F1 grid penalties explained: how a penalised grid is built

A driver sets the fastest time of a qualifying hour, climbs out to the standard congratulations, and starts the race eleventh. Nothing about the session is being disputed. The times stand, the pole position stands in the record, and the grid simply has nothing to do with either.

That gap between what a session produced and what lines up on Sunday is the whole of F1 grid penalties explained in one sentence: qualifying makes a classification, and the grid is a separate object built from it by a procedure written down in the sporting regulations. The penalties never touch the classification. They act only on the second object, and they act in a defined order that produces results most viewers find counter-intuitive, including the familiar case of a ten-place penalty that costs a driver five.

What follows is that procedure, step by step, along with where the penalties come from, what the fifteen-place threshold does, how several penalised drivers are sorted against one another, and why a team sometimes chooses to take a large penalty on purpose at one circuit rather than a smaller one somewhere else.

• 15Unserved penalty places before a driver is sent to the back
• 12Months over which unserved penalties are counted
• 5Reprimands in a season that convert into a grid penalty
• 10Places that fifth reprimand costs

Values set out in Articles B1.9.5, B2.5.4 and B8.2 of the current FIA Formula 1 Sporting Regulations.

A penalty is applied to the grid, never to the session result

The rulebook keeps these two things apart with some care, and the separation is worth holding on to because it explains several arguments that recur every season.

A qualifying session produces a classification: the order of drivers by their best times, subject to the deletion rules and the stewards' decisions about laps set in that session. That classification is a result. It goes into the record, it settles pole position, and no grid penalty alters a line of it.

The grid is then formed from that classification by the procedure in Article B2.5.4, and it is at this point, and only at this point, that penalties are applied. If qualifying never happens at all, the regulations fall back to the drivers' championship order and run the same procedure over that instead. If neither route works, the stewards form the grid at their own discretion, which is the rulebook admitting that some situations cannot be legislated for in advance.

The practical effect is that a driver can hold pole and start from the back on the same afternoon without any contradiction. It also means the sprint and the race are handled by two parallel procedures with two separate penalty pools, which is covered further down.

Where the penalties come from in the first place

Four sources produce most of what a driver carries into a grid.

The largest is component usage. Article B8.2 sets, for each driver across a championship, how many of each power unit element may be used, and Article B8.2.8 sets the tariff for going beyond it. The first time an extra element of a given type is introduced, the driver takes ten places. Every subsequent extra element of that type is worth five. Those penalties are cumulative, so a driver taking a new engine, a new turbocharger and a new energy store at the same event adds three separate drops together. The detail of what counts as an element and how the allowances are structured is a subject of its own, worked through in the F1 gearbox and engine penalty allowances, and the hardware itself is described in how an F1 power unit is put together.

The second is stewards' decisions arising from a lap-time session. Under Article B1.9.4, where an incident happens in practice or qualifying the stewards may delete lap times or drop the driver as many grid positions as they think fit. Those drops are served in the race unless the incident occurred in sprint qualifying, in which case they land on the sprint grid.

The third is stewards' decisions arising from the race or sprint itself. Among the penalties available in Article B1.9.5 is a drop of any number of grid positions for the next race or sprint the driver takes part in during the following twelve months. This is the mechanism that carries a Sunday afternoon's offence into a different weekend, and it is why a driver can arrive at an event already carrying a penalty before turning a wheel.

The fourth is the reprimand ladder. A driver who collects five reprimands in a championship, at least four of them for driving infringements, receives a ten-place penalty on the fifth. If that fifth reprimand comes out of a race, the ten places apply to the next event rather than the one just finished. Reprimands accumulate quietly across a season and this is the point at which they stop being cosmetic.

Two further sources are worth noting because they surprise people. An unsafe release during qualifying is punished with a grid drop of whatever size the stewards consider appropriate, rather than with the stop-and-go it would earn in a race, which is one of the ways the pit lane regulations feed into a Sunday grid. And a five or ten-second penalty, a drive-through or a stop-and-go that a driver was unable to serve, because the race ended or the car retired, may be converted by the stewards into a grid penalty at the driver's next race.

