Defense, Data and Trophies: Decoding the Real Football Behind the Scoreboard

The regular season is entering the stretch where the table starts lying more than it tells the truth. I sat down with fourteen matches I had rewound and rewatched over the past two weeks, taking notes on every duel, every moment a team chose to concede possession, every instant a midfielder dropped deep without anyone noticing. One detail made me pause longer than any other: over their last three matches, the PPDA of the team sitting second in the table had risen from 8.4 to 12.7, meaning they were pressing far less, accepting a deeper block, and still winning. Nobody writes about that. People write about a striker's thirtieth goal, about a spectacular save on television, about the way a star celebrates in front of the stands. But the thing that keeps a team standing through thirty-eight rounds lies elsewhere, in a blind spot the cameras rarely visit.

I have watched football through numbers for twenty-seven years, five of them tied to the Asian market, and the biggest lesson I learned was not how to calculate, but how to doubt the very numbers I trust most. A number is only the beginning; verification is the destination. A beautiful metric can be born from a meaningless match, an impressive statistic can be a mere consequence of a team falling behind and pushing forward, a high possession rate can sometimes be a sign of deadlock. The regular season is the harshest environment for separating real data from data dressed up for display.

This piece is not meant to praise one tactical school or to tear down another. It aims to dissect a question that seems simple: when every club already has data, what still creates a difference? The answer, from what I have observed, lies in three places the media usually ignores — organised defending, the tempo control of invisible players, and the ability to read opponents in the moments when the tactical diagram no longer says anything.

Context: When every club already knows how to read data

There is a paradox in modern football that I call the "paradox of universal data." Fifteen years ago, using an xG model, a RAPM model or progress metrics to scout was an exclusive competitive advantage. The club that moved first would find undervalued players, buy cheap and sell high, and build a squad superior to its budget. But by the current season, nearly every club in the top leagues has its own analytics department, hires specialists, buys expensive data packages from global providers. The exclusive advantage has vanished, and the game has changed entirely.

When everyone knows that a midfielder with pressing success in Europe's top 1% is a valuable asset, that player's price is pushed to a level no longer reasonable. When everyone knows a goalkeeper with good distribution is an inevitable trend, the transfer fees of these goalkeepers soar, even when their fundamental skills — reflexes, positioning, command of the back line — are declining. This is the direct consequence of data becoming a common language: it no longer helps you find hidden value, it only helps you pay more for what others can also see.

In that context, value genuinely returns to qualities that are hard to measure. I am not saying data becomes useless — quite the opposite. But the centre of advantage has shifted from owning data to interpreting data correctly. And correct interpretation demands something a computer lacks: an understanding of context, of historical precedent, of the psychology of a team pushed into a corner during the run-in.

I followed the matches of a mid-table club throughout the first half of the season, and what caught my attention was not the results but the way they changed their pressing structure across phases. Early in the season they pressed high, an average PPDA of 7.5, winning the ball in the opponent's half, and winning open games. By mid-season, when the schedule thickened and the squad thinned through injuries, they began dropping their pressing line, PPDA rising to 11.2, and still maintained a good points rate. This is not a compromise; it is adaptation. This is the kind of change that raw statistical summaries usually miss, because it requires placing the number in the right phase of the season.

The same happens at national level. Looking back at World Cup history, I noticed a pattern that proactively defensive teams often violate: they build a system based on conceding possession, yet fail to prepare for the possibility that the opponent also accepts conceding it. When both teams sit back, the match becomes a battle of set moments, and the better-organised defensive side usually wins — but the side with the more mobile midfield usually has more chances to create those moments. Tactics do not live on the diagram; they live in how you read your opponent.

Core: Decoding the real football

Goalkeepers and the myth of distribution

Let us start with the most sanctified position of the past decade. Over roughly ten years, a goalkeeper's distribution has become a leading selection criterion, sometimes surpassing basic goalkeeping skill. Clubs pay record fees for goalkeepers who can play with their feet, join the circulation, and launch attacks from the back. In theory this is sound: if you control more of the ball, you create more chances, and the goalkeeper is the first link in that chain of control.

