Nine Data Layers of an F1 Season: Reading 2026 Through Systems, Not Standings

01:47, Turin. I open the post-race analysis file and find nine empty worksheets.

Not a system error. I deleted them myself.

That race ended with a victory described as a "statement of intent", a championship table that had just stretched at the top, and roughly twenty headlines enough to build a complete story. I wiped it all: no pit-stop times, no tyre-degradation data, no note on the lap-31 call, no qualifying comparison between two drivers in the same car, no correlation chart between simulation and the real circuit. Nine sheets, nine layers, not a single cell with a number.

I keep this habit four times a season — usually after round four, round eight, round fourteen, and the finale. The purpose is not self-punishment. The purpose is to check how much I still remember without the file.

The average answer: about one third.

The rest is not memory. It is an illusion built from headlines.

August is the most dangerous moment in the year to tell those two things apart. The season is long enough to have produced patterns, but not long enough for those patterns to have disproved themselves. Every team has a story. Very few have a verified system.

Context: a season that has just replaced its entire frame of reference

In June 2026, the FIA published the technical regulations applying from 2026. Two years later, with the new generation of cars halfway through its first season under that rulebook, this championship is in the middle of its biggest rebuild since 2026.

The core, in a few lines. The power unit moves to a near 50/50 split between combustion and electric. Internal-combustion output drops to roughly 400 kW; electric output rises to roughly 350 kW. The MGU-H electric turbocharger is gone entirely. Fuel moves to a 100 percent sustainable blend. Active aerodynamics replaces DRS with two states: X-mode for straights and Z-mode for corners. Cars are shorter, narrower, and about 30 kg lighter than the previous generation.

Off track, the map of power has shifted too. Audi enters as a works team. Cadillac opens the twenty-second seat on the grid. Honda moves to Aston Martin. Ford partners with Red Bull Powertrains. Alpine becomes a Mercedes power-unit customer. Six manufacturers on one grid, where the previous era had four.

If you are reading this and thinking "everyone knows that", you are right. But the analytical consequences are far from common knowledge.

A regulatory rebuild destroys every statistical baseline. The average team gap from 2026 to 2026, the tyre-degradation model, the correlation between qualifying position and race result, even the threshold that separates a fast driver from a consistent one — all of it has an expiry date. Some of it has lost half its value. Some of it has lost all of it.

There is a paradox I have watched for fourteen years in this industry: at the exact moment when every old model loses credibility at once, the volume of published analysis goes up. Because when nobody knows the answer, people talk more.

I choose the opposite. When the frame of reference changes, I rebuild the frame of reference first, and only then speak.

That frame has nine layers. Not nine sequential steps — nine overlapping layers. A conclusion is only allowed to leave the table once it has passed through as many layers as the question demands, and it is stopped at whichever layer still lacks evidence.

The aerodynamics and car-concept layer

The car is the first layer because it is the only thing measurable with a stopwatch without interpretation.

But what is measurable is small, and what decides is large. A lap time tells you the result. It does not tell you where the result came from.

This layer has four variables I always record on the same row: the level of concept advancement, the correlation between wind-tunnel data and the real circuit, resource constraints, and the key data from the event.

Resource constraints are the most underrated part. The FIA allocates wind-tunnel and CFD allowances in reverse order of the previous season's standings: the champion gets the fewest runs, the last-placed team the most. A team finishing seventh has roughly seventy percent of the allowance of the last-placed team, but only about half of the leading team's allocation in the other direction. Add a cost cap of roughly 135 million USD per year on the operational side, and a real production lead time of six to ten weeks from drawing to circuit for an aerodynamic upgrade package.

Those three numbers together generate a simple rule most viewers skip: a team cannot upgrade faster than its factory allows, and cannot operate more freely than the FIA allowance grants. Every promise to "fight back in Europe" has to pass through those two valves.

The 2026 season makes this layer much harder to read. Active aerodynamics means the car's aerodynamic map changes shape as the driver moves between X-mode and Z-mode. That means a correlation model built in the wind tunnel under static conditions holds true for only half of the car's real running time. Correlation error in the first year of a new rule cycle is typically double what it is in year three.

Based on my experience tracking pre-season testing and the opening rounds, this is what I take from it: the launch car is not a statement. The first upgrade package is a statement. The launch car is the product of a year of simulation without feedback. The first upgrade is the product of six rounds with feedback. A team that holds its correlation through the first package has grasped the concept. A team that loses correlation at the first package is guessing.

Across regulation cycles in history, the first-year champion has almost always been the team with the steadiest correlation curve — not the team with the boldest idea. Both 2026 and 2026 repeated the same pattern.

