Colapinto handed Sepang penalty for causing Baku F1 pile-up

At Turn 1, moments after the Safety Car pulled in, the front-left of the Alpine numbered 43 locked solid. Inside the cockpit, Franco Colapinto felt it before anyone in the grandstand could see it: a flat spot burned into the tyre surface, a short screech, and the car sliding straight to the inside. Pierre Gasly was directly ahead of him, same colour, same team. The first contact lasted less than a tenth of a second. I have sat through enough evenings in a Milan newsroom to know that the most memorable collisions of a season rarely begin with a loud noise. They begin with a small signal that only a few people notice — a wheel locking a fraction early, a steering input turned further than necessary, a gap shortened in the instant before the field is released. The Baku incident is an almost textbook example of that pattern. After reviewing the footage from several angles, what caught my attention was not the impact. What caught my attention was the timing. A driver decided to attack the inside at the first corner of a restart less than two seconds after the green light, with brakes still cold and tyre surfaces below their operating window. The stewards later called it an error of control, and they used exactly the three words I always look for in any post-race report: failing to make the corner in a controlled manner. The outcome: Colapinto received a ten-second penalty, but because he retired from the race, that penalty was converted into a different shape — a five-place grid drop at the next round at Sepang, under Article B1.9.6.(e) of the FIA 2026 F1 Regulations. In the same document, the stewards declared no further action against Arvid Lindblad, who had spun Liam Lawson's car. Two incidents, the same corner, the same restart, two opposite conclusions. That is the thread I want to unpick here. I remind myself of one line before every piece I write: every tracking number belongs on the operating table, not on the altar. In this case, the thing that needs dissecting is not a number at all. It is a moment in a race that telemetry was never designed to explain in full. CONTEXT: THE SEQUENCE BEFORE THE FIELD WAS RELEASED To understand Baku, the chain of causation has to be rebuilt in the exact order it occurred. Alexander Albon, the Williams driver, started everything with an incident that forced the race to be neutralised. What followed is what I call the grey zone of a Grand Prix: the field bunched into a single line, the gaps compressed to their minimum, the speed dropped abruptly, and then the Safety Car peeled off. In those final thirty seconds, there is a paradox that anyone who has worked in a pit lane understands: this is the moment when everything needs to be coolest and also the moment when everything is cooling down. Cold brakes. Cold tyres. At the same time, it is the moment when strategy becomes hottest — because every gap has been erased, every advantage accumulated over twenty laps has evaporated, and any driver who attacks successfully at Turn 1 can jump several positions. That is the trap. And in my more than forty years of watching races, it is the kind of trap that young drivers fall into more often than any other. One procedural detail needs to be clear: the collision occurred at the first corner after the race was restarted. This is fundamentally different from an incident on the opening lap. At the first lap of a race, a long-standing convention in Formula 1 holds that stewards will judge more leniently, for the simple reason that twenty cars all arrive at one corner at once and nobody genuinely has enough space to avoid everything perfectly. That convention was not extended to a Safety Car restart. The stewards said so bluntly in their report: although the incident happened at the first corner after the restart, the first-corner leniency was not applied, because no other driver contributed to the collision. That is a sentence worth reading several times, because it marks a change in standard. If you have followed Formula 1 long enough, you know that stewards rarely announce a change of standard. They simply apply it quietly and let the media argue for two weeks before forgetting. This time was different: they said it out loud, in the same report in which they explained why Lindblad was excused — as if determined that nobody could accuse them of inconsistency. THE MECHANICS OF A RESTART: WHY THE FRONT LOCKED An F1 car brakes by shifting the entire load forward. Under braking, weight transfers to the front axle, the front tyres press into the road, and friction rises to allow greater braking force. That is basic physics. The problem lies in the fact that friction does not increase linearly with temperature — it rises very slowly in the cold range, then spikes once the tyre reaches its working window, then collapses when the tyre is too hot. At ambient or near-ambient temperatures, an F1 tyre has a coefficient of grip so low it is hard for an ordinary person to believe. That means a driver must brake earlier, more gently, and for longer than on a normal lap. Yet in the final instant before the Safety Car dives in, competitive instinct says something else: attack, the gaps are gone. The conflict between those two signals — one from the hardware, one from instinct — is where the Baku incident was born. What the stewards described was a lock-up of the inside front, and they concluded Colapinto braked too late and entered the corner on the inside at a speed that did not allow him to make it in a controlled manner. Note the precision here: they did not say the braking system failed. They did not say there was a technical problem. They said