F1 track limits rules, and why the white line decides

A driver crosses the line at the end of a qualifying lap, the timing screen shows a position that changes the front row, and ninety seconds later the number vanishes. The F1 track limits rules have done what they do most visibly: deleted a lap that looked, to almost everyone watching, entirely normal.

The rule underneath that deletion is far simpler than the arguments it generates. The complication is not in the definition. It is in the fact that the same definition has to be enforced at more than twenty circuits with different geometry, different run-off and different amounts of lap time available to a driver willing to ignore it.

The definition, and the part almost everyone gets wrong

The Sporting Regulations set it out at Article B1.8.6, and the wording is worth having exactly. The track is the surface between the white lines that mark its edges, the white lines themselves are part of the track, and the kerbs are not. A car has left the track when no part of it remains in contact with that bounded area. A driver who leaves it must rejoin safely and without gaining a lasting advantage, and the race director may allow an advantage to be given back rather than referring the case to the stewards.

Four separate provisions sit in that paragraph, and almost every argument about track limits is really a disagreement about one of them rather than about the boundary itself.

The clause that decides the most cases is the second one.

Because the line is part of the track, a car touching any portion of that line is still on the track. A car has left it only when nothing remains in contact with the bounded area, which resolves in practice to all four wheels being beyond the outer edge of the line. Three wheels over is not leaving the track. Two wheels over, the version that circulates most widely in casual discussion, is nowhere in the regulation at all.

The choice of an all-four-wheels threshold is not arbitrary generosity toward drivers. It is the only version of the rule that can be judged at two hundred miles an hour, from a fixed camera, thirty times a lap, across twenty cars, without argument. Ask whether a car is fully beyond a line and you get a yes or a no. Ask how far beyond the line it went, or whether the driver meant it, and you have introduced a judgement that has to be made identically by different officials at different corners for the answer to be fair.

That is the trade the sport has made. A bright-line test produces occasional results that feel harsh, and it produces them consistently. A discretionary test produces results that feel fair individually and cannot be reproduced.

• 4Wheels that must be beyond the line to have left the track
• 1Parts of the white line that count as track
• 3Warnings before a race-time penalty under the usual ladder
• 1Laps deleted for one qualifying breach

Structural features of the definition rather than figures from any particular event.

Kerbs, gravel and the surfaces outside the line

Everything outside the white line falls into one of a few categories, and knowing which one a corner has tells you almost everything about how that corner will be policed.

Kerbs sit immediately outside the line. Running across a kerb is not a breach in itself, because the wheels can be on the kerb while the car still touches the line. Kerbs exist to unsettle a car and to discourage the driver from going further, and the aggressive sawtooth versions are a deterrent expressed in suspension loads rather than in the regulations.

Asphalt run-off is the surface that created the modern problem. It was laid for safety, and it works: a car leaving the road at speed decelerates under control rather than digging into gravel and rolling. But asphalt is grippy, flat and fast, and at some corners the line through the run-off is quicker than the line through the corner. That is the entire origin of the track limits debate. The sport built safer circuits and, in doing so, removed the natural punishment that used to enforce the boundary for free.

Gravel and grass punish an excursion automatically. A driver who runs wide into gravel loses time, and frequently loses the car. Nobody needs to police a corner where the penalty administers itself, which is why gravel traps have been reinstated at the outside of several corners in recent years. It is cheaper and more reliable than a sensor and a steward.

The pattern to take from this: the amount of enforcement a corner needs is inversely proportional to how much the geometry punishes leaving it.

Why only some corners are monitored

A common complaint is that enforcement seems selective. It is selective, deliberately, and the logic is straightforward.

Race control publishes event notes before each weekend naming the corners at which track limits will be monitored. Those corners are chosen because they are the ones where leaving the track pays. A corner whose exit feeds a long straight rewards a wider line with a higher terminal speed carried all the way to the next braking zone, and the gain compounds. A slow corner in the middle of a twisty sector offers nothing, because the car has to be repositioned for the next turn regardless.

Monitoring every corner would be possible and would be worse. It would generate a stream of deletions for excursions that cost the driver time, consuming stewarding attention on cases where nothing was gained, and it would slow the reporting of the cases that matter. Enforcement is concentrated where the incentive is.

This is also why the monitored list changes between seasons at the same circuit. Resurfacing, a kerb profile change, a new run-off treatment or a different car generation can turn a corner that never paid into one that does.