This is where I have to warn myself. The conclusion "correlation is king" has one strong counter-example: a team can win a season by locking itself into a correct but undervalued concept, even when its correlation is worse than a rival's early on. The right concept for six months sometimes beats the right process. I keep that counter-example in the file, on the last line, before I sign off on any technical-layer judgment.

The pit window and race strategy layer

Strategy is the most context-dependent of the nine layers, and the one most often misreported.

Pit loss varies widely between circuits. At tracks with a short pit lane and a low speed limit, a stop costs roughly eighteen to twenty seconds. At tracks with a long pit lane, that figure can exceed twenty-five seconds. A seven-second spread sounds small, but multiplied by the remaining laps it decides the entire strategic window.

Every call on the pit wall is a three-variable equation: how much time is lost going in, how much is gained coming out on new tyres, and whether the current gap is enough to rejoin ahead. When a team calls a driver in three laps earlier than planned, that is not inspiration. It is an equation re-run after every lap.

What 2026 adds to that equation is energy.

With roughly 350 kW of electric power and no MGU-H to recover waste heat, drivers must manage an energy budget lap by lap. That budget is not a constant. It depends on whether the driver is running in clean or dirty air, on battery temperature, on how much of it was spent in the first sector.

The strategic consequence is very concrete. Running behind a rival in dirty air reduces drag, and therefore reduces energy consumption. A driver actively chasing for thirty laps can bank the portion of the budget that the car ahead has already burned. On the decisive lap, the car behind has more electricity.

The "de-rating" phenomenon at the end of a straight — when the system has to cut power to recharge — becomes a readable strategic signal. I spend a fair amount of this season logging when it appears on public telemetry, because it shows which phase of the plan a driver is in: attack, defence, or accumulation.

Alongside that comes the safety car and the virtual safety car. A safety car on lap thirty is worth something entirely different from the same safety car on lap fifty, because it compresses the entire field and gifts free time to those who were already ahead of plan. This is the only part of the strategy layer I call luck, and I always separate it from the skill part when I write.

The minimum evidence to assess a strategic call: the stop lap, the tyre compound before and after, the stint lengths, the pit loss at that circuit, and the points situation at the moment of the decision. Missing any one of those, any judgment should stay at directional level only.

The pairing and people layer

In an F1 team's entire data system, there is exactly one clean comparison: two drivers in the same car.

Every other comparison is polluted by car performance. The gap between a driver in team A and a driver in team B only says something if you have already factored the car variable out of the equation — and in the early phase of a new rule cycle, factoring it out is close to impossible.

So I read this layer through four indicators. Qualifying comparison between teammates. Race pace on the same compound. Consistency across rounds, measured by standard deviation rather than average. And the balance between the two cars in the team.

The balance between the two cars is the most neglected indicator. When a team has one car consistently at the back and one consistently at the front, the cause is usually not the driver. It is the process: two cars assembled at two different times, using two different batches of components, sometimes running two different aero configurations to gather control data. That is technically legitimate, but it ruins every conclusion about people.

2026 returns this layer to the centre, exactly as 2026 did. When the power unit is more complex and the energy budget becomes a strategic variable, a driver's energy-management skill becomes a genuine differentiator. A driver able to distribute electric power across a lap — knowing which section needs all of it, which section will accept losing a tenth — creates a gap that is not pure driving skill.

I watched this at testing sessions: two drivers in the same car, same tyre set, same fuel load, separated by nearly half a second in the third sector. That half-second did not come from the corner. It came from one of them already understanding the circuit's energy map and the other not.

This layer carries a causality trap I have to remind myself of every time I write. A driver finishing ahead of a teammate in three consecutive rounds does not prove he is faster. It only proves that in those three rounds he finished ahead. To use the word "faster", I need at least half a season and a large enough sample to strip out safety cars, pit errors, and strategy.

On the track there are twenty drivers, but the real race happens between two brains — and the third brain, on the pit wall, is often the one that decides.

The competitive hierarchy layer

A championship table is a photograph. The competitive hierarchy is a film.

I build this layer by dividing the grid into four groups: title contenders, podium contenders, midfield, and backmarkers. Which group a team belongs to is not decided by its current position but by the trend in its gap over the last eight to ten rounds, after removing the high-variance events — wet races, safety-car-heavy races, races with widespread technical failures.

The most important thing in this layer is distinguishing two completely different states of a rule cycle: convergence and divergence.

When the rules are stable, gaps between teams tend to narrow. Weak teams copy strong teams' solutions, wind tunnel and simulation converge, and eventually everyone hits the ceiling of the same rulebook. When the rules change, the trend reverses. Nobody knows where the ceiling is, so whichever team guesses the right direction moves far ahead of the rest before the others work out what is happening.

The 2026 season sits in the divergence phase. That has two opposite consequences I have to hold at once.