the fault lay in the driving. This is an important distinction in the language of steward reports. When they want to exclude a technical cause, they use words such as error of control, failing to make the corner, braking too late. When technical doubt exists, they say further investigation is needed or leave the matter open. In this case, nothing was left open. In my experience, a front-axle lock-up at the first corner of a restart almost always carries the signature of three factors combined: cold brakes, tyres below the thermal window, and an attacking decision taken too early in the re-acceleration cycle. None of those three is abnormal. What is abnormal is all three meeting on the same metre of asphalt. There is one technical point I want stated clearly to avoid misunderstanding. When brakes are cold, the front-to-rear brake distribution set by the driver remains the same on paper, but its practical consequence changes completely, because the front axle cannot generate enough temperature to absorb the braking force being requested of it. Put another way, the same braking input can be entirely safe on lap ten and entirely wrong two seconds after a restart. That is why the restart phase is regarded as one of the highest-risk windows of any Grand Prix. And that is also why I believe the real story is not about Colapinto personally. TWO INCIDENTS, TWO RULINGS: THE CAUSATION TEST Within the same report there was a second incident. Arvid Lindblad, the Racing Bulls driver, spun his team-mate Liam Lawson's car. By ordinary reflex, one would put the pen down immediately and write that the young driver had made a mistake. The stewards did not do that. They explained that the collision was primarily the product of restricted space and the chain reaction caused by Colapinto's shunt, rather than an error wholly attributable to Lindblad. In other words, they applied a causation test: who was wholly or predominantly to blame for this incident? Place the two conclusions side by side and the logic is clear. For Colapinto, nobody else contributed to the collision — he caused it and bore it himself. For Lindblad, there was an obvious external cause — a car ahead created an unavoidable situation. The same corner, the same instant, but two different causal chains, and two different outcomes. This is the point I believe the media passed over too quickly. Coverage focused on Colapinto's penalty, on Lando Norris's call for a race ban, on McLaren losing a car. But the principle established in the Lindblad case may prove far more consequential. It establishes that a driver can make an error at Turn 1 of a restart and still escape sanction, provided that error was induced by an upstream event. This is a downstream-victim precedent. And any precedent built on the concept of an external cause carries an unanswered question: how large must an upstream event be before it excuses a downstream error? That question now sits with the people who will occupy the stewards' room at the next rounds, and they will have to answer it with a concrete decision, not an abstract principle. One further detail deserves recording: the stewards' decision to explain both rulings fully within the same document was a deliberate act of transparency. Anyone who works in this environment knows that issuing two contradictory decisions in one report opens the door to protest. The only way to close that door is to explain the differentiating variable before anyone asks. They did exactly that. ALPINE: THE SELF-INFLICTED DOUBLE DNF AND ITS PRICE Of all the consequences of the Baku incident, the heaviest is not the penalty handed to Colapinto. The heaviest is that Alpine lost both cars in a single incident, and one of those cars was eliminated by the team's own driver. In my own classification of loss types across a season, this is the worst category. It is not a car retiring through a power-unit failure — the kind of incident where you can point at the engine supplier and at least lose nothing internally. Nor is it a collision with a rival — the kind you can call a racing risk and move past. This was an incident in which one of the team's drivers eliminated another of the team's drivers, and the team lost both cars in a race where both could plausibly have scored. No strategy protects against this category of risk. No data predicts it. And no technical meeting at the factory can repair it, because the problem lies in a decision taken in less than a second at Turn 1. I have watched enough seasons to recognise a pattern in how teams handle these situations. In the first forty-eight hours, the team issues a unified message about team spirit and moving forward together. In the following week, there is an internal meeting whose contents nobody outside learns. Over the next two rounds, some adjustment appears — usually a clearer rule about which driver may attack which driver in specific circumstances. I would not be surprised if Alpine is walking that exact path now. What interests me is which specific rule emerges. A rule stating no attacking a team-mate at a restart sounds entirely reasonable on paper, but it violates a basic principle of racing: drivers must be free to compete. Any rule restricting that produces a driver stripped of instinct — and in Colapinto's specific case, a driver who needs to prove himself, removing the attacking instinct may do more harm than good. This is where I want to say something that may not please many people. A contract only looks good on paper until someone tries to fit it into a running system. The same logic applies to a team's internal rules. A rule written in a factory meeting can be logical and entirely sound. At the next