1. 1The car crosses the boundaryAll four wheels go beyond the outer edge of the white line at a corner named in the event notes. Nothing has been decided yet; the crossing is the trigger, not the offence.
2. 2Detection reports itA loop sensor, a dedicated camera or a marshal post flags the excursion automatically or by report. The system records the car, the corner and the lap.
3. 3Race control reviews the caseThe question asked is whether the driver gained a lasting advantage, and whether the excursion was forced by another car or by avoiding an incident.
4. 4Qualifying: the lap is deletedThe time simply does not stand. There is no warning ladder and no accumulation, because withholding the lap already removes the entire gain.
5. 5Race: a warning is issuedThe team is told by message. Warnings accumulate across the race, and the driver is expected to correct the line rather than keep taking the time.
6. 6Repeat breaches reach the ladderOnce the warning count is exhausted the stewards apply a time penalty, and further breaches attract further penalties on the same basis.
7. 7The decision is publishedEvery deletion and penalty is issued as a stewards' document naming the car, the lap and the corner, which is what makes the enforcement auditable after the event.

The standing sequence. Individual corners and thresholds are set in the race director's event notes and vary by circuit.

Qualifying and the race are enforced differently, for a reason

The asymmetry confuses people, and it follows directly from what each session is measuring.

Qualifying is a series of individual lap times. If a lap was set partly off the track, the remedy is to not count the lap. That is complete: the driver gains nothing, loses nothing they were entitled to, and can go again if the session has time left. No warning system is required, because there is no residue to punish. This is why a driver can have several laps deleted in one session without any further sanction.

A race is one continuous event. Deleting something is not available, because there is no discrete unit to delete. A driver who runs wide at the exit of a quick corner has carried the advantage into the next straight and into the position they hold at the end of it. The remedy has to be applied to the race result instead, and it arrives through the warning ladder and then a time penalty added to the driver's race time or served at a stop.

There is a second reason the race is handled with warnings rather than immediate penalties. Track conditions evolve, cars go off while avoiding incidents, and a driver is sometimes forced wide by another car. A ladder gives room for the difference between a driver taking liberties repeatedly and one who ran out of road once.

The advantage test, which is the actual rule

Leaving the track is not by itself an offence. What the regulations care about is whether the driver gained a lasting advantage by doing so.

This distinction does a lot of work and is the reason several excursions each race produce no consequence at all.

A driver who runs deep into a corner, takes to the run-off and rejoins having lost half a second has gained nothing and is not penalised. A driver forced off by a rival alongside them has not chosen to leave the track. A driver who takes to the run-off to avoid a spinning car has done what the flags and the safety framework ask of them.

The obligation on rejoining is to give the advantage back if one was taken. That is why a car will sometimes visibly lift on the following straight or let a position go: the driver has been told, or has worked out, that keeping the gain converts an excursion into an offence. Handing it back promptly and unambiguously ends the matter.

Where the advantage is a position rather than lap time, the standard is stricter, because a position held to the end of the race cannot be unwound afterwards without changing the result. This overlaps with the wider set of in-race penalties and how each is served, which is where the sanctions themselves are set out.

How a breach is actually detected

Three mechanisms operate together, and which one a corner uses is a practical decision rather than a regulatory one.

Loop sensors buried in the surface beyond the line detect a car passing over them. They are objective, they work in any visibility, and they report instantly without anyone watching. Their limitation is that a loop reports a crossing at the point it is buried, so the geometry of the installation defines what it can catch.

Dedicated cameras are positioned to view the boundary at a specific corner. They give a reviewable image, which matters when a case is disputed, and modern installations pair the image with automated detection so a human is confirming a flagged event rather than watching for one.

Marshals report from post. This is the oldest method and it remains in use where a corner does not justify permanent instrumentation, or as a cross-check.

The reason the mix matters to a viewer is that it explains the timing. A sensor-policed corner produces a deletion within seconds. A corner relying on camera review and confirmation produces one minutes later, sometimes after the session has ended, which is why a qualifying result can change once the cars are already in parc fermé.

• Time gained by going wide
• Enforcement needed

Asphalt run-off onto a straight9/1010/10

Asphalt run-off into a slow corner4/105/10

Aggressive kerb then asphalt3/104/10

Gravel trap1/101/10

Grass and a barrier1/101/10

Relative, illustrative scale. The point is the inverse relationship between natural punishment and required policing, not the values.

Show the numbers

How much enforcement different corner treatments need

Item	Time gained by going wide	Enforcement needed
Asphalt run-off onto a straight	9/10	10/10
Asphalt run-off into a slow corner	4/10	5/10
Aggressive kerb then asphalt	3/10	4/10
Gravel trap	1/10	1/10
Grass and a barrier	1/10	1/10

Why the sport keeps changing the answer

Track limits have been policed by several different methods over the years, and the churn is not indecision. Each approach solves one problem and creates another.

