Madring: 19 Red Flags, 11 Crashes, and the Silence Nobody Timed

The Car on Fire and the Silence Before the Extinguishers Arrived

Tasanapol Inthraphuvasak's car came to a stop at the exit of a high-speed corner at Madring, and then the fire began to crawl out of the rear. On the timing screens, the F2 driver's name turned red. The session was halted. He got out, walked clear of the danger zone on his own.

The part I keep re-reading is further down: it was some time before marshals reached the scene with fire extinguishers in hand. Nobody stated how long that interval lasted. Nobody rebuilt it as a timeline. For anyone who works with data, an unmeasured interval is the most dangerous kind — it cannot be challenged, cannot be improved, and will never appear in any safety report.

I wrote the first line into my notebook: "Madring, test day — car fire, driver unhurt, fire response delayed." Then a second line: "19 red flags across two days, 11 due to crashes." Placed side by side, those two lines tell a story the original wire report only grazed. One is an individual accident attributable to a driver. The other is a systemic pattern, and the systemic pattern is what interests me.

What Madring Is, and Why It Has Not Earned Forgiveness Yet

Madring is the newest circuit in the top category, designed with 22 turns, a banked corner, and a hybrid street-circuit / purpose-built layout. Twenty-two turns places it among the densest venues on the calendar. The hybrid structure inherits the weaknesses of both families: the average speed of a purpose-built track, with the error margin of a street circuit — walls close, exits narrow, run-off minimal.

In the paddock, people compare Madring to Saudi Arabia's Jeddah. The comparison is not arbitrary. Jeddah is also high-speed with walls close, and it debuted under immediate driver questioning of its safety. But there is a difference rarely mentioned: Jeddah launched in a pandemic season, with a compressed calendar, and political pressure on the FIA was heavy enough that those questions were pushed into silence. Madring launches in a more stable cycle, with a media apparatus that has learned how to exploit risk.

That is why I do not read Madring as a circuit. I read it as a test of governance capability.

The FIA's licensing framework names its highest tier Grade 1 — a prerequisite to host a round. Grade 1 means the track passed inspection for barriers, run-off, medical systems, and response protocols. But inspection is a static state. It proves the circuit is adequate when nobody hits a wall at 250 km/h. It proves nothing when somebody does.

Every tactical diagram begins with a trembling hand-drawn line on PowerPoint. And my first trembling line for Madring had three marks: 22 turns, one banked corner, and an unbroken sequence of walls.

The Geometry of 22 Turns

Back in Vietnam, I learned to measure the space between two defensive lines by radius and opening angle. Here, I apply the same measurement to a circuit: corner radius, braking point, exit angle. The principle never changes. Space is never empty. It is only waiting for the right reader.

For Madring, the measurement produced three troubling parameters.

First, the banked corner. Banking lets a driver carry more speed through the apex, but it also generates sustained lateral load. Sustained lateral load means the tyre's tolerance budget is drained continuously, and any small error at entry is amplified across the whole sequence beyond it. A banked corner is an amplifier. It amplifies skill and error alike.

Second, corner density. On average, a direction change arrives every 250 metres. At that density, the driver gets no interlude to reset focus. Errors do not accumulate per lap; they accumulate per corner. And when errors accumulate per corner, the probability of a spoiled lap rises arithmetically.

Third, the walls. This is what I call a circuit's tolerance budget. At Barcelona you get several metres of grass and gravel to correct a mistake. At Madring you get a wall. The smaller the tolerance budget, the larger the consequence of the same error. That is why drivers describe Madring in the language of risk rather than the language of challenge.

Max Verstappen and George Russell both framed the circuit in terms of high-speed corners, close walls, and risk. Russell went further, calling it arguably the riskiest circuit he has driven. I do not read that as sentiment. A Formula 1 driver discusses circuit risk against a very narrow reference frame — dozens of tracks they have already driven. When someone with a reference frame that wide puts Madring at the top, that is data.

