The Empty Report in the F1 Backroom and the Lesson of a Lagging Sensor at San Siro

Summer 2026 and a Number That Refused to Match

In the summer of 2026, at 48, working as a member of the AC Milan coaching staff, I was handed a task nobody wanted: validating the motion-tracking dataset from 20 Serie A matches in the 2026-17 season before it fed into the tactical report. Milan's expected-goals figure at San Siro was 1.85. Away from home it dropped to 1.02. Actual goals scored in the two settings were almost identical.

A team creating far better chances at home while scoring at the same rate on the road always comes with a very easy story attached: psychology, crowd pressure, confidence. I accepted none of it. I pulled the match footage and cross-checked every build-up from the goalkeeper, and found what I was looking for: a sensor in the south-west corner of San Siro was responding 0.2 seconds late. Every ball played out of the box was recorded off-beat.

The 14-page internal report I wrote recommended recalibrating the equipment. Head coach Vincenzo Montella used the finding to shift more circulation to the right flank. Milan won 5 of their last 8 matches and claimed a Europa League place. I retell that old story because this week I found its mirror image a thousand kilometres from San Siro: the analysis backroom of Formula 1.

A Two-Layer Pipeline and an Empty Envelope

Professional sports analysis now runs on a two-layer pipeline. The first layer deconstructs the source: title, outlet, article type, one-sentence summary, author stance, purpose, information points, core viewpoints, entities, time sensitivity and source quality. The second layer is where deep analysis happens: technical and car, race strategy, team and driver, competitive landscape, regulations and governance, driver market, risk profile, public narrative, and industry transmission.

The report I read this week carried all nine dimensions of the second layer. It had comparison tables, a risk matrix, a transmission-chain diagram, a glossary at the bottom, even a disclaimer. Formally, it was flawless. But when I flipped to the first layer, I found blank space: no title, no source, an empty summary, an empty information-point list, no entities, time sensitivity unassessed, source quality ungraded. All nine analytical dimensions returned the same sentence over and over: insufficient information.

The most telling detail sat in one small cell. The domain label should have been normalised to "F1/Motorsport" — capital F, capital M, a slash between. It arrived as "f1", lowercase, with the tail missing. Which means some normalisation step ran halfway and stopped. For anyone who reads system errors for a living, a half-finished signal is always worth more than total silence.

The Best Engineer Is the One Who Notices a Dead Channel

On any pit wall there is an unwritten rule I learned across more than four decades in this trade: the best engineer on the pit wall is the one who spots which data channel has gone dark, not the one who reads the most numbers. A dead telemetry channel does not cause a crash. An engineer who believes it is still live and extrapolates from it does.

Three paths to an empty envelope. When the deconstruction layer returns every field blank, three causes are possible, ranked by probability. First, the fetch failed: the source page sits behind a paywall, or is JavaScript-rendered so the scraper only received an empty body, or the server blocked access. Second, the extraction model errored or timed out and emitted an empty schema instead of raising a flag. Third, the source document really was empty — a placeholder draft nobody ever wrote.

The lowercase "f1" leans toward the second cause. A total fetch failure would leave the domain label entirely absent, not truncated to a partial value. The fact that the pipeline ran as far as normalisation before breaking suggests the system started, processed part of the input, then failed mid-way. For an observer, that is evidence with weight — weak, but directional.

A Completeness Gate

What deserves attention is that this entire incident could have been stopped by an extremely simple gate: if the information-point count equals zero, halt every downstream analysis step. The minimum threshold for a valid deconstruction could be set at one title, three information points and one named entity. Those three numbers need no sophisticated algorithm. They need one person to decide that publishing nothing beats publishing a hollow analysis.

In F1, the line between "no data yet" and "data equals zero" gets blurred by time pressure. A strategist on the pit wall has a few dozen seconds to decide whether to call a driver in or leave him out. If the fuel-consumption channel goes silent at that exact moment, he has two choices: say plainly over the radio that the channel is dead and he has no number, or extrapolate from the previous lap. The second option sounds brave. It is also the option that turns a correct strategy into an unfixable mistake.

The Lesson of a Sensor 0.2 Seconds Late

Back to San Siro. What makes the 2026 story memorable is not the 1.85 against 1.02 expected-goals gap. What makes it memorable is that the sensor was not broken. It was working, transmitting, producing numbers that looked entirely normal. It was simply 0.2 seconds late. In a sport where a counter-attack is decided in less time than that, 0.2 seconds is enough to turn an offside-breaking pass into a wasted ball, and a wasted ball into an offside-breaking pass.

The pipeline failure I read this week sits on the opposite side of the same problem. At San Siro the data was wrong, but there was data. Here the data was right, but there was nothing at all. Both cases lead into the same trap: the analyst has to fill the gap with something, and the cheapest thing to fill it with is memory, instinct, and pre-existing expectation.

