top of page
Search

What Norris' DNF Revealed About F1's Telemetry Blind Spot

  • Writer: Tim Harmon
    Tim Harmon
  • 5 days ago
  • 4 min read

McLaren’s 1000th Grand Prix gave us more than a lap-45 retirement — it showed why teams need a physics-aware way to detect when live telemetry stops matching reality.


The article was originally published on LinkedIn on June 10, 2026. Read the original here: LinkedIn Article


McLaren MCL40 in the team’s 1000th Grand Prix livery, used here to frame the Monaco telemetry case study behind Project Apex.
The MCL40 in its 1000th Grand Prix livery — Monaco 2026, where reliability became the story. Image: The Race / McLaren Racing

Monaco Made The Problem Visible


On McLaren’s 1000th Grand Prix weekend in Monaco, Oscar Piastri delivered fourth place while Lando Norris retired on lap 45 with a power-unit issue. McLaren said the team identified “an anomaly with the power unit” that had not appeared before the race, tried mitigation steps and steering-wheel setting changes from the cockpit, and still lost the car before the finish.

That is the kind of failure that matters to me with Project Apex. It is not just a retirement; it is a live test of whether the telemetry still reflects the car’s actual physical state as the hardware begins to drift from normal behavior.

What The Public Result Misses


From the outside, the timing screen reduces the event to “RET — Power Unit.” From the inside, the more important story is that the anomaly escaped pre-race checks, evolved in real time during the Grand Prix, and forced the team to manage an unknown failure mode.


That is the blind spot Project Apex targets. I want a physics-aware validator running alongside the normal telemetry stack, continuously asking a simple question: do the numbers still add up to something this car can physically do?

I Tested The Idea on Norris’ Monaco Lap


To explore that question, I used FastF1 to pull Lando Norris’ fastest race lap from Monaco 2026, exported the telemetry to CSV, and analyzed speed, throttle, brake, RPM, gear, and distance in MATLAB.


Speed-versus-distance plot of Lando Norris’s fastest 2026 Monaco Grand Prix lap, highlighting braking zones, acceleration phases, and corner-by-corner pace changes.
Norris’ fastest lap maps Monaco’s violence in one trace.

The speed trace shows the circuit’s signature immediately: heavy braking into Sainte Devote, the long acceleration through the tunnel, and the low-speed sequence through the Swimming Pool and Rascasse. The lap does not prove the later failure on its own, but it gives us a clean baseline for what a representative Monaco lap looked like before the retirement.


Throttle and brake traces versus distance for Norris’s Monaco lap, showing driver input transitions and control patterns through key corners.
Monaco forces brutal transitions between full commitment and heavy braking.

The throttle-and-brake trace shows where the car demands maximum commitment and where it sheds speed aggressively in a narrow operating window. Those are exactly the regions where a power-unit anomaly, sensor inconsistency, or control-state mismatch can become operationally expensive very quickly.

From Post-Race Plots to Real-Time Validation


Pretty plots do not solve the problem. A useful system needs to define a physics envelope before the session starts, then compare every live sample against both the car’s prior behavior in the event and the physical relationships that should still hold between speed, RPM, gear, acceleration, and energy state.


That is the core of Apex. If RPM and speed no longer match the known transmission state, or if the implied energy behavior no longer matches the declared control state, the system should raise a clear, explainable warning before the car reaches a terminal fault.


Gap-detection visualization from Project Apex showing where missing or inconsistent telemetry samples create discontinuities that can distort motorsport analysis.
Apex flags the moments the telemetry stream stops behaving normally.

As a first step, I examined sampling regularity by measuring the distance between consecutive telemetry samples. In this Monaco lap, the distribution clusters around a normal step size, but several spikes break sharply away from that baseline, which is exactly the kind of integrity signal a real-time validator should surface rather than leave buried in the raw data.

Why This Matters Now


McLaren’s Monaco report describes the exact kind of problem a validation layer should help expose earlier: an unexpected power-unit anomaly that was not visible before the race and could only be managed for a limited time once it appeared. That doesn’t identify the root cause, and I am not claiming this lap-level analysis does. It does show why teams need a shared, physics-based way to see when the telemetry narrative starts diverging from the car’s physical reality.


That is why I am sharing Project Apex now. I am building an always-on, explainable telemetry-validation layer for high-stakes motorsport systems — not to replace race engineers, but to give them earlier, clearer evidence that a car is moving toward a failure state.

Why The Right People Should Care


Reliability is not a back-office concern. In modern F1, it is a competitive, operational, and safety problem that sits at the intersection of race engineering, controls, software, data integrity, and decision-making under pressure.


If you work in Formula 1, Formula E, WEC, IndyCar, or any other high-consequence series, this is the conversation I want to have: how do we detect the moment a live telemetry stream stops telling the truth? That is the problem space for Project Apex.

Thanks for reading. If you found this useful, feel free to share it with other engineers, data scientists, and motorsport folks who care about reliability and integrity as much as they do about outright speed.

Timothy D. Harmon, CISSP, is a Lead Enterprise Architect, motorsport official, and the author of Project Apex, a 2026-focused cyber-physical telemetry validation concept for Formula 1. He presented “Hacking Physics at 300 KPH” at BSides San Diego 2026 and writes at The Secure Accelerator.

References

Comments

bottom of page