But here is my self-correction: driver perception is not a measurement. It is a signal. A signal must be cross-checked against the red-flag count. And the red-flag count exists.

Nineteen Red Flags and the Denominator Problem

Nineteen red flags across two test days. Eleven due to crashes. The rest for other reasons.

I place that figure against other test days in my memory. At Barcelona, at Silverstone, at Bahrain, three or four red flags in a day already felt heavy. Nineteen across two days is a different pattern in kind, not merely in volume.

But I have to be careful with myself. This is a classic trap: take an anomalous phenomenon and conclude something about the nature of the system before excluding the confounding variables. There are at least four I must subtract first.

The first is the track surface. A new circuit has asphalt that has not been rubbered in — no accumulated tyre rubber to raise grip. New asphalt is slicker than familiar asphalt. This is a consequence of a debut, not of design.

The second is car configuration. A test day is when teams run unvalidated configurations, sometimes deliberately pushing outside the operating window to gather data. An unbalanced car is normal in testing, and an unbalanced car crashes more.

The third is the driver pool. There are rookies, drivers moving up a category, drivers hunting the limit of a track nobody knows. The limit of a new circuit is usually found by exceeding it.

The fourth is running volume. On a test day, laps per hour run far higher than in an official session, because there are no segments and no time pressure. More laps means more opportunities to crash, even if the per-lap crash probability is unchanged.

After subtracting those four, what remains is still notable. Eleven crashes in two days. For a Grade 1 circuit that has passed inspection, that is higher than I would expect at a new venue. But it is not enough to claim the design is flawed. It is only enough to claim the design is being tested at its upper threshold.

The Fire Silence: The Part That Cannot Be Blamed on Asphalt

This is the section I want to isolate, because it is different in nature.

Slick asphalt is a consequence of a debut. Car configuration is a team's choice. Rookie drivers are a choice of the development system. All three can be argued, explained, and will fade as the circuit ages.

Fire response cannot.

When a car is on fire, a clock starts. There is no confounding variable in that equation. No asphalt, no car configuration, no rookie. There is one question: how long until suppressant reaches the fire.

The report states it was some time before marshals arrived with extinguishers. That is a description without a unit. And a description without a unit, in the domain of safety, is a data gap.

I have no figures to conclude the protocol failed. I have a single observation, and a sample of one cannot support a conclusion. But I have the right to ask a structural question: if a new circuit is granted Grade 1, is its response system tested under real load, or only under simulation? If the answer is simulation, then the silence we just saw may not be an isolated incident. It may be a product of process.

The summer of 2026 taught me that empty space is never empty; it is only waiting for the right reader. The gap between the sound of impact and the sound of an extinguisher is the same. It is not empty. It is full of questions nobody has asked.

The F1, F2, F3 Hierarchy and the Role of Canaries

One detail is easy to miss in the report: alongside the F2 fire, an F3 session was also halted for a crash.

In a race weekend's structure, feeder series run before the main event. In media terms they are the opener. In safety-data terms they are canaries in the coal mine.

The reason is simple. F3 drivers race the same circuit at relatively high speed with less experience, and in cars without the same passive safety as a Formula 1 machine. If a track has a narrow tolerance budget, the canaries fall first.

That raises a question the system rarely asks itself: when a new circuit enters the calendar, which reference frame evaluates it — the main series or the feeder series? If the main series, then F2 and F3 crashes are treated as the feeder series' problem. If the feeder series, they are evidence about the circuit's overall risk threshold.

I lean toward the second reading. Not because I want to indict Madring, but because a circuit is a physical system. Physical systems do not distinguish categories. The same banked corner, the same wall, the same tolerance budget. Only approach speed and driver tolerance differ.

The Political Economy of a New Circuit

A circuit does not exist in a technical vacuum. It exists inside a contract.

To place a new round on the calendar, a local promoter must sign a multi-year deal, commit to a hosting fee, and commit to infrastructure. In return it receives the right to host, ticket revenue, and media value. In that structure, safety is a variable with a cost. Every extra metre of barrier, every metre of widened run-off, every additional response station is money.