Every tracking number belongs on the operating table, not on the altar. I told my colleagues at Milanello that in 2026, and I still use it when working with lap data. But it needs completing: an empty channel belongs on the operating table exactly like a beautiful number. Blank space in a data table is no place to keep writing from instinct.

Russia 2026 and the Limits of Turning Numbers into Images

In 2026, thanks to the previous year's internal report, Sky Sport Italia invited me as a specialist commentator for the World Cup. Germany against South Korea, minute 70, I posted one short line: Germany's defensive line was sitting an average of 68 metres high, pressing had failed 17 times, South Korea already had 12 counter-attacks; without dropping the block, the goal would come from an aerial situation. In the 90+3rd minute Kim Young-gwon scored exactly to that script. Thousands of accounts mocked me for turning sporting emotion into arithmetic. Gazzetta dello Sport still republished the piece alongside the distorted-trapezoid diagram I drew of Germany's back line.

From that night I drew a professional rule: numbers must be translated into spatial images before readers remember them. I stopped writing "68 metres high" and started writing "the zipper has burst open to the valve box". I stopped writing isolated figures about the gap between centre-back and goalkeeper, replacing them with the image of a gap as wide as a vertical rectangle.

That rule carries a consequence few notice: when the data is empty, you cannot translate it into an image. There is no image to draw from blank space. If that nine-dimension analysis were filled in by someone from memory of the season, it would stop being analysis and become a piece of storytelling dressed in statistical clothing. In Vietnam, sports data platforms such as VuaBong.vn face precisely this problem when they cross-check sources before publishing any figure.

Why a Hollow Report Is More Dangerous Than a Wrong One

A wrong analysis can be caught. You check a second source, you re-examine the measurement window, you discover the sensor was running late. A hollow analysis has nothing to catch, because it has not said anything yet. It is simply waiting for someone to fill it in.

In sports media that temptation is larger than outsiders imagine. Deadlines run ahead of kick-off. An editor holds a complete analytical skeleton, all sections present, all tables present, and a source dataset that is empty. Fixing the source costs time. Filling the blanks costs minutes. This is where the F1 data pipeline and the football data pipeline share the same hole: both are designed to produce conclusions, and neither is designed to refuse to produce one.

Data only tells part of the story; the rest lives in whether people know how to listen. But listening requires a voice first. When the deconstruction layer returns blank space, the correct action is not to guess what that voice should have said, but to record that it was absent, where, when, and how far down the pipeline things ran before the break.

The Blind Spot Sits Where Nobody Wants an Empty Table

The counter-intuitive angle I want on the table is this. The most correct conclusion in this week's report is not inside its nine analytical dimensions. It sits in the note stating that a hollow payload should be treated as a pipeline incident requiring a halt, not as content of low value.

That sounds like a dry administrative finding. Placed next to the San Siro story, it becomes one of the week's most valuable discoveries: the likeliest path for a sports analysis to go wrong is someone deciding they cannot tolerate an empty table, not dirty data.

One more point, stated plainly. F1 teams almost never disclose the nights their telemetry channels died. They publish upgrade packages, lap times, contracts. Internal blanks stay internal. So when a hollow analysis leaks into the light, it accidentally reveals something public data never shows: this industry's data systems still have joints nobody has inspected.

Every collapse has a premise; few people bother looking before it happens. This time the premise lived in a truncated label, in an extraction model returning an empty schema instead of an error, and in the absence of any gate to stop the next step. For Germany in Russia in 2026, the premise was 17 failed presses and 12 counter-attacks counted before the goal arrived. For the data pipeline, the premise was three blank fields and a misformatted label.

From a training ground in Milan to a data screen in Milan, the law of empty space is the same. Wherever nothing exists, it gets filled by the cheapest available thing.

What I Will Check Before Next Season

In my own analytical work I will keep a habit that has followed me since 2026: before reading any conclusion, I verify the source of the numbers. From this week I add one more step — checking whether that source actually exists, or is merely a carefully packaged envelope.

A nine-dimension analysis with full tables can still contain exactly zero information points. Readers deserve to know that before they are persuaded by its form. And when a team enters a decisive phase of the season, the first thing I want to know will be: is the data pipeline behind the numbers we trust still flowing?

Glossary

• ATR (Aerodynamic Testing Restriction): the system allocating wind tunnel and CFD allowances in reverse order of the previous season's constructors' standings.
• Technical Directive (TD): a governing-body document clarifying or tightening the interpretation of existing rules.
• Scrutineering: post-session technical inspection confirming a car's regulatory compliance.
• Silly season: the peak period of driver-transfer rumour activity.
• Gardening leave: the mandatory waiting period for an engineer between two teams.
• Completeness gate: the minimum input threshold that permits the next analytical step to run.