But there is a problem that data models often conceal. When I reviewed the goals conceded by several goalkeepers rated highly for distribution, I found a worrying pattern: the very overconfidence in playing with the feet became the source of serious errors. A careless short pass inside the box, an elaborate touch under pressure, a quick distribution when the defence was not ready — these mistakes do not appear in passing-accuracy metrics, because they are rare exceptions. But in elite football, where a single goal can decide a season, rare exceptions are the most expensive thing of all.

I once followed a goalkeeper through an entire season, logging every situation in which he distributed under high pressure. The result showed something interesting: in matches where opponents pressed high, his success rate fell sharply, but the number of goals conceded directly from personal errors rose exponentially. In other words, his distribution was an asset under ideal conditions and a risk under harsh ones. The problem is that valuation models usually average the two, thereby underpricing tail risk — situations that are rare but severe in consequence.

This is not an argument against modern football. It is an argument for a more nuanced reading of data. Defence is what people despise, until it lifts the trophy. A goalkeeper with excellent reflexes, good positioning, effective command of the back line and awareness of his limits in playing with his feet can deliver more points than a goalkeeper with flashy distribution who errs in decisive moments. But the transfer market does not price it that way, because the market is driven by what is easy to see.

I once erred in the opposite direction. Years ago, I undervalued a goalkeeper simply because he did not stand out in distribution metrics. After rewatching all the footage, I realised he had concealed the defence's weaknesses through such excellent reading and positioning that the defenders looked better than they were. That is the kind of value no metric fully captures, and that is why I always add a "data limits" section at the end of every analysis.

Running metrics and the trap of pretty numbers

Let us move to another equally abused metric. Distance covered and sprint counts have been packaged as measures of effort, as the moral indicators of football. Players who run a lot are praised as dedicated; players who run little are suspected of laziness. Heat maps are printed on broadcasts, distance figures compared across matches, and fans form impressions of players based on those numbers.

But here is the core problem: running a lot does not mean running effectively. A player can cover thirteen kilometres in a match and have nearly all of that distance be meaningless — chasing a ball already passed away, chasing opponents in risk-free phases, running to cover space after already losing position. Conversely, a defensive midfielder may run only ten kilometres but every metre has purpose, every metre part of a system that cuts off dangerous passing lanes.

I spent time analysing the running metrics of several clubs this season and noticed something remarkable: the clubs with the highest average distance covered are usually not the best defensive sides. In fact, there is a mild inverse correlation between distance covered and defensive efficiency in certain phases. This can be explained by the concept of "wasted running": when a team loses structure, players must run more to compensate, and that very running makes them lose position further, creating a spiral. Good defensive sides run less because they are already in the right place before the ball arrives.

This leads to an important consequence for the transfer market. High running metrics are often used to justify a player's high price, especially young midfielders. But if that metric reflects a system's chaos rather than individual quality, paying a high price for it is a strategic mistake. I have witnessed such deals, and it usually takes two to three seasons for the truth to surface — when that player moves to a club with better structure and his metrics suddenly drop, yet the team's efficiency rises.

I always remember a principle I set for myself after years of analysis: never judge a player by a single metric, and never judge a metric without asking what context produced it. Every media wave mixes trash and gold; our job is to sift. In an era where every number can be cited and shared, the ability to sift becomes more important than the ability to calculate.

Invisible midfielders and value that never reaches the highlights

There is a type of player I respect most, and also the type the media ignores most: defensive midfielders who do their job so well that no one notices. They do not score, do not assist, do not produce touches that make the stands rise. They simply are in the right place, at the right time, cutting passing lanes, shielding the back line, keeping the match tempo within control.

Their value lies in what does not happen. A dangerous counter-attack cut off at the root will never appear in the summary, never become a highlight. A dangerous pass read in advance and intercepted will not be recorded as a spectacular tackle. These players often have modest statistics — few goals, few assists, passing rates not especially impressive — but when they are absent, the whole team collapses.