race, applied to a driver being squeezed from behind at Turn 1 with cold brakes, it can become meaningless. What I genuinely want to know is whether the Alpine pit wall had any realistic window to intervene. At a green-flag restart, the interval between the race director's message and Turn 1 is far too short for an engineer to transmit any meaningful strategic instruction. Hold position and attack, at that stage, are the same thing — words spoken in a window during which the driver has already begun to act. Which means responsibility returns to the driver, and to whoever trained him. MCLAREN: EXOGENOUS RISK AND THE QUESTION OF POSITIONING In this incident McLaren made no mistake at all. Lando Norris did not decide to attack, did not decide to brake too late, did not lock a wheel. He was simply there, in a position many would call a good one — until the car ahead decided to do the wrong thing. This is the kind of risk that is hard for a front-running team to manage, because in theory it lies outside their control. But I do not think it lies entirely outside their control. In recent years I have begun noticing something teams rarely put in their strategy documents: positioning strategy during restarts. That is, not merely where your driver sits on the order, but who he sits beside. If you are racing at a circuit where you know young drivers with a history of aggressive restart attacks will be present, there is a strategic argument for being on the outside rather than the inside. There is an argument for holding slightly more distance than normal in the opening seconds. There is an argument for choosing an entry line with more escape space, even if it is marginally slower. All of that trades a little position for a little safety. In an ordinary race, that trade may not be worth it. But in a season where the gaps between the leading teams are measured in single points, losing a car to someone else's error at a restart is the kind of loss a team can absolutely reduce the probability of. I am not saying Norris was wrong to be there. I am saying there is an unexploited category of analysis in Formula 1 — the analysis of geographic risk during neutralisations. In a race where everything is optimised to the thousandth of a second, ignoring a variable that can cost an entire Grand Prix is a waste. From a training ground in Milan to an esports screen, the law of space remains the same. People always talk about filling space. Fewer talk about choosing which space to stand in. RACING BULLS: LINDBLAD AND THE SQUEEZED SPACE There is a detail in the report that I think will be cited repeatedly in the coming months: the way the stewards described Lindblad's incident. Restricted space. Chain reaction. No further action. For a rookie driver, this is a far more valuable outcome than it appears. Imagine the reverse: Lindblad penalised for spinning his team-mate at a restart. A penalty like that would follow him all season — not because it is heavy in sporting terms, but because it becomes part of the file every team reviews when assessing a young driver. Every time he enters a similar situation, commentators will recall Baku. Every time a contract rumour surfaces, it will be raised. The stewards took that possibility away. And they did so on the basis of a causation analysis, not on goodwill. This is a point where I think the stewards performed better than their average. For years I have criticised F1 steward decisions for inconsistency and for a tendency to sanction based on consequence rather than conduct. A collision causing greater consequences is not necessarily a worse act. In this case, they drew the distinction correctly. I still hold one doubt, however. If Colapinto's shunt had affected one car instead of two, would Lindblad have been excused? If the chain reaction had not been large enough to produce such an unmistakable scene of chaos, would the stewards have decided that he remained responsible for spinning his team-mate? I do not have the answer. But I know that question is being asked inside team briefing rooms. FROM TEN SECONDS TO FIVE PLACES: THE MECHANISM AND WHAT IT DOES NOT EXPLAIN Procedurally, the conversion is correct and clearly regulated. A driver who retires from a race cannot serve an in-race time penalty. If it cannot be served, the penalty is converted into another form, in this case a grid drop at the following round. The mechanism exists for a very specific reason: without it, a driver could cause a major collision through his own error, destroy his own race, and escape any sanction because he was no longer on track. Here, Article B1.9.6.(e) of the FIA 2026 F1 Regulations is the provision enabling conversion. The citation itself is a signal — it shows the 2026 rulebook has been restructured in its codification, and precedents from the previous rule cycle may no longer map directly onto the new one. There is, however, a comparative question the report does not resolve. A ten-second penalty becomes a five-place grid drop. Where does that ratio come from? No formula appears in the report. In a sport where everything else is measured to the thousandth of a second, a penalty converted according to an unpublished ratio is a detail worth tracking. More important is the structural asymmetry. A driver who causes a collision and finishes serves his penalty within the race, in a specific window, at a specific cost — he knows exactly how many track positions he has lost. A driver who causes a collision and retires serves his penalty at a different round, at a different time, at a different circuit, with an entirely different strategic consequence. In this specific case, that difference is not small. A five-place grid