First consequence: the gaps in the standings at this stage exaggerate the truth. A team leading by forty points after ten rounds might be leading by sixty points in concept, or it might be leading by ten in concept and thirty through luck and rival unreliability. There is no way to tell those two possibilities apart by looking only at the standings.

Second consequence, and the one I consider more important: in a divergence phase, a team behind is limited in how fast it can catch up by the two valves already mentioned in the aero layer. The wind-tunnel allowance and the cost cap. A team behind has more testing allowance, but must spend most of it understanding where it went wrong rather than developing. This is the rarely mentioned trap: the testing allowance advantage is only worth something if the team knows what it is looking for.

There is another signal I track at this layer: the flow of technical talent. Historically, regulation rebuilds are followed by a wave of chief-engineer moves. Contracts include mandatory gardening leave before an engineer may work for a new team, and that clause exists because the knowledge inside an aerodynamicist's head is worth something for about six months. After that it depreciates.

That is why every piece of technical-talent movement news must carry a date. A name without a date is a meaningless name.

The regulation and governance layer

This is the layer most sensitive to precise language, and the one sports journalism gets wrong most often.

Regulatory analysis cannot rest on a summary. It must rest on the exact article, the exact appendix, the exact penalty schedule. Information paraphrased through three intermediaries can be wrong to a degree that changes the conclusion entirely.

Four clusters of issues I always check here: technical compliance through post-race scrutineering, the cost cap and financial regulations, sporting penalties and points deductions, and the impact of mid-season rule changes.

In 2026, the grey zone is wider than usual in three places.

The first is engine compression ratio. The rules measure compression ratio in static conditions, at ambient temperature. At operating temperature, metal expands and the effective ratio changes. The gap between those two measurements is a real grey zone, and it became a formal point of debate before the season began. How the FIA handles it — a technical directive, a regulation amendment, or an informal agreement — will determine the real value of a chunk of power across the whole grid.

The second is energy management. When performance depends on software that allocates electric power, the line between legitimate strategy and electronic intervention blurs. The sensors confirming active-aero state must prove that X-mode and Z-mode are genuinely used as declared, and not optimised around the boundary.

The third is the start procedure, where I expect arguments for the rest of the season. Whenever a launch process is standardised in software, some team will always find a way to shave the timing boundary.

The defensive authorship I recognise in myself has a flip side: once I have built a conclusion, I tend to protect it by ignoring contradictory facts. My defence against that habit in this layer is to always write out the strongest version of the opposing argument before writing my own conclusion. If I cannot write the other side persuasively, I do not yet understand the problem well enough to conclude.

The grey zone is not where the light is missing. It is where this sport is most real.

The seats and talent ecosystem layer

Driver markets are the layer where source quality matters more than content.

In transfer analysis, who reports something usually carries more information than what is reported. A journalist with direct paddock relationships, a mainstream sports outlet, and an account aggregating someone else's reporting are three entirely different confidence tiers. Ignore the tier, and you have no way to distinguish a real negotiation from a media pressure campaign.

The 2026 season introduces a variable unseen in years: a completely new team needing to hire two drivers at once, while the new rule cycle makes the value of old experience and the skill of new adaptation two sides of the same coin.

Three kinds of force drive the driver market, and they run on three different rhythms.

Performance force: a driver outperforming expectations raises his own value. This is the slowest rhythm, taking a season for the market to reprice.

Contract force: option clauses and performance clauses create hidden deadlines the public never sees. Some seats are decided two years before the news is published.

Media force: this is the fastest rhythm and the noisiest. A rumour appearing mid-summer-break usually has a specific purpose: to pressure a stalled negotiation, or to prepare a coming announcement.

I grade every rumour at three levels. High: reported by a source with direct relationships, with at least one independently verifiable detail — current contract length, a release clause, a meeting that took place. Medium: mainstream sports reporting without verifiable detail. Low: everything else.

What I do not do: grade a rumour high just because many outlets reprinted it. Three copies of the same source are still one source.

Every new contract is a hypothesis. The race is the experiment.

The risk profile layer

Risk is the layer I process first, even though it is usually presented last.

I split risk into six groups. Sporting: points lost to competitive error. Technical: power-unit reliability, the speed of getting upgrades into operation. Personnel: dependence on one key individual, or internal conflict. Regulatory and financial: breaching the cost cap, procedural risk. Public opinion: reputational risk with sponsors. Systemic: calendar, geopolitics, supply chain.

2026 pushes technical risk higher than it has been in years. A new power unit, with a much higher electrification ratio and no waste-heat recovery, means more undiscovered failure modes than usual. Across engine cycles in history, power-unit retirements in the first year are typically around double the rate of the third year of a cycle.

