Empty Cells, Empty Verdicts: The Integrity Test of Football Tactical Analysis

It is 1:47 in the morning. At my data desk in Valencia, my second cup of coffee went cold long ago. On the screen sit two columns — input on the left, output on the right. The right side is completely blank; not a single cell is filled. The first stage of the analysis pipeline has come back empty-handed. I moved the mouse, refreshed the file, switched the encoding and ran it again — the same result. Zero. Eight years ago I might have filled those empty cells with my own guesses. But since 2026 I no longer fill cells with feeling. That day I paused the tape at frame 47, and the whole newsletter was hiding right there. The lesson was plain — build a story out of what is absent and it stops being analysis and becomes imagination. That blank that evening became my most valuable piece of information. Late in 2026, at Valencia's Mestalla, I wrote a twelve-part thread on Marcelino's 4-4-2 mid-block. Valencia had beaten Athletic Club 2-1, but the scoreline was not my subject. My subject was the measured distance between the two banks of four. Forty-seven annotated freeze-frames, each carrying a number — how many metres of gap, at what angle the passing lane opened. The thread drew 2.1 million impressions, a reply from a La Liga analyst, and by December my weekly space-map newsletter had 38,000 subscribers. From that day my writing changed shape. I stopped describing what players did and started measuring where they stood. Every piece is now anchored by a distance, a passing-lane angle, or the gap between banks. That numeric spine is my signature readability. Football tactical analysis today stands on two stages. The first is deconstruction — recording, event data, passing networks, press maps, xG, PPDA, from which information points are separated out. The second is analysis — using those points to read formations, space, and coaching decisions. The two depend on each other. If the first stage returns empty, the second has nothing in its hands. That is the brutal rule of my trade. Without a passing network I cannot talk about formation. With a blank press map I cannot measure the height of the mid-block. If a coaching change is missing from the record, what do I evaluate the decision with? Every sentence of analysis is a debt owed to an information point — and a claim written without repaying that debt is a default. I learned this rule in the bone at Russia 2026. In Moscow, for Spain's round-of-16 tie with Russia, I charted passes live. Spain completed 1,029 of 1,137 attempted passes, held 74 percent possession, took 25 shots — and still drew 1-1, losing 4-3 on penalties, Igor Akinfeev saving Iago Aspas's kick. Ninety minutes after the final whistle I filed the breakdown, showing that most of Spain's passes arrived in zones where shot probability was near zero. That night, sitting in the Moscow press tribune, I heard the journalists around me marvel at Spain's passing numbers; someone called it a World Cup record. But my ear caught a different sound — a silence. After 1,029 passes there was no joy behind the net. In the ledger of ball usage, Spain were nearly bankrupt. The lesson was simple — possession volume is a diagnosis, not a virtue. The possession ledger said 62 percent; the truth lived in the other 38. The real question is where the ball died; where it circulated is secondary. A side that takes 1,029 passes and cannot build a single goal has a real problem in how it uses the ball. I was fortunate to learn this on the evening of a defeat; otherwise the dazzle of the statistics would have deceived me. Since then every piece opens with a where-the-ball-went ledger — completed passes by zone, then one sentence naming the zone that actually mattered. Today the rhythm of my work is the same — metres, not metaphors; distances, not drama. If there is genuinely a three-metre gap between a full-back and a centre-back, that is my pen. This is why an empty input no longer frightens me; it calms me. An empty input is an honest answer. The false answer comes when someone forces the blank cell full. The most dangerous moment in my trade is when, under the pressure of an empty cell, an analyst invents a team, a scoreline, a cause — just to fill the space. Every day I ask one question — what else changed? When a team suddenly plays well, my first question is not praise of that team; my first question is what changed in the opponent. Which player returned, which defender was missing, was the pitch dry or wet? Cause and result are two different things. Assuming that whatever changed is the cause is the most common error in analysis. I read a player as an account with a balance and an outstanding sum. How many minutes were spent, how much load remains. If a side plays again within five days, its ledger is different. Analysis that ignores this load-budget thinking and looks only at the scoreline stays incomplete. But all of this has one precondition — the information must exist. Measuring a load budget needs a player's minutes. Checking a confound needs the opponent's line-up. Verifying the chain needs the event log. Without information the method is inert. In a tournament cycle this lesson matters even more. During a World Cup or a continental finals, information is lost in the storm of emotion and flags. The stands fill up, the analysis runs empty. That is exactly when verifying the input file matters most. Before every match of a tournament I ask three questions — who is playing, how much load remains, and how reliable the information is. The third question is the most neglected, and yet the most important. Consider the three stages of a data pipeline — collection, analysis, publication. The first stage failed and returned empty. If the second stage now invents teams, players, competitions on its own, and the third prints it as a finished report, the reader receives a tidy, fluent, clean — and entirely wrong — analysis. The error will not be noticed, because the error is smooth. In 29 years of work I have seen this trap many times. In 2026, after finishing a civil-engineering degree, when I entered journalism, I kept one rule from the start — no claim without a source. What I have done for decades as editor of Krira Jagat is the same work: verify the information, verify the source, then publish. The true beauty of analysis lies in its honesty. Sitting in Valencia's press box, the first thing I still write in my notebook is: what is the input? If the input is empty, the conclusion stays empty too — that is methodological honesty. Zero decisions from zero information is the only respectable outcome. Here the counter-argument arrives, and it is this — the biggest enemy of analysis is empty information; bad information at least gets caught. A wrong pass count raises suspicion, the source is checked, it is corrected. But nobody suspects a blank space. Faced with an empty cell, the reader's mind raises no question, because there is nothing there to question. The rules of data lineage are strict. A number can be traced back three places — the match event log, from there the pass map, from there the decision. Cut one link in this chain and the whole decision hangs loose. But in practice we do not verify the chain; we only look at whether the final result is pretty. The risk of my trade is therefore moral more than technical. When a generative layer runs over an empty input, it can conjure teams, players, scorelines out of nothing — without shame. This invented information is so clean it looks more credible than the truth. The reader never gets the chance to verify. That is why I consider it more important in analysis to show where the fact came from than to surprise the reader. More important than a striker's cool head in front of goal is how strong each link in the data chain is. Before opening the next match's scoreboard I will open the input file first. I will see whether the list of information points is full or empty. Because an empty list is not a failure to me — it is a question. The professional analyst's job is not to fill the question with a blank answer, but to place the question honestly before the reader. In my next thread I will do exactly that — I will count the cells first, and the score after.