Judging every case on whether an advantage was gained is the most philosophically correct method and the least reproducible. It requires a human decision per incident, it produces inconsistency between corners and sessions, and it generates precisely the accusation that some drivers are treated differently from others.

Policing the white line absolutely is reproducible and produces results that feel disproportionate: a lap deleted for a few centimetres at a corner where nothing was gained.

Building the punishment into the circuit removes the need to police at all, and costs money, requires the circuit owner's cooperation, and interacts with safety requirements that were the reason asphalt run-off was laid in the first place. A gravel trap that punishes a driver who runs wide also complicates the recovery of a car that has crashed there, and can pitch a car that arrives at a bad angle.

The current settlement leans on the third option where circuits will pay for it and the first two where they will not. That is why the same nominal rule produces visibly different racing at different venues, and why the argument recurs every season rather than being settled once.

The pit lane has its own boundary, and it is absolute

One line on a circuit is enforced without any advantage test at all: the one at the pit exit.

A car leaving the pit lane must keep to the correct side of the line marking the exit path until it ends. This is not judged on whether crossing it gained time. It is a safety provision, because a car rejoining at low speed alongside cars arriving at racing speed is one of the more dangerous configurations in the sport, and the line exists to keep the two apart until the merge is complete.

The same absoluteness applies to the pit lane speed limit and the other procedural requirements around a stop, which sit alongside the rules governing what happens during a pit stop itself. The distinction to hold onto is that safety boundaries are enforced on the crossing, while racing boundaries are enforced on the gain.

What a driver actually experiences

From inside the car this is not a rules question, it is a feedback problem.

A driver at the limit is placing the car within a few centimetres of a line they frequently cannot see, at a corner exit where the car is loaded and the view is of the horizon rather than the ground. The reference is a kerb, a marker board, a change in surface colour, sometimes a shadow. Modern cars give a warning through the wheel when a breach is recorded, so a driver knows the count is climbing, but the correction has to be made by feel on the next lap.

The consequence is that drivers deliberately build a small margin at monitored corners and spend it back at unmonitored ones. That is a rational response and it is visible in onboard footage: the same driver will run right to the paint at one corner and leave a car's width at another twenty seconds later.

It also explains why deletions cluster. A driver who has found a quicker line that happens to breach the boundary will keep producing the same lap until told, because the line feels right and the timing screen rewards it.

The corner-cutting case is the same rule pointed the other way

Almost everything above concerns a car running wide. The mirror image, a car cutting the inside of a corner, sits under the same provision and behaves differently in one respect worth understanding.

Cutting a corner is far rarer, because the inside of a turn is usually protected by a raised kerb, a bollard or a wall, and because the geometry means a cut line is frequently slower rather than quicker. Where it does happen, it tends to happen at chicanes, where straightlining the sequence is genuinely faster and the inside is open enough to permit it.

The regulation does not distinguish between wide and inside excursions. The boundary is the boundary, and the advantage test applies identically. What changes is the enforcement emphasis. A chicane cut usually produces a large, unambiguous gain rather than a marginal one, so it is treated less as a lap-limits matter and more as a case for handing the time or position back immediately. Circuits with a history of it install a physical deterrent, which is why so many chicanes acquired kerb blocks and bollards rather than being monitored electronically.

The other structural difference is that a cut is more often a consequence of a mistake under braking, a lock-up, a missed apex, than a deliberate line choice. That makes the forced-or-chosen question easier for the stewards, which is part of why these cases generate less argument than a repeated wide exit at a fast corner.

Why this argument will not be settled

It is tempting to treat the recurring track limits controversy as a failure of governance that better drafting would fix. It is not, and seeing why is the most useful thing to take from the subject.

The rule has to satisfy four demands at once. It must be objective enough to enforce identically across every car, corner and session. It must be cheap enough to apply thirty times a lap without a queue of decisions building up. It must not punish a driver who was forced off or who gained nothing. And it must not require circuits to rebuild run-off that exists for safety reasons in the first place.

Any rule that maximises one of those weakens another. Absolute white-line enforcement is objective and cheap and occasionally absurd. A pure advantage test is fair and unenforceable at scale. Physical deterrents settle the matter permanently and cost money the circuit may not have, and reintroduce recovery and safety problems the asphalt was laid to solve.

So the sport runs a compromise, adjusts which corners sit on which side of it each season, and absorbs a fresh round of complaint every time the adjustment is visible. That is not a symptom of the rule being badly written. It is what a genuine trade-off looks like when it is being managed in public.

Where track limits meet other rules

Two overlaps are worth separating, because they get conflated in commentary.

Track limits and the flag system are different things. Running wide under a yellow flag is two separate matters: the excursion, judged on advantage, and the obligation to reduce speed, judged on its own terms. The meaning of each flag and what it requires is a distinct body of rules, and a driver can breach one without the other.

