A Blank Cell Is Not a Zero: How Football Decides When the Data Goes Silent

A forty-page report sat untouched on my screen in Seoul, and every cell in it read the same thing: insufficient information. No team name, no match date, no metric, no player. The nine-section skeleton was still standing, with room for tactics, finance, medical, regulation, dressing room, risk, media and industry value chain, yet each section repeated a single sentence. My editor sent one line: so which club has a problem?

There was nothing strange about the question. It is the question this industry asks every week, and answers wrongly nearly every week. What kept me awake was not the nine empty sections but the reflex of an entire newsroom when it sees a blank cell: fill it. In a room full of confident men, I am the only one who brings the tape. That night my tape was blank too. But I knew something the spreadsheet did not say: a blank cell and a zero are two different things. A zero is data. A blank is the absence of data. This industry survives by treating them as the same.

In football that confusion has concrete consequences. A club that publishes no injury news is not necessarily healthy. A season with no financial sanctions is not proof of clean books. A player with no headline defensive metric is not necessarily a player who does not defend. Every time a column goes empty, a decision still gets made, and it gets made by something else: memory, instinct, crowd pressure, or simply the confidence of the loudest voice in the room.

Four hundred set-piece situations taught me that chaos follows an order. It took three more years to understand the reverse of that lesson: order only exists when someone bothers to count. When nobody counts, what remains is not chaos but a very smoothly told story.

When the toolkit becomes armour

Over fifteen years, football analysis moved from hand-written notes to a complete data supply chain. Upstream sits the academy and scouting network; midstream the clubs, leagues and governing bodies; downstream the broadcasters, sponsors, derivative product markets and data platforms selling to fans. Every link produces a metric, and every metric produces a genre of narrative.

We have learned to name them like tools in a box: expected goals to measure chance quality, passes allowed per defensive action to measure pressing intensity, financial fair play and profit and sustainability rules to measure financial health, third-party ownership to measure the legality of a deal, the solidarity mechanism to trace training footprints, the post-international-break syndrome to measure physical drain, the new-manager bounce to measure psychological recovery. This armour lets us speak far more precisely than the previous generation.

But every armour has gaps at the joints. The problem of this era is no longer a shortage of metrics; it is a shortage of the habit of naming which metric is missing. When a report comes back empty, the system behind it rarely stops. It switches to default mode: reading silence as a positive signal, reading the absence of bad news as good news. That is the moment analysis stops doing its job and starts doing public relations.

Expected goals measure chances, not intentions

The first commercial expected goals model was presented by Sam Green during his time at Opta, somewhere between 2026 and 2026, and was later developed into many versions by providers such as StatsBomb and Understat. The core idea is simple: assign every shot a scoring probability based on location, angle, shot type, number of defenders in front and a few other variables, then add them up. Thanks to it, a team that loses 0-1 with 2.4 expected goals is judged to have played well.

The first problem is cross-provider variance. The same shot can produce figures differing by three tenths of a goal depending on the model. That is not mathematically wrong, but it means no single correct expected goals figure exists to argue over. Every time I read a piece claiming team A deserved to win because its expected goals were higher, I wonder which model was used, and whether the author knows they are arguing with one specific version.

The second problem is more serious: expected goals measure chance quality, not decision quality. The model does not know a team deliberately ceded the ball to draw the opponent up, and does not know a shot in the 89th minute came from a counterattack rehearsed since the 60th. In Kazan, on 27 June 2026, Germany held close to seventy percent of possession and produced roughly twice as many attempts as Korea. Korea 2-0 Germany was not an earthquake; it was a formula that lazy people call luck. Kim Young-gwon opened the scoring in the 90+2nd minute after a video review, and Son Heung-min sealed it in the 90+6th with the German goalkeeper stranded upfield. Look only at the final expected goals column and you see a one-sided match. Look at the geometry of the German back line in the last twenty minutes and you see an eighteen-metre gap behind both full-backs, and that is what decided the result.

Pressing numbers and the small-sample trap

The metric counting how many passes an opponent completes per defensive action, commonly known as PPDA, became the standard measure of pressing intensity in the 2010s. Lower is more aggressive. Marcelo Bielsa's Leeds United in 2026-21 is the textbook case: the lowest PPDA in the Premier League, dropping below eight for much of the season, and the eye confirmed it.

But PPDA does not distinguish good pressing from chaotic pressing. A team that charges forward without structure can post a flattering number over a few matches, because every time an opponent passes and one of their players swings a leg at it, a defensive action is recorded. The paradox is that the best pressing team and the most naive pressing team can sit side by side in the same table.

