The Empty Dossier: Reading What No One Wrote Down

In the 34th minute, Aaron Hunt went down in the middle of the pitch. Hamburger SV against RB Leipzig, in the season when I was the club's only doctor-liaison reporter. I was sitting less than twenty metres from the touchline, eyes fixed on the live GPS tracking board. Hunt's deceleration speed in the first half was 7.2 metres per second. He went into the medical room, then came back out. Ten minutes later, that figure was 5.8.

No announcement came from the technical area. No assessment record existed. The coaching staff let him play out the match.

When I tried to enter the men's dressing room to cross-check the data with the club doctor, an assistant coach blocked me at the door: "Women don't understand tactics, get out!" I stood still. I did not argue, did not raise my voice. I waited for the doctor to come out and confirm what I had already seen on the screen.

That night I opened the match's official medical file to compare. Forty pages of forms. Every administrative signature in place. And empty on the only thing that mattered. No hamstring note. No test result. No name attached to the decision that sent Hunt back onto the pitch.

An injury dossier does not know how to lie — only the person reading it knows how to hide the truth.

From then on I developed a professional habit that later became my main working method: read the blank sections before reading the written ones.

Professional football runs on an administrative paradox. Federations require clubs to submit a medical report after every match. But the minimum standard for that report is so low that a blank sheet with a signature still counts as valid. Nobody checks whether the doctor recorded enough, only whether the doctor recorded anything. The result is that within the same league, two clubs can submit two completely different classes of document, and both are accepted equally.

This differs fundamentally from Formula 1, where a driver's injury record is bound by safety inspection protocols, by disclosure obligations to the governing body, and by the commercial pressure of sponsors. A driver with a wrist injury cannot hide it from the medical commission before the next race. A midfielder with a grade-two hamstring tear can — if the file is written skilfully enough.

That skill has a name in the trade: a file that is too clean.

I have spent most of my career categorising the types of medical documentation produced in professional sport, and I now believe only three types are worth analysing. The first is the too-clean file: every metric agrees, every timestamp is round, there are no internal contradictions. The second is the noisy file: too many technical details, too much jargon, designed to make the reader give up before finding the anomaly. The third, and by far the most dangerous, is the empty file.

The first two are intentional. The third is usually the consequence of something far more mundane: a failure of record-keeping capacity.

In March 2026, the Bundesliga stopped. I was twenty-nine, working mid-level at a sports data analytics company in Hamburg. During the shutdown I discovered something I had until then only suspected. Werder Bremen and Schalke 04 had no full-time club doctor. They used outsourced medical services, by contract, by appointment. Which meant that from May onward, when football returned with a compressed calendar, those two clubs entered the highest-risk phase of the season with nobody holding the full injury history of their players in their head.

I built a spreadsheet comparing the injury records of 412 Bundesliga players across five seasons. When the ball rolled again in May, the data showed hamstring re-injury rates up 19 percent compared with previous seasons of similar fixture density. That was a quantitative finding. But what made me pause longer was the annotation column: a disproportionate share of re-injuries occurred at the clubs with the thinnest medical files.

That correlation does not prove causation. It points to something simpler: where nobody records, nobody sees the early signal.

Three years of pandemic taught me that the gap between two football clubs can always become a bridge.

At this point the story leaves Germany.

World Cup 2026 in Russia. Germany were eliminated in the group stage after a 0-2 defeat to South Korea, in a match where they held only 35 percent possession. The media piled blame on Mesut Özil. I was twenty-seven, working for an independent sports outlet, and I had a question nobody was asking on the evening bulletins: could Özil's numbers at that tournament have been affected by a pre-existing physical condition?

I approached the national team doctor, verified through the treatment log, and cross-referenced two further independent medical sources. The result: Özil had undergone three corticosteroid injections for his back before the tournament began. His pressing capacity was down 28 percent compared with the qualifiers.

A back ache can tell you a story about dressing-room politics, if you are willing to listen.

But I also have to be honest about my limits. I did not examine Özil. I did not read his MRI. I have no authority to diagnose. What I could do was show that a tactical conclusion drawn on an undisclosed medical foundation cannot be treated as a complete conclusion. That is the line between analysis and judgment, and I have learned to stand on the right side of it.

I do not trust a medical report before I understand the pressure weighing on the doctor's signature.