drop at a high-degradation circuit is a far heavier penalty than five places at a circuit where overtaking is comparatively easy. And if the Sepang reference is accurate — a tropical circuit with high tyre degradation and extreme track temperatures — this form of penalty may be considerably heavier than it looks on paper. This is the kind of detail nobody debates on social media, but it is exactly the kind of detail that decides a Grand Prix. THE FIRST-CORNER LENIENCY DOCTRINE REFUSED: A CHANGE OF STANDARD This is the part I regard as the most important in the entire report, and also the least noticed. A convention exists in Formula 1 that collisions on the opening lap are treated more leniently. The reason is practical: on lap one, twenty cars arrive at one corner together, nobody has enough information, nobody has enough space, and penalising every collision would produce a system in which any driver starting mid-field could be sanctioned for an incident he did not control. The stewards declined to extend that convention to a restart. And the reason they gave — that no other driver contributed — is a reasoned one. But think about what it implies over the longer term. If restarts are judged by a higher standard than race starts, it follows that drivers must treat a restart as a situation demanding more caution, not less, than the opening lap. Intuitively this is entirely correct — the field is compressed, brakes are cold, tyres are cold, and every strategic gap has been erased. But culturally, it runs against a powerful instinct: the restart is an opportunity. Every driver knows it. Every strategy engineer knows it. And precisely for that reason, restarts produce both the finest overtakes and the worst collisions. The stewards have now said they will scrutinise restarts more strictly. That is a change of standard. And if applied consistently, it will change how drivers approach one of the most important moments of a Grand Prix. Every collapse has a precondition; few people bother to look beforehand. In this case, the precondition was written publicly in a stewards' report. The question is how many drivers will read it. CONTRARIAN: A PENALTY NOT MEASURED IN SECONDS Let me now offer a view I know will not be widely shared. In the public debate about Baku, most attention focused on whether Colapinto should be banned. That is an emotionally compelling question but an analytically wrong one. Because the penalty he actually received — a five-place drop at Sepang — may harm Alpine more than a one-race ban would have. Run the numbers. A one-race ban removes a driver from one round and inserts a reserve. At Alpine, that means losing a driver but still having a car on track. The team can still score with the remaining car, and in some cases an experienced reserve delivers a result no worse. A five-place grid drop is different. It removes nobody. It simply starts a car from a worse position, at a circuit where overtaking is difficult, in high temperatures, with a strategy constrained because the driver must attack more to compensate, which accelerates tyre degradation, which forces an earlier pit stop, which traps him in traffic. In other words, this penalty does not eliminate Colapinto's opportunity. It makes that opportunity far more expensive. And there is another dimension I regard as more important than all of it: psychological pressure. In the round following a collision like this, every movement Colapinto makes will be dissected. Every braking input will be compared with Baku. If he enters a first corner cautiously, people will say he has lost his instinct. If he enters decisively, people will say he has learned nothing. That is a trap with no perfect escape. And it is why I always tell young people in this trade that sometimes the heaviest part of a sporting penalty is not its form, but the story it creates around it. Data only tells part of the story; the rest lies with those who know how to listen. And at Baku, what needed listening to was not the noise of the impact, but the silence of the seconds before it, when everyone knew what was coming. EXECUTION BLIND SPOT: WHEN A WHOLE TEAM IS SHAPED BY ONE DECISION There is an aspect of this incident nobody has discussed, and I believe it will become more important in the coming months. In the modern sport, teams have built elaborate risk-management systems. They have brake-temperature data, strategy simulations, algorithms predicting rival performance. They can calculate to a fine margin whether pitting on lap eighteen or lap nineteen is better. But none of those systems manages the instant when a driver decides to attack at Turn 1 of a restart. That is a structural blind spot. Teams optimise what can be measured, and what can be measured is what occurs over intervals long enough to carry statistical meaning. A decision taken in seven hundred milliseconds generates no statistically meaningful data. It generates only an outcome. Which means the only way a team manages this category of risk is through training — through building a driver's instincts, through repeating restart scenarios in simulation until the response becomes automatic. And here is the crux. That training only works if it runs against the driver's natural competitive instinct. A young driver at Alpine, fighting for his place, is handed an opportunity at Turn 1 of a restart. The opportunity is right there — his team-mate ahead, the gaps erased, and everything he has been taught about seizing opportunity says attack. To change that behaviour, a team must do more than write a rule. It must make the cautious instinct stronger than the attacking instinct, in a situation where every external factor says