Empty Stadiums and the 68% Number: How Data Is Reshaping the Way We Read Football

Across Levante UD's 47 matches in the 2026-17 season, one number kept me awake: 68% of their goals conceded came from the left flank. Not from lightning counter-attacks or long-range shots outside the box, but from a specific zone on the pitch, repeating an almost identical pattern of movement. I rewatched 31 hours of footage and drew 214 attacking diagrams to be sure I was not misreading it. The number was not wrong. The problem was that no one wanted to look at it.

That was my debut piece as an independent analyst. The editor of a new outlet was not convinced. He asked what gave me the right to be sure. I answered: 47 matches, 214 diagrams, 31 hours of tape. The piece correctly predicted 3 of Levante's next 4 matches. From then on, I set myself a rule I have never broken: every conclusion must stand on a number.

In 2026, I left my assistant-coach seat to become an independent tactical analyst in Valencia. It was the moment the new sports-media wave exploded, when everyone began talking about xG, about PPDA, about metrics that had previously lived only in club analysis rooms. But most of the people talking about them were not actually reading them. They used numbers as badges to show off knowledge, not as questions to dig deeper. I decided to do the opposite: start from raw data, build my own archive, then let the match tell its own story.

Following Levante across an entire season was not a glamorous project. No one pays you to watch a mid-table team defend. But that is exactly why it suited me. When you follow a small club across 47 matches, you are not distracted by the glittering moments of superstars. You are left with structure, habit, and repeated errors. And repeated errors, by definition, are things that can be measured, predicted and fixed.

I built my database by hand. Every set piece was logged, tagged, classified by zone, by timing, by the player taking it. After 47 matches, I had a picture no Levante coach had the time to draw. Not because they were poor. But because within a football club's week, the time to look back at yourself systematically barely exists. There is a match on Saturday, then one on Wednesday, then Saturday again. Between those are recovery, travel, meetings, training. The truth is that insiders are often the ones who see best, but only when they have the time to see.

They did not have that time. I did.

What I found at Levante was not in the 68% figure. It was in this: when you break a team down into thousands of individual actions and rearrange them, you begin to see what the naked eye misses in 90 chaotic minutes. Football is a system, and a system always leaves traces, only the traces rarely sit where we are looking.

Levante's left flank did not collapse because one full-back was weak. It collapsed because of a chain of tactical decisions. Levante played with an attacking left-back to create width, a left-leaning central midfielder to support him, and a left centre-back slower than his partner. When the ball was lost in midfield, the space behind the left-back opened for two or three seconds, enough for a well-coached side to exploit. The issue was not who was at fault. The issue was a design that placed three players in a situation where none of them could cover for the others in time.

The nine points Levante dropped from corners did not come from poor aerial defending. They came from an opponent movement pattern Levante never adjusted to. The opponent pulled a striker away from goal to drag out a centre-back, pushed a midfielder near the front post to block the keeper's view, then delivered the ball into the zone between the near post and the penalty spot. Three goals came from that same pattern. Not two. Not four. Three. And in all three, the Levante defenders' positions almost matched to the metre.

That is when I understood something about the nature of data. Data does not lie, but it does not tell the story by itself either. The 68% figure is only the door. Behind it is a story about how a team repeats the same mistake without anyone noticing in time, because each individual match looks different. Only when you stack 47 matches on top of each other does the pattern emerge. And once it emerges, it asks another question: if this error repeats, why did no one fix it all season?

The answer lies in the structure of a football week. After every defeat, the natural reaction is to fix something immediately, something tangible, something visible in the next training session. But Levante's blind spot was not visible. It was a gap created by the team's own design. Fixing it required changing how the team attacked, not just how it defended. And changing how you attack is far more expensive than changing one person.

Two years later, on a completely different stage, I met the same lesson in another form. World Cup 2026, round of 16, Spain against Russia. Spain completed 1,029 passes and held 74% possession. Reading the stats sheet, it was a dominant display. But when I mapped their 47 attacking sequences, a different picture appeared: 82% of the passes were lateral circulation in front of the box, creating no penetrating angle. They had only 8 shots on target.

That is the definition of false possession. The ball was at their feet, but the match was not in their hands. Every lateral pass in front of the box is a refusal: a refusal to take risks, to penetrate, to accept danger in order to create a real chance. We once believed in possession, until the ball was no longer at our feet. Spain controlled the ball until the shootout arrived, and there the only thing they controlled was their own disappointment.

What is striking is that Spain did not lack talent. They had players like Isco, Andrés Iniesta, David Silva, men who could unlock any defence in a single moment. But their system did not give those moments a chance to happen. The ball moved too safely, too predictably. A defence that has parked the bus does not fear lateral passes. It only fears penetrating passes. And Spain, in that match, almost refused to play them.

There is a detail I never forget. In the 80th minute, with the score still 1-1, Spain had the ball in the opponent's half with six players against four Russian defenders. That was the chance to penetrate. But the ball was played wide, then back to midfield, then to the other flank. Three passes, no breakthrough. Russia did not need to do anything. They just had to stand still and wait. In football, sometimes the strongest opponent is not the one who attacks you, but the one who is patient enough to let you attack yourself.

When I presented this argument live before two million viewers, the reaction was not gentle. Many said a woman does not understand tactics. I did not argue. I simply showed the 47 sequences, the 1,029 passes, the 8 shots on target and the heat map of touch points. My data was validated shortly after. But what I remember most is not the validation. What I remember most is the feeling that I was right because I had prepared, not because I had shouted loudest.\nThen came 2026. The pandemic forced football to stop, then to return in empty stadiums. It was a laboratory no one could have planned, and I decided to use it. I reviewed 63 post-lockdown La Liga matches and compared them with 63 pre-pandemic matches. Same league, almost the same teams, the same rules. The only variable changed was the presence of a crowd.