That signature carries at least four pressures at once. Commercial pressure from the sponsor, who does not want their star sitting out. Sporting pressure from the coaching staff, who need points. Contractual pressure from the player himself, who can lose his place if he rests for three weeks. And professional pressure from the doctor, who knows that a signature that is too cautious may cost him a contract renewal at the end of the season.

Those four pressures explain why the empty file is so common. Writing nothing is the safest option. A blank space cannot be rebutted before a disciplinary panel.

When the dressing-room door closes, I understand that tactics are not on the whiteboard.

And that is where I have to be precise about my method, because it is easily misread.

When I say that the gap is data, I do not mean that every gap is a conspiracy. In this trade it is very easy to slide from analysis into conspiracy theory, and I have seen enough colleagues slide that way to know how fast it destroys credibility. So I categorise gaps into three distinct groups, and each demands a different treatment.

The first group is structural absence. The club has no full-time doctor, no archiving system, no one responsible for data continuity. The gap here is a consequence of missing resources. The treatment is not to hunt for someone hiding information, but to look for the traces of that shortfall — the volume of outsourced services, the number of doctors per head, the update frequency of records. Werder Bremen and Schalke 04 in the 2026-2026 season belong to this group.

The second group is deliberate absence. The information exists, was recorded somewhere, but was excluded from the published document. The tell lies in the shape of what remains: a report with deliberate omissions usually has a structure that is too even, timestamps that are too round, and not a single surplus detail. Özil's 2026 file sits closer to this group.

The third group is unassessed absence. Nobody ever asked the question, so nobody ever wrote it down. This is the most common group and the least discussed, because it has no villain. It only has indifference.

These three groups lead to a practical consequence that I apply to every piece I write.

First, I never conclude from a single source. Before writing anything about an athlete's physical condition, I cross-check a minimum of three independent medical sources. If there are only two, I write it as a question. If there is only one, I do not write it.

Second, I use the language of probability rather than absolutes. I write "the file indicates a likelihood" rather than "the player is injured". The difference sounds small, but it is the entire distance between a journalist and someone playing doctor.

Third, and most importantly, I record the gap itself. Every dataset I build has a dedicated column for cells with no data, annotated with the reason for the absence wherever I can determine it. That column is usually the longest one.

There was a period when I thought technology would solve this problem. Wearables, accelerometers, positioning systems, heart-rate monitors — all of them became cheaper and more accurate with each season. Clubs now capture more physiological data than at any point in the sport's history.

But more data does not mean more understanding.

I have watched this trend in football and seen it repeat almost intact in other sports: the heat map has become a new form of fortune-telling. People look at the red patches on the pitch and believe they are seeing a player's true role. But a heat map only shows where a person was. It does not show where he was told to be, where he was forced away from, or what he was concealing by moving constantly. The data thickens, and the meaning thins.

That is why I do not trust the abundance of data as a guarantee of objectivity. Data does not select itself. A hamstring test is only performed if someone decides to send the player for a test. An injection is only entered in the log if someone decides to enter it. The entire medical record of a sport is the product of thousands of small decisions about whether to record or not.

Data has no gender. Only the person reading the data carries prejudice.

That is not a slogan I deploy against the people who once threw me out of a dressing-room door. It is a technical description. When I presented Hunt's deceleration figures — 7.2 down to 5.8 metres per second — the assistant coach that day did not dispute the numbers. He disputed the person holding the numbers. Those are two different things, and that difference is the whole problem.

So I moved to a different strategy. I wrote more drily. I cited a source for every fact. I described specific injury counts, speeds, intensities, timings, instead of using adjectives. My writing became less appealing, and precisely for that reason harder to rebut. Male colleagues had to read carefully before speaking up in denial, because to deny a sourced dataset you must produce another sourced dataset.

But if I had stopped there, I would have been only an auditor. And auditing is not what I am after.

What I am after is the central question of the trade: what actually happens to an athlete's body in the interval between two matches.

That interval is almost entirely blind. No cameras there. No spectators there. There are only the doctor, the player, the fitness coach, and conversations that are not recorded. I have spent years trying to reconstruct that interval from indirect fragments. An unexplained rest day in the training schedule. A substitution in the 63rd minute instead of the usual 70th. A sprint profile quieter than every other match that month. A player talking to the club doctor for longer than two minutes in the tunnel area.

None of those fragments is evidence. But when they are laid on the same timeline, they form a shape. And that shape usually matches what is announced two weeks later as a sudden event.

Injury at the elite level is rarely sudden. It is the final outcome of a chain of decisions that began long before the moment the athlete went down.