Unserved, and the twelve-month clock

Two words in the procedure do more work than their length suggests.

The grid is built using each driver's cumulative unserved grid penalties imposed in the previous twelve months. A penalty that has been discharged, because the driver has already started a race from the position it produced, drops out of the count. A penalty that has not been discharged stays in it.

The twelve-month window matters because grid penalties are not confined to the event at which they were imposed. A penalty handed down for an incident late in one season can be sitting on a driver at the first event of the next, and the rolling window is what stops such a penalty being kept alive indefinitely. It also means the count is a live figure rather than a season total: a driver's number goes up when a penalty is imposed and goes down when it is served.

This is also the reason the phrase used on television, that a driver is carrying so many places of penalty, is more literal than it sounds. The number belongs to the driver, not to the weekend.

The arithmetic that produces a temporary position

Here is the procedure itself, in the order the regulations set out, because the order is what generates the strange outcomes.

The grid starts nominally empty. No driver holds a slot, including the pole sitter.

The first drivers placed are the penalised ones, provided their unserved total is fifteen or fewer. Each of them receives a provisional slot found by adding their qualifying place to the sum of their penalties. A driver classified fourth with a five-place drop lands provisionally on nine. A driver classified second carrying fifteen lands provisionally on seventeen.

Only after every penalised driver has a provisional slot are the unpenalised drivers dealt with. They are placed into whatever positions remain empty, taken in the order of their qualifying classification. They keep their order relative to one another, always, whatever the penalised drivers have done around them.

The final step is the one people forget. Penalised drivers holding provisional slots are then moved forward to fill any position still unoccupied. The grid is closed up so that it has no gaps in it.

1. 1The team decides to fit fresh hardwareAn element is introduced beyond the allowance in Article B8.2, usually because reliability or performance across the remaining events is worth more than one weekend's grid position.
2. 2The element counts as usedThe regulations tie the moment to the timing transponder: an element has been used once the car has left the pit lane with it fitted. Running it in practice is enough.
3. 3The tariff is appliedTen places for the first additional element of that type in the championship, five for each one after it. Separate types are counted separately and the totals are added together.
4. 4The penalty joins the driver's unserved totalIt sits alongside anything else imposed in the previous twelve months and not yet discharged. That running total is what the grid procedure reads, not the individual decisions.
5. 5Qualifying runs unaffectedThe driver's laps count, the classification is set on merit, and pole position is awarded on time alone. Nothing about the penalty touches the session result.
6. 6The grid is assembledPenalised drivers take provisional slots equal to their qualifying place plus their total, unpenalised drivers fill the gaps in qualifying order, then the penalised drivers are pulled forward into whatever is still empty.
7. 7The provisional grid is publishedIt appears no later than two hours before the formation lap. Any car withdrawn after that leaves its slot vacant rather than closing the grid up again.
8. 8The penalty is dischargedOnce the driver has started from the resulting position, that penalty is served and stops counting towards the fifteen-place threshold at the next event.

The sequence a component penalty passes through under the current sporting regulations. A stewards' penalty enters the same path at the notification stage.

Two penalised drivers who land on the same slot

Ties are common, because a field of twenty produces plenty of ways for two drivers to arrive at the same number. The regulations break them with qualifying pace rather than with the size of the penalty or the order the penalties were imposed.

Where two or more penalised drivers share a provisional slot, the slower qualifier keeps it, and the faster ones are given the positions immediately in front. A driver classified sixth with a three-place drop and a driver classified seventh with a two-place drop both arrive at nine; the seventh-place qualifier stays on nine and the sixth-place qualifier takes eight.

That is a deliberate choice, and it is the fairer of the two obvious options. Ordering by the time a penalty was notified would reward a team for getting its paperwork in early, which is not a form of merit anyone wants to reward. Ordering by qualifying pace at least uses something the drivers earned on the circuit.

Why ten places so often costs less than ten

Take the closing-up step seriously and the familiar complaint answers itself.