the opposite. That is a very difficult coaching problem. And I suspect very few teams address it seriously. NORRIS'S VOICE AND THE LIMITS OF EMOTION There is a detail here I think needs careful handling, because it concerns a driver I hold in high regard. Lando Norris, immediately after the collision, publicly called for a race ban for Colapinto. He made that call before reviewing the footage. I understand the emotion. I have watched many drivers climb out of a damaged car and say things they did not mean. I have seen multiple champions react in ways they later had to retract. This is a sport built on reflex, and reflex does not distinguish between driving and speaking. But there is a lesson here that has nothing to do with Norris. The stewards were not swayed by that call. They imposed a penalty far lighter than the public demand of a front-running driver. That matters, because it shows the decision-making process still rests on evidence, not noise. It also shows something about how this sport operates. When a driver at the front says publicly that another driver should be banned, that is not a comment. It is an intervention in a public debate. And if stewards yielded to such interventions, the system would collapse. I write this not to criticise Norris but to record that the gap between the public demand and the actual decision is an important indicator of the health of this sport's governance. SEPANG: WHERE THE PENALTY IS HEAVIER THAN IT LOOKS If the Sepang reference in the report is accurate, the analysis of this penalty becomes considerably more complex. Sepang is a circuit I remember clearly, not for its finest races but for its physical conditions. Track temperatures in Malaysia can reach levels at which tyre management becomes the only variable that matters. Under those conditions, starting from a poor position means a driver falls into a spiral: attacking to recover positions, overheating the tyres, pitting early, returning into traffic, attacking again. Very few drivers escape that spiral. Which is why a five-place drop at Sepang can be equivalent to a much heavier penalty at a circuit like Monza, where overtaking is easier and tyre management less brutal. The Sepang reference also carries a calendar implication. Malaysia has not hosted a Formula 1 round since 2026. Its appearance in a stewards' report for the 2026 season suggests the calendar has been expanded, and that the Asian regional weighting has been restored to some degree. This point needs independent verification before any commercial conclusion is drawn. I say that deliberately, because I have seen far too many analyses written on the basis of a single reference in a single document. REMAINING RISKS AND WHAT TO WATCH Taking it together, I would rate the overall risk level of this incident as medium. The penalty is decided, no regulatory appeal is pending, and the exonerated driver was publicly named with clear reasoning. Procedurally, everything is closed. But there are five signals I will be tracking in the coming rounds. First, the frequency of restart collisions. If the stewards have genuinely raised the standard, we will see its effect in how drivers approach Turn 1. If we see three or four more incidents of comparable severity in the next five rounds, the new standard will have changed nothing. Second, internal relations at Alpine. This is the kind of incident that may leave no public trace for months and then surface suddenly in a statement after another race. I will watch how the two drivers speak about each other in press conferences, and whether the pit wall issues differentiated restart instructions. Third, Colapinto's Super Licence penalty-point total. The report does not mention it, and that is a notable information gap. The twelve-point threshold triggers a one-race ban. If he is near it, the Sepang penalty is only the smaller part of a larger story. Fourth, how an upstream event is quantified in future cases. The principle established by the Lindblad ruling will be tested again at the next round, and how the stewards draw the line between an event large enough and one not large enough will determine whether this precedent has real value. Fifth, the composition of the 2026 calendar. If Sepang returns, the entire strategic calculus of the teams changes, and penalties like this one carry a different weight. A CLOSING THOUGHT I have spent more than forty years watching this sport, and what keeps bringing me back is never the victories. Victories are recorded, counted, ranked. They leave little to argue about. What brings me back are the moments before an incident happens, when everything could still go another way. A wheel locking slightly early. A steering input turned a little too far. A driver deciding to attack while every data point on his dashboard says wait. At Baku, that moment lasted under a second. Two Alpines retired. One McLaren retired. One driver received a five-place grid drop at a circuit where he will need every place he has. And in a room somewhere, a panel of stewards wrote a new principle for an old phase of the race. An empty grandstand does not kill a race, but it takes away something numbers cannot measure. At Baku, the opposite also held: the silence before the field was released took away the space a driver needed to correct his mistake. The question I leave for Sepang is not how many cars Colapinto can pass. The question is whether a driver, arriving at Turn 1 with cold brakes and an opportunity within reach, can learn to choose a different piece of space. If he learns it, we will know within three rounds. If he does not, we will know that too — and this time there will be another driver ahead of him.