The SPEC Specification

If we want to explore performance, we must first accept a simple truth: performance is not absolute. It only exists relative to a baseline.To speak meaningfully about performance, we must compare: against peers, against expectation, against an agreed reference. And ultimately, compare to ourselves. The PACE domain sets a foundation where these comparisons are meaningful. First and foremost it allows a direct comparison to the best version of ourselves. Transitively it allows a comparison between drivers, but even the race leader is racing against his own imperfections. This is a fundamental principle in PitWallGeek.

In Formula 1, information is profoundly asymmetric.

Teams operate with vast quantities of data that are never public: full telemetry, simulator models, practice runs, correlation tools, and years of historical context. Even so, that information is largely confined within team boundaries. Each team has deep visibility into its own two cars — and only partial visibility into the rest of the grid.

PitWallGeek does not live inside that world.

PitWallGeek operates strictly from the outside, using a much narrower but universally shared dataset: the official race timing records. This is the only data that is complete, consistent, and equally available across all drivers and teams.

That constraint is not a limitation — it is a design choice.

By relying exclusively on race time sheets, PWG accepts the same information boundary for everyone. What is lost in depth is gained in symmetry. What is lost in precision is gained in comparability.

Everything that follows — Race, PACE, and SPEC — is built within that boundary.

But there is a deeper problem.

Every race lap is inherently different. Fuel load changes lap by lap. Tires degrade, Track position, traffic, weather, and incidents continuously reshape the context in which a lap is driven.

The PACE domain sets a baseline where performance can be measured against the personal best and the field. It does so by reorganizing the race timetable, isolating the race circumstances; and by normalizing every lap for fuel and tire degradation.

It is also worth remembering what Formula 1 is — and what it is not.

Formula 1 is, first and foremost, a constructors’ championship.

The sport is explicitly designed to reward engineering excellence, not to produce a spec-racing world cup for drivers.

If what we wanted was pure spec racing, we already know where to find it. Formula 2 and Formula 3 provide tightly controlled machinery and a clearer view of driver skill — but they do so at a cost. Those series showcase the best rookies in motorsport, not the twenty most complete and refined drivers in the world.

Formula 1 sits at a different intersection: the highest concentration of driving talent, operating inside a deliberately unequal technical landscape.

SPEC does not attempt to turn Formula 1 into something it is not. It does not deny the constructors’ nature of the sport. It simply asks a counterfactual question — what patterns emerge if we temporarily constrain that inequality by rule?

At this point, a hard limit must be acknowledged.

The mathematical separation of car and driver performance in Formula 1 is indeterminate. With only two drivers per team sharing nominally the same machinery, the system is under-constrained. There are not enough independent observations to uniquely solve for driver contribution versus car contribution.

Teammate comparisons provide partial information, but they do not close the system.

As a result, any attempt to isolate driver performance requires assumptions. SPEC makes these assumptions explicit. Rather than pretending to solve an unsolvable equation, it applies a set of simple, empirical rules to constrain the problem and explore its consequences.

SPEC is not a decomposition. It is a rule-based abstraction.

SPEC therefore rests on a working hypothesis.

Within the current Formula 1 grid, the twenty drivers represent the highest tier of elite motorsport talent. While their average performance may be shaped by machinery, strategy, and circumstance, it is reasonable to assume that each of them is capable of extracting peak performance from their car — at least on a limited number of normalized laps.

In other words, while no driver may sustain peak performance continuously, all drivers occasionally touch it.

This hypothesis is supported, albeit imperfectly, by anecdotal evidence: when drivers change teams or machinery, their underlying pace often reappears, even as absolute results fluctuate. The signal is not constant, but it is persistent. SPEC is designed to look for that signal.

And yet, this hypothesis is imperfect.

There is clear evidence that contradicts it. The principle is well known in Formula 1: your teammate is not your mate. Even within identical machinery, large and persistent performance gaps sometimes emerge between teammates. These gaps are not always transient, nor can they always be explained by circumstance alone.

This does not invalidate the hypothesis — but it does bound it.

Not every driver extracts peak performance with the same frequency, consistency, or reliability. Psychological factors, confidence, adaptation windows, and driving style interactions with the car all play a role. The signal exists, but it is uneven.

SPEC does not deny these differences. It is explicitly designed to expose them.

The SPEC transformation therefore rests on two simple rules.

First, car performance is leveled. Differences attributable to machinery are constrained by a transparent, empirical normalization rule. This creates a common reference frame across teams.

Second, slower drivers are not promoted. SPEC does not elevate average or inconsistent performance to artificial parity. Normalization is anchored to demonstrated peak extraction, not to mean performance.

Together, these rules enforce a critical balance:

SPEC reduces the influence of the car without inventing performance that was never there.

The P80 Normalization Rule — leveling car performance

SPEC anchors car performance using the P80 lap time for each driver.

Rather than using best laps or averages, the 80th percentile represents a repeatable, representative level of performance that is largely free from outliers, incidents, or single-lap heroics.

For each team, the P80 lap times of its drivers define the team’s effective performance envelope. These are then compared across teams to establish a relative car-performance scale.

All laps for a given driver are scaled by a normalization factor derived from this P80 comparison, such that team performance is leveled to a common reference while preserving the shape of each driver’s lap-time distribution.

In short: P80 aligns machinery without rewriting driving style or race dynamics.

The n-5 Constraint — preventing artificial promotion

However, normalization alone is insufficient.

If applied globally, it could allow slower drivers to be unrealistically promoted ahead of clearly faster competitors — violating the working hypothesis and the observed reality of elite performance gaps.

To prevent this, SPEC applies a positional constraint:

a driver’s normalized performance may not be promoted beyond n-5 positions relative to the original competitive order defined by P80 performance.

This rule preserves competitive hierarchy while still allowing meaningful movement within it. It constrains SPEC to explore plausible counterfactuals, not fantasy outcomes.

In short: Cars are leveled, but drivers are not reinvented.

Taken together, these two rules define the SPEC transformation.

By leveling car performance and preventing the artificial promotion of slower drivers, SPEC creates a counterfactual comparison — one that is rigorous in method and honest about its limitations. It does not claim to solve the unsolvable separation of car and driver. It simply constrains the problem enough to make meaningful patterns visible.

SPEC is not reality. It is an alternative reference frame.

Within that frame, driver performance can be compared more directly, more fairly, and more transparently than in the raw race outcome. What emerges is not a ranking to be believed, but a question to be examined.

If you prefer, you can call it Fantasy F1.

But it is fantasy with rules, grounded in data, and explicit about what it assumes — which is more than can be said for most debates about driver performance.

And yes — it’s also a lot of fun.

SPEC is meant to be explored, argued with, and disagreed over. It’s an invitation to look at Formula 1 from a different angle, not a verdict to be defended.

So feel free to treat it for what it is: Fantasy F1 with a spine. Share it, challenge it, and take it apart — preferably over a pint with your buddies.

One final clarification.

The SPEC ranking is not absolute. It is not a declaration of who would win a spec championship, nor an attempt to crown an undisputed best driver.

SPEC is deliberately scoped to identify podium-capable contenders — those whose normalized performance places them plausibly within reach of the front. It answers a narrower and more interesting question: who belongs in the conversation when car performance is constrained.

In that sense, SPEC is less about ordering one through twenty, and more about separating signal from noise at the sharp end of the field.

What happens on the top step is still racing.