Suppose, on an invented grid with round numbers, that the drivers classified third, fourth and fifth each carry a ten-place drop and nobody else carries anything. Their provisional slots are thirteen, fourteen and fifteen. The unpenalised drivers are then poured into the empty positions in qualifying order, which means the driver classified sixth takes slot three, the driver classified seventh takes four, and so on down. When the field has been placed, the highest empty positions are well above thirteen, so the three penalised drivers are pulled forward into them.

Each of them has lost fewer than ten places. Nobody has been let off anything and no discretion has been exercised. The grid has twenty slots, three of them were vacated by drivers who moved down, and the rule requires the grid to be continuous.

The same effect appears whenever a penalty would push a driver past the end of the field. A driver classified eighteenth in a twenty-car field with a ten-place drop cannot start twenty-eighth. They start last among the drivers placed by the arithmetic, and the effective cost is two places rather than ten.

This is the single most misreported thing about grid penalties, and the fix in a reader's head is to stop thinking of the number as a distance travelled downward. It is an input to a sorting procedure, and its effect depends entirely on what everybody else is carrying.

• Places on the penalty
• Places actually lost

Only driver penalised, qualified 3rd1010

Three drivers penalised together, qualified 3rd107

Driver qualified 18th of 20102

Driver qualified 2nd, six others penalised104

A constructed twenty-car example with invented qualifying positions, used to show the effect of the closing-up step. Not drawn from any real session.

Show the numbers

Illustrative grid: what a ten-place penalty actually costs

Item	Places on the penalty	Places actually lost
Only driver penalised, qualified 3rd	10	10
Three drivers penalised together, qualified 3rd	10	7
Driver qualified 18th of 20	10	2
Driver qualified 2nd, six others penalised	10	4

A worked grid, built from scratch

Numbers make the ordering obvious in a way prose does not, so here is a constructed field of ten with invented penalties. Nothing below describes a real session.

Say qualifying finishes in the order A, B, C, D, E, F, G, H, I, J. Driver B is carrying five places. Driver D is carrying five places. Driver E is carrying twenty. Nobody else has anything.

Step one places the penalised drivers who are under the threshold. B qualified second and carries five, so B takes a provisional seventh. D qualified fourth and carries five, so D takes a provisional ninth. E is over fifteen and is set aside entirely for now.

Step two places the unpenalised drivers into the empty slots in qualifying order. The empty slots at this point are one, two, three, four, five, six, eight and ten. A is the fastest unpenalised driver and takes one. C takes two. F takes three. G takes four. H takes five. I takes six. J takes eight.

Step three closes the grid up. B is sitting on seven and D is sitting on nine, and the highest slot still empty after the unpenalised drivers have been placed is nine, with ten free behind it. B is pulled up to nine and D follows into ten.

Step four handles E, who started this weekend with more than fifteen unserved places. E goes behind every other classified driver, so E starts eleventh in a field that has ten placed cars, which in a real field of twenty means last.

Look at what happened to B. The penalty said five places. B qualified second and started ninth, which is a loss of seven, not five. That is not an error; it is the consequence of D also being penalised and of E being removed from the middle of the order entirely. The closing-up step does not always help the penalised driver, and this is the case where it does the opposite: the drivers who moved down created gaps that unpenalised cars filled from below, and B was pulled into a slot that was lower than the provisional one.

The general lesson is that no driver's penalty can be evaluated on its own. The grid is produced by a single pass over the whole field, and every driver's outcome depends on what everyone else is carrying. A team can compute its own likely starting position on Saturday evening only because it knows the whole field's totals, which are published, and not because the arithmetic is local.

First extra element10places

Second extra element15places

Third extra element20places

Fourth extra element25places

Derived from the tariff in Article B8.2.8, which is ten places for the first additional element of a type and five for each subsequent one. Separate element types are counted and added separately.

Show the numbers

The cumulative cost of extra power unit elements of one type

Item	Cumulative penalty
First extra element	10places
Second extra element	15places
Third extra element	20places
Fourth extra element	25places

What a grid penalty leaves untouched

It is easier to reason about penalties once you know what they do not reach.

They do not affect points already scored. A drop applied at one event has no retrospective effect on any result, and a penalty carried across from a previous race does not remove anything the driver banked at it.