Track limits and penalty points are connected but not equivalent. Deletions in qualifying carry no licence consequence. Repeated race breaches produce time penalties, and the stewards may attach penalty points where the conduct warrants it, but the ordinary case is a time penalty alone. The licence points system and how sanctions escalate covers where those points lead.

Why the rule is written as a property of the car, not the driver

A detail in the wording repays attention. The regulation asks whether any part of the car remains in contact with the track. It does not ask what the driver intended, how fast they were going, or whether the excursion looked deliberate.

That is a deliberate piece of drafting, and it has two consequences.

The first is that intent is irrelevant to whether the boundary was crossed. A driver who is pushed wide, one who makes a mistake, and one who has decided the run-off is quicker all cross the line in exactly the same way, and all three are detected identically. Intent only re-enters at the second stage, when the question becomes whether an advantage was gained and whether the driver was forced there. Separating those two questions is what allows detection to be automated at all: a sensor can answer the first and no sensor could ever answer the second.

The second is that the test is about the car's footprint rather than its trajectory. This sounds pedantic and is not. A car can be travelling entirely beyond the racing line, sliding, sideways, and still be within limits if one wheel remains on the paint. Equally a car perfectly under control and pointing straight ahead can be in breach. The rule measures where the machine is, not how it got there or how it looks.

That is also why onboard footage is so often unpersuasive. From inside the cockpit an excursion that breaches the rule can look identical to one that does not, because the difference is a few centimetres of rubber against a painted line that the camera cannot see and the driver cannot feel.

The parts of the paragraph people forget

Two of the four provisions in the regulation get almost no attention and both matter.

The obligation to rejoin safely is separate from the advantage question. A driver who runs off, gains nothing, and then rejoins across the racing line in front of a car at speed has committed a different offence entirely, and one the stewards treat far more seriously than a lap-time gain. Unsafe rejoining is a safety matter, and it is judged on the danger created rather than on whether any time was taken. A driver can be entirely blameless on track limits and penalised for how they came back.

The race director may allow the advantage to be given back. This is the mechanism behind those moments where a driver visibly lifts, or hands a position back, and nothing further happens. It is not an informal courtesy or a team decision made in a vacuum: the regulation explicitly contemplates the remedy being applied without the stewards being involved at all. It exists because referring every recoverable case to the stewards would clog the process with incidents that fixed themselves within a lap.

The practical result is a two-tier system. Recoverable advantages get handed back and disappear. Unrecoverable ones, and repeated breaches, go up the ladder. Understanding that division explains most of the apparent inconsistency in how similar-looking incidents end up being treated.

Kerbs are excluded, and that exclusion is doing work

The regulation's statement that kerbs do not form part of the track is easy to skim past and is the reason a great deal of apparently marginal running is legal.

Because the kerb is outside the track, a car with wheels on the kerb has wheels outside the track. But the test is whether any part of the car still touches the bounded area, and a car with two wheels on a kerb almost always still has two wheels on the asphalt or the paint. So kerb-riding, which looks like the most flagrant possible boundary abuse, is routinely and entirely within the rules.

This is why circuit designers reach for kerb profile as a tool. If a corner is being abused and the governing body would rather not police it with sensors, making the kerb more punishing changes the driver's calculation without changing a single word of the regulation. A sawtooth kerb that unsettles the car, or a raised block that risks the floor, imposes a cost measured in lap time and component wear rather than in stewards' documents.

It is enforcement by engineering rather than by adjudication, and where it works it is the cleanest solution available, because it produces no arguments and requires no officials.

What to watch for as a viewer

Track limits are one of the more readable parts of a race weekend once you know where to look.

• Read the event notes if you can find them. The monitored corners are published before the weekend, and knowing which three or four corners matter turns a vague sense of unfairness into a specific expectation.
• Watch the exit kerb, not the apex. Almost every deletion happens on exit, where the car is being fed power and running out of road, rather than at turn-in.
• Note how quickly a deletion appears. Seconds means a sensor. Minutes means review, and a result that may still change after the session.
• Watch for a driver lifting on a straight. That is usually an advantage being handed back, and it tells you a decision was made on the pit wall before the stewards had to make one.
• Count the warnings in a race. Broadcast graphics show the tally, and a driver on their last warning is a driver about to change their line at one specific corner.

The thing to carry away is that the rule is not really about the white line. The line is only the measuring instrument. What the regulations are trying to protect is the idea that the lap time a driver sets was produced on the circuit everyone else is using, and every enforcement argument in the sport is a disagreement about the cheapest reliable way to guarantee that. More on how the sport's rules fit together sits in the motorsport archive, and the wider set of explainers is collected on the blog.