That does not mean promoters cut safety corners. It means a negotiation zone exists between technical requirement and budget, and that zone is rarely made public.

This is where I audit my own argument. I have no evidence Madring cut safety corners. I have two public data points: a delayed fire response and a high crash rate across two test days. Those two are not enough to conclude a root cause. But they are enough to demand an answer with a unit attached.

And that, I believe, is what drivers are waiting for too. The report notes drivers held back from voicing safety fears. I read that two ways. First: they want more data before speaking, which is professionalism. Second: an unspoken paddock agreement that nobody challenges a new circuit before it runs its first official race, which is politics.

The two readings do not exclude each other. And both send the same signal: pressure is accumulating, unvented.

The Counterintuitive Angle: The Circuit Is Not the Most Dangerous Thing Here

I want to push back against my own analysis, because that is how I protect myself from the storytelling trap.

The easiest story about Madring is: a new circuit is too dangerous, 19 red flags prove it, drivers say so, end of story. That is a tidy story. It has characters, conflict, evidence. And it may be wrong.

Because 19 red flags on a test day belongs to a context that does not repeat. Asphalt not rubbered in, configurations unlocked, corners not memorised, and most importantly — a test day has no price. A crash in testing costs repair money. A crash in a race costs points. When the price differs, behaviour differs. Teams push beyond limits in testing far more than in a race. So a testing crash rate cannot be used as a direct forecast for the race.

If I read Madring only through the number eleven crashes, I am comparing two states that are not comparable.

But here is the crux: fire response does not depend on price. It does not depend on whether the session is testing or official. It does not depend on whether the asphalt is rubbered in. It depends only on whether a response system works as designed.

In other words, I can reject the concern about circuit design and still keep the concern about operations. And if forced to track only one through the race weekend, I choose operations. Design only hurts you when you make a mistake; operations hurt you after you have already made it and are sitting in the car.

A misplaced pass is not a mistake. It is data the system is trying to send you. The fire at Madring sent me a very clear message, and it was not a message about 22 turns.

Transition: The Silence Between Incident and Process

Transition is not a stretch of running. It is the silence between two intentions that few can read.

In motorsport, I define transition as the interval between the end of an incident and the moment a process takes it over. Between impact and the red flag. Between flame and extinguisher. Between a driver's sense of risk and a governing body's action.

All three intervals exist at Madring, and none has been measured. I am not writing this to say the circuit is dangerous. I am writing this to point out three places where data is missing, and all three sit exactly where a process change could flip the entire picture.

If fire response had been fast, the Madring story changes key in a single note. A difficult circuit, a chaotic test day, a driver who escaped, and a system you can trust. That is an entirely different story from the same dataset.

What to Watch During the Race Weekend

When the race weekend begins, I will track three signals, and I recommend readers track them the same way.

The first is the red-flag count in official sessions. If it drops to a normal level for a circuit, the test-day pattern was largely a format artefact. If it holds or rises, that is data about design and must be treated as data about design.

The second is response time in any subsequent incident, however minor. An improvement here will not be announced as a victory. It is merely evidence the problem was recognised. Silence is also a signal.

The third is driver speech. If concerns continue to be withheld, the paddock has not reached the action threshold. If two or more drivers speak up systematically, the governance structure must reopen the file.

I redrew Madring on PowerPoint once more after finishing this piece. Still a trembling line. But this time I added three question marks at three different positions on the map, and I recognised something familiar: the points we mark with question marks are rarely where the race is decided. They are where the race is organised correctly or incorrectly.

When there was no football, I drew football. It turned out drawing is also a way of understanding. I still draw. The only difference is that instead of drawing the space between two defensive lines, I now draw the silence between a burning car and an extinguisher that arrived late.

That silence is not on the blueprint. It is on the process. And process can always be fixed — if someone is willing to count it as a number.