That is why I always add a second column next to PPDA: how many times a team is played through after losing the ball in the opponent's half. For Leeds that season, the second column was healthy. For other teams with equivalent PPDA, the second column was an open wound. With one number, you tell a story. With two numbers set against each other, you begin to have a hypothesis.

The blank cells in the medical file

No area misreads silence more than medical information. Clubs publish injuries to different standards; some state recovery windows, others write only not available for selection. A player absent from a squad list is not necessarily fit; sometimes it means the club is keeping quiet for transfer reasons.

On 6 February 2026, the Premier League announced charges against Manchester City, a figure widely reported as 115 alleged breaches of financial rules spanning many years. The independent hearing began in September 2026. My point is not the substance of the case but how public debate handled the gap behind it. For nearly twenty months between those two dates there was no ruling. And many articles filled that gap with one of two extremes: reading the delay as proof of innocence, or reading it as proof of guilt. Both are fake zeros written into a blank cell.

From financial fair play to the seventy percent ceiling

UEFA's financial fair play rules were introduced in 2026 and applied from the 2026-12 season, requiring clubs in European competition to balance income and spending. In April 2026, UEFA replaced them with the Financial Sustainability Regulations, including a squad cost rule: spending on player wages, transfer fees and agent commissions may not exceed a percentage of revenue, on a declining path from ninety percent in 2026-24, to eighty percent in 2026-25, and seventy percent from 2026-26.

The framework looks elegant on paper, and it carries one structural blind spot: revenue is defined, while contract structure is flexible. A club can keep total costs identical while changing how they appear in the accounts through contract length, signing-on fees and deferred bonuses. When the rules change, the money does not disappear. It moves to another column.

Four points deductions and a new benchmark

In England, the Profit and Sustainability Rules permit a club to lose a maximum of one hundred and five million pounds across three seasons. The 2026-24 season became the first in which this mechanism produced direct table consequences. On 17 November 2026, Everton were deducted ten points; on 26 February 2026, the sanction was reduced to six on appeal. On 18 March 2026, Nottingham Forest were deducted four points. On 8 April 2026, Everton received a further two-point deduction for a second breach.

What matters is not the numbers but the fact that before 2026 almost the entire debate about football finance took place in a space with no penalties. Without precedent, people assumed there was no breach. That was a logical error, but a comfortable one, and it survived fifteen years.

Eight-year contracts and the loophole closed in 2026

In the January 2026 transfer window, Benfica announced the sale of Enzo Fernández to Chelsea for one hundred and twenty-one million euros. The interesting part is not the fee but the term: a contract running to 2031, roughly eight and a half years. In August 2026, Chelsea completed the signing of Moisés Caicedo from Brighton for a reported one hundred and fifteen million pounds, on a deal to 2031 with an option for a further year.

In football accounting, transfer fees are amortised over the contract term. Dividing one hundred and twenty-one million euros across eight and a half years produces a much smaller annual figure than dividing it across four. This is not fraud; it is using the existing rules correctly. But it shows what I repeat in every transfer-market piece: any spending rule can be neutralised by changing the axis of time.

In July 2026, UEFA amended its rules to cap amortisation at five years for new contracts. The Premier League later approved a similar limit. The door closed, but deals already signed kept their original terms. This is the common feature of every reform in football: it applies only to the future, while the past has already been sealed.

The 2026 West Ham lesson and third-party ownership

In 2026, West Ham signed Carlos Tevez and Javier Mascherano in deals involving an investment company holding the economic rights to the players. In April 2026, the Premier League fined the club five and a half million pounds for breaching third-party ownership rules. Sheffield United, relegated that season, later received a reported settlement above twenty million pounds.

FIFA subsequently banned third-party ownership, effective from 1 May 2026. From a governance standpoint this was one of the clearest reforms of the decade. From an enforcement standpoint it is a perfect example of how incomplete data creates grey zones. A company may not own a player's economic rights while still holding a share of future cash flow through sponsorship contracts, loans, or relationships with agents. No data cell records that, and because no cell records it, it barely exists in public discussion.

Where the five percent goes

FIFA's solidarity mechanism distributes five percent of the value of an international transfer to the clubs that trained the player between the ages of twelve and twenty-three. The split is fairly specific: each season a player spent at a club between twelve and fifteen counts for five percent, and each season between sixteen and twenty-three counts for ten percent, capped in total at five percent of the transfer value.

