F1 Injury Files: When a Report Is Too Clean, the Gap Is the Story

On 6 December 2026, the FIA medical bulletin issued after the Sakhir round ran to four lines. Romain Grosjean had pulled himself out of a burning Haas at Turn 3 after 28 seconds; the car's accident data recorder logged the first impact at 67 g. The bulletin mentioned burns to the back of both hands, one surgical procedure, and a single sentence saying the driver was “stable”. No acceleration data. No thermal exposure time. No neurological assessment, not even as a conclusion.

I read those four lines four times over one evening in Hamburg, and only on the fourth pass did I realise I was looking in the wrong place. A four-line bulletin is not short of information. It was designed to be exactly four lines.

Fifteen days later, Grosjean was back in the car at Abu Dhabi.

Formula 1 does not publish driver medical files. That is regulation, not habit: the FIA medical delegate holds the original record, the team receives only the conclusions it needs to run the car, and the rest sits behind a confidentiality layer tightened after every fatality. The confidentiality has a real professional rationale — a driver whose spinal file became public would lose a seat before losing form.

Which means every piece written about injury in F1, including this one, is built on an information structure with holes in it. Nobody drilled those holes. They are part of the design.

I learned to read those holes before I ever covered F1. In 2026, working as the team doctor liaison for Hamburger SV in the Bundesliga, I logged Aaron Hunt's GPS data against RB Leipzig: minute 34, speed falling from 7.2 m/s to 5.8 m/s after one acceleration phase, with the coaching staff leaving him on. I took those numbers to the men's dressing-room door and was stopped with one line: “Women don't understand tactics, get out.” I did not argue. I stood and waited for the team doctor to confirm the numbers.

From that day I have written to one rule: only what has a source. Data has no gender. Only the person reading the data carries bias.

In 2026 I built a comparison table of injury records for 412 Bundesliga players across five seasons. When football restarted after the shutdown, hamstring re-injury rates rose 19 percent. Injuries do not recur because players are weak. They recur because of the calendar.

Three F1 files.

Romain Grosjean, Sakhir 2026: a 67 g impact, 28 seconds in fire, burns to the hands, one race missed, a return at Abu Dhabi after 15 days. From a sports-medicine standpoint this is a fast comeback that is not unreasonable — second-degree burns on the back of the hand do not affect spinal function, and post-concussion assessment can be completed in seven to ten days absent residual symptoms. What stands out is not the recovery speed. It is that we know the 67 g figure from a car sensor, not from a medical file. Without that sensor on the Haas, the only remaining fact would be the word “stable”.

Robert Kubica, 2026 to 2026: a rally accident in February 2026, severe injury to the right hand, eight years before a full-time F1 race seat at Melbourne 2026 with Williams. This is the longest file and the most sealed one. We know he came back; we do not know wrist range of motion, baseline pain level, or grip-strength threshold. An F1 driver needs heavy steering torque through high-load corners, so a limitation in the right wrist becomes a limitation at specific corners. Nobody publishes which corners.

Daniel Ricciardo, 2026: a metacarpal fracture during practice at Zandvoort, five races missed, a return at Austin. This is the cleanest of the three files, and it shows most clearly the limits of how we read. We have the fracture date, the number of races missed, the return date. We do not have the decisive item: the load through the hand under late braking at Turn 1, and the pain threshold the driver accepted.

Three files, three levels of transparency, one common point: the information that actually determines performance is never published — not because it is hidden, but because nobody is obliged to publish it. An injury record does not lie; only the person reading it knows how to hide the truth.

That is why I do not trust a medical report before I understand the pressure bearing down on the doctor's signature. That signature does not merely confirm the state of a body. It confirms an agreement between four parties: the team doctor, the FIA medical delegate, the team, the driver. Four people, four different levels of risk, but only one of them pays with a spine — so the signature always leans toward the other three.

The harder thing to see inside the story of “the team rushed the driver back” is that the pressure does not come from a human decision. It comes from the race count. A 24-round calendar leaves no room for a six-week reconditioning block, at every team, for every driver. The driver is not being pushed by his team principal. He is being pushed by a calendar nobody in that meeting room wrote.

The common reading is that a short medical bulletin means concealment. I read it the other way.

A four-line bulletin is the only thing that allowed Grosjean to return after 15 days without carrying a label for the rest of his career. If the full file were published — 67 g, exposure time, post-smoke-inhalation markers, every neurological test result — the next team to hire him would read that before reading his results. Brevity protects the driver. It also conceals the second injury. Both are true at once, and the problem has no clean solution.

The blind spot is this: when a file is too clean, readers assume there is nothing to worry about. In medical records, cleanliness is inversely proportional to completeness. A complete report is always messy — unresolved markers, values outside threshold, a note saying “monitor further”. A report with no notes at all has been through editing: by communications, by lawyers, or by the doctor who knows exactly who will read it.

When the dressing-room door closes, I understand that tactics are not on the whiteboard. In F1 that door is the medical room, and what is not on the whiteboard is real recovery time.

What I want is not full disclosure. It is a minimum data set: injury date, mechanism, days lost, races missed, and one line on residual functional limitation. Five fields. Not enough to sell papers, enough that a journalist does not have to guess, and enough that a 25-year-old driver knows what he is signing.

A good file is not a clean file. It is a file with room for uncertainty.