This is where I diverge from most people in this business.

The most popular way of telling an injury story is dramatic. One moment, one cry, one season collapsing. That telling sells, spreads easily, and is almost always wrong about the mechanism. A hamstring injury does not happen in the 34th minute. It happens over the preceding three weeks, in training sessions where nobody counted the sprints, and the 34th minute is merely the moment the system reports the error.

The second most popular telling is heroic. The athlete returns from injury, plays his first match, scores, and the story closes. This telling ignores what interests me most: the degree of tissue remodelling, the recovery of neuromuscular control, and the re-injury risk over the following six months.

I understand the pressure to return early. I have seen it from the inside. A player in the final year of his contract, a coach on a hot seat, a club needing points to survive. Those three pressures combine into a medical decision that nobody in the room wants to name correctly.

And in almost every such case, the empty file is the loyal companion. It does not lie. It simply says nothing at all.

That is why I argue the big debate of this decade in sports medicine is not the debate about treatment methods. Treatment methods are advancing quickly, and most of those advances are real. The real debate lies elsewhere, and it gets far less attention: the debate about the obligation to record.

A sport that records fully would automatically expose decisions that can currently be hidden in silence. If every posterior-chain test had to be logged, if every corticosteroid injection had to appear in a published record, if every instance of a player competing for more than sixty minutes in a medically sub-optimal state left a trace — commercial pressure would still exist, but it would have to operate in the light.

Light does not make anyone more moral. It only makes pretending more expensive.

And that, I think, is what clubs will soon have to face. As the cost of physiological analytics falls and the number of wearables rises, the gap between clubs that record well and clubs that record poorly will stop being an academic gap. It will be a points gap.

A club that knows its midfielder's sprint capacity has dropped three percent over ten days will substitute him in the 70th minute. A club that does not know will substitute him in the 85th, and lose him for six weeks.

The difference between those two clubs is not budget. It is record-keeping discipline.

What bothers me most is not the files that have been cleaned up. Those, after all, can still be read backwards. What bothers me is the files with nothing to read, because they force me to talk about absence instead of about a human being.

An injured player is a story about a body, a career, a family, a contract, a season. An empty file is a story about paperwork. It has no characters. It only has a system.

That is why I always start from a specific person before going anywhere near a dataset. Aaron Hunt in the 34th minute. Mesut Özil at the third injection. 412 Bundesliga players across five seasons, each of whom woke up one morning and wondered whether today was the day his body would give up.

Data answers the question of pattern. People answer the question of meaning. If I could only do one, I would choose the second — but I will always keep the first beside me to check myself.

There is one thing I have never written down in nineteen years in this trade, and perhaps now is the time.

The first time I recognised the value of a blank space was not in a medical file. It was in a conversation with a club doctor in a stadium corridor, after he had signed the match report. I asked why he had not recorded the metric I had handed him. He did not answer my question. He said something else: "If I write it, I have to explain it. If I have to explain it, on Monday I will be sitting in a meeting with the sporting director. And I will not still be working here in June."

That is the entire mechanism, spoken in four sentences, in a corridor where nobody could hear.

After Germany were eliminated from the 2026 World Cup, I checked Özil's treatment log a fourth time. Three corticosteroid injections. Two sources independent of the national team doctor. One tracking analysis recording his contested possessions in the opposition half. None of it existed in any official document published for that tournament.

I wrote the piece. It did not say Özil played well. It did not say he was blameless either. It said only that the conclusion the public was handed rested on an incomplete set of facts, and that the missing facts could be restored if someone bothered to read the treatment log.

That was the only thing I had the right to say, and it was also the only thing that needed saying.

If there is one lesson I want to leave for young people entering this trade, it has nothing to do with sensors or predictive models.

The lesson is this: when you open a dossier and find it has nothing to read, do not jump to the conclusion that there is nothing to know. Write down that you opened it, what you were looking for, and that you did not find it. Record that absence with a date and the name of the provider. Then come back the following week, and the week after that.

If the blank is still there after you have asked several times, you have a story. Not a story about an injury. A story about a system that chose not to look.

And a system that chooses not to look always has a reason. The writer's job is to reconstruct that reason from documents, not from imagination.

The season goes on. There will be more mornings when someone goes down in the middle of the pitch at some minute or another, and more dossiers that get closed without anyone writing anything inside. I will still be in the same place, with my notebook and my tracking board, filling in both columns: the one with data, and the empty one stretching away to the right.

That empty column has never been the less important one.