This is one of the few mechanisms that turns a training footprint into verifiable data. It also carries a familiar paradox: the money flows to small clubs, where record-keeping is weakest and where few people can pursue an international procedure. If you want to see how silent data causes material harm, find an academy in a provincial town with no accurate record of how many seasons a player spent there.

The money nobody audits

This is the position I have held for a decade, and it gets clearer every year. Signing-on fees for free agents damage football's transparency more than transfer fees do, because they sit outside almost every core monitoring mechanism.

Look at the sequence of free transfers in recent seasons. In 2026, David Alaba left Bayern Munich for Real Madrid on a free transfer; Gianluigi Donnarumma left AC Milan for Paris Saint-Germain, also free; Lionel Messi left Barcelona for Paris Saint-Germain, also free. In 2026, Antonio Rüdiger moved from Chelsea to Real Madrid on a free. In June 2026, Kylian Mbappé joined Real Madrid after his Paris Saint-Germain contract expired, officially announced on 3 June 2026.

In summer spending tables, every one of those deals displays a value of zero. But no club acquires a world-class player without paying something. That something exists under the names of signing-on fee, agent commission, contract bonus, loyalty bonus. It does not appear in spending tables, does not appear in articles ranking the biggest spenders, and in many cases is not treated like a transfer fee in squad cost calculations. Real money recorded in an unnamed column. That is the most precise definition of a dangerous blank cell.

I am not saying free transfers are wrong. I am saying a monitoring system that only counts what passes through the front door will always undervalue what passes through the back.

Barcelona, the levers, and the price of creative accounting

In 2026, Barcelona sold twenty-five percent of their La Liga television rights for twenty-five years, along with stakes in Barça Studios. In the accounts these appeared as revenue, helping the club meet its spending limit. In reality they were future revenue spent in advance.

This is the clearest illustration of the argument I keep making: for the same economic event, classification decides the entire story. No rule was broken. But read only the revenue column and you would believe the club was getting stronger, when the substance of the deal was selling off the cash flow of the next twenty-five seasons.

The post-international-break syndrome and the ten-match sample

The phrase post-international-break syndrome appears in media to describe clubs losing players to injury and overload after each national team window. The problem is that it is rarely measured. Leagues do not publish injury data to a common standard, and clubs have incentives not to disclose detail. The result is a widely acknowledged phenomenon that cannot be fully quantified.

The same applies to the new-manager bounce. Studies typically record short-term improvement of around three to five tenths of a point per match across the first ten matches after a change, before the team returns to its previous trajectory. But a ten-match sample is too small to separate genuine tactical change from regression to the mean. I ran this against K League data myself and found the same thing: most of the jump came from a bad run ending, not from a reorganised back line.

The blind spot: this industry rewards fake completeness

Here I want to argue against myself. My job pays for pieces with conclusions, and a blank cell produces no conclusion. Nobody shares an article saying there is not enough data to judge. Market incentives lean toward whoever dares to assert, and so the whole industry develops a systemic bias: turning scarcity into certainty.

I have made exactly that mistake, in a different form. For years I explained almost every goal through a set-piece situation. Set pieces are my mother tongue, and when you master a language you start translating everything into it. But on the night I checked my own work, across the last twelve matches of one K League club, four goals conceded came from passing errors in their own half under no pressure at all. No dead ball. No diagram. Just a centre-back playing a lazy square pass, and an opponent who still remembered his former teammate's name. That kind of error sits in none of my models, and if I refuse to look at that gap, I will turn four hundred set-piece situations into a religion instead of a tool.

Football is not only dead balls, and a team is not only variables. I once stood in Seongnam on a rainy afternoon, watching a thirty-four-year-old run laps around the touchline after his club had already been relegated, with no camera recording it and no metric capturing it, and the only thing I could honestly write about him that day was that he was still there. Prejudice is like a high defensive line: it only takes one correct pass for it to fall apart.

The test for next week

A pandemic season, four hundred set-piece situations, and I learned to speak the language of space. But that language is only useful when I admit I have not heard the whole sentence.

So from the next matchday, do one simple thing. Every time you read an analysis, find out which metric the author used, and whether the author names the metric that is missing. If a piece claims team A defends well without citing how often they are played through, treat it as a blank cell. If a piece discusses a club signing a free agent without mentioning the signing-on fee, treat it as a blank cell. And if we have the patience to count those blank cells across a full season, we may discover that what football lacks is not data, but the habit of admitting we do not have it yet.