What matters in this layer is a rule I learned the uncomfortable way: the absence of a flagged risk does not mean the risk does not exist. It only means I have not found it yet.

In any analysis file I leave behind, unfilled risk cells are always explicitly marked as unfilled, never left blank. A blank cell in a spreadsheet reads like a zero. Zero is a conclusion. Unfilled is not.

That is the lesson I learned from an empty file at 01:47 in the morning.

The public expectation layer

This layer measures the distance between what is being told and what is happening.

I use four tests.

Fundamentals test: is the result being praised supported by underlying data, or only by the result itself.

Sample-size test: how many rounds that conclusion rests on. Three rounds do not create a trend. They create three data points.

Equipment-stripping test: if you remove the car variable entirely, what is left of that driver or team.

Heat-cycle test: whether a story is in its warming phase, its peak, or its cooling phase. The lifespan of a narrative usually matches the time it has gone unverified.

2026 generates a particularly contagious narrative: the story of a new era. It is compelling because it is structurally correct and vague in content. Anything can be a sign of the new era: an unexpected win, a slipping team, a young driver shining, an older driver slowing down.

That makes this the most manipulable of all nine layers. Any fact can be attached to the new-era frame and look like evidence.

My defence: whenever a narrative appears, I write down the condition under which it would be falsified. If I cannot write a falsification condition, it is not a narrative. It is a belief.

Esports taught me that the meta always changes. Football is the same, just one beat slower. F1 is two beats slower, but the rule cycle is not slow at all.

The industry transmission layer

The final layer is where a sporting event leaves the circuit and becomes an economic one.

The transmission chain has three tiers. Upstream: manufacturers, power units, driver academies. Midstream: teams, race promoters, and the commercial rights holder. Downstream: media, sponsorship, and derivative markets.

2026 hits all three tiers at once, which is rare.

Upstream, the number of manufacturers rises to six. That means the fixed cost of the whole championship rises, but also that the level of long-term commitment rises. A manufacturer that has already sunk money into an engine programme is unlikely to walk away within three years.

Midstream, the grid expands to twenty-two cars. More cars means more seats, more data, more content for media platforms. But it also means the gap between the front and the back of the grid may stretch once more in the first year, because a new team has to build from zero while existing teams already own factories.

Downstream, the commercial agreement between teams, promoters and the commercial rights holder is being renegotiated for the next period, with all eleven teams involved. This is the kind of event audiences barely notice, yet it determines revenue distribution, and therefore the financial competitiveness of the entire grid for the next half-decade.

I track this layer for a simple reason. Changes in the transmission layer do not show up on a stopwatch. But they decide who is still on the grid in five years, and therefore what all the numbers in the other eight layers mean.

I do not believe in titles. I believe in the system that operates to produce titles.

The contrarian angle: the blank-cell trap

This is where I return to the nine empty worksheets from the start.

The industry's common conclusion is that regulation rebuilds reward boldness. The team that dares to think differently wins. It is a compelling story and it is partly true.

But looking at the history of rule cycles, the pattern that repeats is not boldness. It is organisation. The winner of a cycle's first year is usually the team with the fastest feedback-loop process — simulation to wind tunnel to circuit and back to simulation. A bold team without that process is usually right in idea and wrong in execution, and being wrong in execution inside a new rule cycle is very hard to fix because you do not know where you are wrong.

The execution blind spot is the hardest thing to detect, because it does not appear as an error. It appears as a gap.

And that is the real lesson of the nine empty worksheets. The greatest danger in F1 analysis is not wrong data. Wrong data can be caught, because it contradicts other data. The greatest danger is a blank cell read as a zero — a gap read as an answer.

When a team does not publish the reason it dismissed an engineer, that cell is blank. When a driver does not discuss his contract clause, that cell is blank. When a team cannot explain why its upgrade package did not deliver the predicted time, that cell is blank.

Every blank cell is an unasked question. And in a season whose frame of reference has just changed, there are more blank cells than usual.

My theorem does not predict the champion. It predicts who collapses first.

And the first collapse is almost always the blank cell nobody was willing to fill in.

What to verify over the rest of the season

I am leaving three open questions for the finale, and I will answer them with data rather than with feeling.

Whether the leading group's second upgrade package delivers the time the model predicted. This is the real correlation test of the whole rule cycle, because the first package is always distorted by track learning.

Whether the power-unit retirement rate in the second half of the season falls compared with the first half. If it does not, the problem sits in design rather than assembly, and it will follow the teams into next season.

And finally: which midfield team has the flattest gap curve. A flat curve inside a divergence cycle is the most credible sign of a correct concept being built correctly.

An empty stadium is not abnormal. An empty stadium is an operating theatre.

An empty file is the same. It is only waiting for the next person willing to fill it in.