They do not affect the qualifying record. Pole position, session results, and any lap record set in the session survive intact, which is why the phrase "started from pole" and "qualified on pole" are not always the same statement.

They do not attach to the car or the team in the constructors' standings directly. A grid penalty follows the driver, which matters when a team fields a replacement: the regulations treat a substitute as the original driver for the purposes of power unit usage, so the allowance and the penalties that flow from it stay with the seat rather than with the person.

And they do not stack with in-race sanctions. A driver who takes a ten-place grid penalty and then earns a five-second penalty during the race serves both, but the two live in different systems and neither converts into the other unless the stewards make that conversion explicitly, which they may do when a driver could not serve an in-race penalty at all.

The fifteen-place threshold, and what happens beyond it

The arithmetic above applies only to drivers whose unserved total is fifteen or fewer. Beyond that number, the procedure stops calculating and starts relegating.

A classified driver who has accrued more than fifteen cumulative unserved grid penalties over the previous twelve months starts behind every other classified driver. There is no provisional slot, no closing-up, no interaction with anybody else's penalty. They are simply placed after the field.

That threshold exists because of the closing-up step. Without a cliff, a team facing an expensive component weekend could keep stacking drops onto one driver, knowing that the grid would close up and limit the damage. The sixteenth place of penalty would cost nothing. The cliff removes that incentive by making a very large total qualitatively different from a large one rather than merely bigger.

Where several drivers exceed the threshold at the same event, the group is ordered among themselves by qualifying classification. The faster qualifier in the relegated group starts ahead of the slower one. This is the change that most rewards a driver for taking qualifying seriously on a weekend where the result appears irrelevant, and it is why teams still send heavily penalised cars out for a proper lap in Q1 rather than treating the session as a formality. That interaction with the session itself is followed further in how F1 qualifying is structured.

Behind the relegated group sit any unclassified drivers who have been permitted to start by the stewards, ordered by the classification rules that would otherwise apply to them.

One consequence of the cliff is easy to miss. A driver whose total sits at exactly fifteen is still handled by the arithmetic and therefore still benefits from the closing-up step, while a driver at sixteen is not handled by it at all. The gap between those two situations can be worth several positions on a grid where a number of cars are penalised, and it is one of the few places in the rulebook where a single unit changes the category of the outcome rather than its size. Teams building a component plan for the back half of a season are aware of it, which is why penalties tend to be taken in blocks that either stay comfortably under the line or go well past it, rather than landing just above.

Back of the grid is not the same as a pit lane start

These two outcomes get used interchangeably in commentary and they are different things with different causes.

A back-of-grid driver still occupies a grid slot, still takes the formation lap with everybody else, and is still bound by the parc fermé condition of the car, meaning the car must be in the specification it qualified in.

A pit lane start is a separate consequence and usually a chosen one. A team that wants to change the car in ways parc fermé forbids, a different setup, different aerodynamic parts, a rebuilt car after a heavy qualifying accident, accepts a pit lane start as the price. The car then lines up in the fast lane, and if more than one car is starting from there they queue in the order the grid procedure produced for them. Work in the fast lane is permitted only until fifteen seconds before the formation lap, and the list of what may be done is short.

The practical difference is worth about the length of the pit lane in time plus whatever the field does on the first lap, and it is bought with the freedom to run a setup the team could not otherwise use. A team that has already lost the grid to penalties frequently takes the pit lane start as well, on the reasonable ground that the marginal cost is small once the position is gone.

The gearbox rule that no longer exists

For several seasons the gearbox was one of the components that carried a grid penalty. A driver had to keep a gearbox for a defined run of consecutive events, and an unscheduled change cost five places. It produced a familiar weekend ritual of teams weighing a suspect gearbox against a known penalty, and it was one of the standard reasons a driver's Saturday result and Sunday grid diverged.

That limitation is not in the current sporting regulations. The component allowances that generate grid penalties are the power unit elements set out in Article B8.2, and the transmission does not appear among them at all. A team that wants to change a gearbox between sessions still has to work within parc fermé, which is a real constraint, but the automatic five-place tariff is gone.