The result made even me pause. Successful pressing fell 12%. Goals from fast counter-attacks rose 18%. The average high line of home teams dropped by 4 metres. Home advantage, long treated as an immutable law of football, almost vanished when 40,000 spectators were no longer there to pressure the referee and the players' minds.

That 4-metre figure is my favourite. It does not speak of spirit. It speaks of behaviour. Without a crowd, home teams pushed their defensive line high less often, pressed less, and accepted risk less. They became more cautious. An empty stadium does not erase the match; it strips away the excuses. For years, people spoke of home spirit, of the pressure of the stands, of the twelfth man. When the stands were empty, those words disappeared, and what remained were pure behavioural numbers.

What is interesting is that not every team reacted the same way. Teams with well-drilled pressing systems, teams that treat pressing as a principle rather than an option, were affected less. Meanwhile, teams that relied on inspiration and crowd pressure to press declined sharply. The empty stadium did not create a new league. It merely exposed the difference between teams with a real system and teams with only slogans.

Three weeks after I published a 12-page report, a La Liga assistant coach cited it in an official press conference. I do not know whether he knew who the author was. It does not matter. What matters is that an independent analysis, built outside the system, entered the system itself.

If you put these three stories together, Levante 2026, Spain 2026, and the crowdless season of 2026, you will see a common thread. All three show that behaviour on the pitch is not the result of pure will. It is the result of conditions: squad design, match context, playing environment. When conditions change, behaviour changes, and it changes in measurable ways.

That is why I never begin an analysis with the question of which team is stronger. I begin with the question of which conditions are shaping the style of play. Tactics are not a diagram; they are how a team responds to chaos. A diagram is only the starting point. What truly decides a match is how eleven people react when the plan collapses, when the ball is lost in the wrong place, when the referee awards a controversial free kick, when a centre-back must leave the pitch injured in the thirtieth minute.

At Levante, the chaos came from losing the ball in midfield with the full-back pushed high. The team's default response was to drop back, but dropping back did not fill the gap that had opened. At Spain 2026, the chaos came from being unable to break through Russia's concrete defence. The default response was more lateral passing, which meant slowing themselves down. In the crowdless season, the chaos came from losing outside pressure. The default response was to be more cautious, safer, less ambitious.

In all three cases, the default response was the wrong response. And it was wrong systematically, because it was pre-programmed into how the team was built. A system that functions when the opponent is in chaos is what truly needs coaching. Every team knows how to play when everything goes to plan. Very few teams know how to play when the plan collapses, and that is precisely when matches are decided.

There is a paradox I want to place on the table. We praise teams that defend well, yet we measure them by attacking metrics. We praise teams that attack beautifully, yet we rank them by defensive points. Levante in 2026-17 is one example. They did not lose because they failed to score enough. They lost because they conceded along a predictable pattern. But in press conferences, the question was always why the attack did not score, never why the defence conceded the same way. Media attention and data truth often run in opposite directions.

This leads me to a remark on how we read the table. The table is a photograph, not a film. It tells you where a team is, but not where it is going. A team can sit eighth on a lucky run while its playing pattern is worsening. Another can sit twelfth on a cruel run while its pattern is improving. If you only look at the table, you will sell the second and buy the first. Process data, not results data, is what tells you what is about to happen.

Of course, this is where I must say something my followers may not want to hear. I too have fallen into the traps I now warn others about, more than once.

The first trap is believing data can replace the eye. It cannot. Data tells you what happened, but it does not always tell you why. The 68% figure at Levante said the goals came from the left, but you had to watch the tape to understand that the cause was a three-man structure, not an individual. Had I read only the stats sheet, I might have proposed replacing the left-back, a wrong solution to a right problem.

The second trap is using data to defend a conclusion you already believe. This is the analyst's greatest temptation. When you have spent 31 hours on tape, you want to be right. You begin selecting the numbers that support you and ignoring those that do not. I learned to guard against this with a simple question: which data would make me change my mind? If there is no answer, I am not analysing, I am merely decorating a bias.

The third trap is confusing contrarianism with counter-argument. Not every against-the-grain opinion is correct. Sometimes the consensus is right, and the contrarian is merely trying to look clever. The difference is this: a real counter-argument stands on data and logic, while a fake one stands only on a desire for attention. I have seen too many analyses of the latter kind, and I do not want to become part of it.

There is one more thing I must admit. My findings, whether at Levante, at the World Cup or in the crowdless season, all came from small samples. 47 matches, 63 matches, 47 sequences. These samples are enough to point to a pattern, but not enough to assert a law. Good data does not answer questions, it teaches us to ask better ones. And the best question my data raises is not what is certainly true, but what deserves a closer look in the next match.

This is also why I always tell young editors never to begin an analysis with a conclusion. Begin with an overlooked detail. An action in the eighty-third minute. A position two metres out of place. A backward pass instead of a penetrating one. These small details, placed side by side, form a story that a grand statement could never create. An empty stadium is still loud enough, if we know how to listen to every touch of the ball.

So, as this season unfolds and the table begins to take shape, I will not look at the teams that are winning. I will look at the teams repeating the same mistake without realising it. I will look for flanks exploited three times the same way, for corners defended with the same wrong pattern, for lines pushing four metres less high without a crowd, or with a crowd but having lost belief. I will look for teams holding 70% possession with no shot on target in the second half. I will look for teams winning three matches in a row through last-minute goals, because that run is beautiful but not sustainable.

The ball is only a variable; how it moves is the message. And that message, this season, may be sitting in a flank no one has bothered to look at. The question for you is not where your team stands in the table. It is this: if you stacked all their matches on top of each other, what pattern would you see, and do you have the courage to look at it?