. No tactical errors recorded means the system is working. That logic sounds reasonable, but it reverses the causal relationship. The absence of evidence of risk is not evidence of the absence of risk.

In my trade this is the deadliest trap, and it is especially dangerous for anyone who writes fast. After the final whistle, the pressure to publish within two hours forces the writer to fill every empty box with something that sounds plausible. No injury news? Write "full squad available". No statements from within? Write "no signs of unrest". And so an empty file becomes a clean file, and readers trust a tranquillity that has never been verified.

I remember a midwinter match in which the away side conceded twice in the last ten minutes. The stats showed 61% possession and twice as many shots as their opponent. Anyone reading the table would think a dominant performance had been undone by luck. But rewinding the tape, I counted fourteen occasions on which the away side lost the ball immediately after crossing the halfway line, and all fourteen originated from their midfield being squeezed into the wide channels. No metric on that sheet said so. I had to count it myself. The miracle on the pitch is only a calculation the audience has not yet read — but there are also calculations the analyst himself has not bothered to read.

Injury and the price of silence

There is another data zone that is routinely left blank: injuries and comebacks. The report says "player X returns from injury" and stops there. Nobody records that he endured four months of rehabilitation, that he is expected to "prove himself" in his very first start, that every sprint in the opening twenty minutes is a gamble with his own ligaments.

Demanding that a player just back from injury prove himself immediately is a cruelty disguised as professional standards. It creates re-injury pressure that no statistical box records. The crowd reads the team sheet, sees his name among the registered players, and assumes he is ready. But readiness is not a line of text. It is a long process for which the team sheet has no box.

At 63, I no longer chase the ball, only the intent behind it. And the intent of a player just back from injury rarely appears in his numbers from the first match of his return.

Context that is not on the tape

There is one section I always keep to myself in every analysis, and I call it context. Tactics do not exist in a vacuum. They exist inside a congested fixture list, in the cold of a January evening in Germany, in the mood of a dressing room after three straight defeats, in the noise of a full stadium and the silence of an empty one.

The regular season is a long-distance race, and the most important signals usually appear before they become headlines. A team's PPDA declining over three matches is not because the coach changed his philosophy, but because the midfielders' legs have accumulated enough fatigue that they can no longer spring up in time. A centre-back standing three metres higher than usual may be a tactical decision, or it may be a sign that he is trying to compensate for a teammate who is half a step slower.

Those things take time to read, and they rarely appear in any automated data table. They live in the matches you have to watch over and over, sometimes with the commentary muted so you can hear the rhythm of the team's breathing.

Open hypotheses, not rushed conclusions

I set myself one rule: if the causal chain is not closed, I must clearly mark it as "this is still a hypothesis", rather than pretend I have understood everything. With the team in my video room that May 2026 night, I was too hasty. I saw a blank table and called it calm, when I should have called it by its real name: I knew nothing yet.

The lesson is not to abandon data. The lesson is to keep one box reserved for the unknown, and to label it correctly. An empty risk matrix must be read as "risk undetermined", never as "low risk". That is the boundary between an analyst and a typist who knows how to use the right template.

What I will verify next round

Over the coming stretch of the season, I will pick the three teams with the prettiest defensive statistics and re-examine every goal conceded that was missed within the data block. My main scenario: at least one of those three will concede in a way the metrics never warned about — not from a high-quality shot, but from a midfield turnover that the spreadsheet mislabelled as "harmless". Reversal condition: if all three preserve their defensive structure through the round without increasing their running volume, I will admit I was wrong and record the reason.

If the hypothesis holds, I will take it as evidence of what I have believed for years: an empty data table is not the frightening thing. The frightening thing is that we have learned to read that emptiness as a compliment. In the video room in Hamburg, I keep an old habit: before trusting a number, I ask where it came from, and what it is hiding. Football does not reward the fastest writer. Football rewards the most careful reader.