The reason for removing it was the same reason a lot of reliability-based penalties have been trimmed. A penalty that punishes a mechanical failure the driver had no part in makes the championship a partial function of component lottery, and the sport has slowly moved those decisions from automatic tariffs towards cost control instead. The power unit allowance survives because it is doing a different job: it is rationing development and usage rather than punishing breakage.

Taking a penalty on purpose, and choosing where

The most interesting decision in this area is not about the rules at all. It is about which weekend to spend.

A team that knows it will exceed its power unit allowance at some point in the season is not choosing whether to take a penalty. It is choosing when. The penalty costs a fixed number of positions everywhere, but the value of a position is not fixed anywhere. What a team is really deciding is how many of those places a fast car can win back on Sunday.

That makes the calculation a function of the circuit rather than of the penalty. Where the layout gives a quicker car a genuine route past a slower one, a fast driver starting near the back can recover most of a large drop by the flag, and the real cost of the weekend is a handful of points rather than the whole of them. Where passing depends almost entirely on the pit lane, the same penalty converts directly into a lost result, because the driver will finish roughly where the grid put them.

Three secondary factors push the decision around. A long race with high tyre degradation helps a recovering car, because it produces more strategic divergence for a fast driver to exploit, which is why this decision is usually taken alongside the tyre strategy for that weekend. A weekend with a sprint is more expensive to spend, because it puts two results at risk rather than one. And a weekend where the team is already compromised, by a crash, by a session lost to weather, is a cheaper weekend to spend, because the result was going to be poor anyway.

The counter-argument teams hold against all of this is championship position. A team fighting for a place in the constructors' standings cannot always take the theoretically correct weekend, because the rival it is fighting will score at that event whatever the model says. Deliberate penalties are therefore taken earliest by teams whose season is already settled and latest by teams still in a fight, which is why the pattern of new components across a season looks so different at the two ends of the grid.

Sprint weekends run two grids and two pools

At an event with a sprint, the procedure runs twice over two separate sets of penalties.

The sprint grid is built from the sprint qualifying classification using the same steps and the same fifteen-place threshold, counting unserved penalties for the sprint. The race grid is built from the main qualifying classification counting unserved penalties for the race. A stewards' penalty arising from an incident during sprint qualifying attaches to the sprint grid; almost everything else attaches to the race grid.

Component penalties belong to the race. A driver who takes a new engine on a sprint weekend serves that penalty on Sunday and starts the sprint from wherever sprint qualifying put them, which occasionally produces the odd sight of a heavily penalised driver starting the sprint near the front.

When the grid is published, and what a withdrawal does

The provisional starting grid appears no later than two hours before the scheduled formation lap. A team whose car cannot start, or which has good reason to think it will not be ready, must tell the stewards as early as possible and in any case no later than an hour and a quarter before the formation lap. If a car is withdrawn by then, the grid closes up behind it.

After that deadline the treatment changes: the slot of a car withdrawn or unable to start simply stays empty. The final grid is published an hour before the formation lap.

That distinction is the reason a grid sometimes has a visible hole in it and sometimes does not, and it is a small piece of procedural fairness. Beyond the deadline, everybody has planned a start from a known position, and moving them all up one to tidy the picture would change what they had prepared for.

How to read a penalised grid quickly

Four habits will get you to the right answer faster than the graphic on screen.

Add up the totals before you look at anybody's position. The grid is a sorting problem, and the only inputs are the qualifying order and each driver's unserved total. If you know both, you can build the grid yourself in under a minute.

Check whether anybody is over fifteen. That single fact changes the shape of everything below it, because a relegated driver leaves a slot that the closing-up step will hand to somebody else.

Do not treat the penalty number as places lost. Ask instead how many drivers are penalised and how far down the field the penalised drivers qualified, because those two facts determine the real cost.

Watch what a heavily penalised driver does in qualifying. If the team sends the car out for a serious lap when the grid position looks irrelevant, they are qualifying for the tie-break inside the relegated group, and that is a signal about how many other cars they expect to be down there with them.

More on the rules that decide how a grand prix weekend is ordered sits in our motorsport section, and the full run of explainers is in the blog.