I verified this by analysing the records of several clubs in matches with and without their key defensive midfielder. The results were clear: when these players were absent, the average goals conceded rose significantly, and more importantly, the chances opponents created from counter-attacks soared. This is a contribution that cannot be measured by goals or assists, only by advanced defensive metrics — and even those capture only a portion.

Interestingly, the transfer market sometimes recognises this value, but in a distorted way. Good defensive midfielders are often bought cheap when young, and only once they have proven their importance over several seasons does their price soar. This is one of the points where my valuation model often diverges from the market: I rate these players earlier, based on progress metrics rather than results.

But there is a downside that must be acknowledged. Over-focusing on defence can lead to a trap I once fell into: undervaluing the importance of attacking efficiency. A team that only defends well but cannot score will not win the big matches. The trophy is not awarded to the prettiest team, but to the team that errs least. But "errs least" does not mean "never attacks" — it means knowing when to attack and when to control. This is the balance I always try to reflect in my analyses, even though I myself lean toward defence.

The transfer market and the loophole of invisible fees

Now let us talk about money. For years I have built a database of transfers, focused not only on transfer fees but on contract structures, release clauses, and the hidden fees the media usually overlooks. And what I found is that one of the biggest problems of the modern market lies in the least-noticed deals: free-agent signings.

When a club signs a player on a free transfer, no transfer fee is recorded. This means the deal escapes the scrutiny of financial fair play regulations, which mainly focus on transfer fees and contract amortisation. But in reality, the club still pays a large sum — as signing bonuses, agent fees, and higher wages to compensate for the player costing no transfer fee. These can equal a significant transfer fee, yet they do not appear in financial reports in the same way.

This is a loophole I believe is more toxic than traditional transfer fees, because it circumvents the core oversight of FFP. I have analysed several free-agent deals and found that the true total cost — including signing bonuses, agent commissions, and total contract value — is often far higher than what is reported. In some cases, this total exceeds what a club would pay for a player with an equivalent transfer fee.

This raises a question about market transparency. When clubs can hide the true cost of a deal by structuring it as a free transfer, financial fair play regulations lose part of their effectiveness. And when regulations lose effectiveness, the advantage returns to the clubs with the highest spending power, the clubs that can pay enormous signing bonuses without being bound by amortisation limits.

I once wrote about such a deal and was criticised for digging into unnecessary details. But I believe those details are exactly where the truth lies. History does not repeat, but precedent always knocks at the door of a crisis. If we do not understand how money truly moves in football, we cannot understand why some clubs can spend far beyond their means without punishment, and why others fall into financial crisis.

Historical precedent and how to read a crisis

Speaking of crisis, there is a lesson I learned from my own experience. In 2026, when the pandemic paralysed the entire global sports system, I did not join the wave of optimistic predictions about the return of the leagues. Instead, I dug into data from previous disruptions — the 2026 NBA lockout, the same year's NFL lockout — and analysed the average break length, its effect on playing tempo, and injury rates after the return.

The data showed a clear pattern: after long breaks, teams with many older players usually suffered heavier injury losses. This is physiologically sound — older players need more time to recover fitness and maintain match sharpness, and a long break disrupts that rhythm. I published a prediction that teams dependent on stars over thirty-two would face greater difficulty in the season after the return.

When one of those teams won a title under centralised conditions, many laughed at me. But the following season, when one of their oldest stars was injured and the team was eliminated early, my judgement was proven right in trend, though wrong in timing. This is an important lesson on reading historical precedent: it does not tell you exactly what will happen, only what is more likely to happen. Crisis does not ask whether you are ready; it only asks whether you have seen it before.

I learned that citing a historical precedent is only valuable when I can point to at least three substantive similarities between the current situation and the past one. If there is only one similarity — for example, both are crises — then it is not a precedent but a superficial resemblance. I made this mistake in the past, comparing situations that resembled each other only on the surface, and the result was flawed judgement. Now I always self-check before drawing any historical comparison.

The contrarian angle: when progress data deceives

At this point I must address a downside of the very method I use. In recent years, progress data — especially xG and related metrics — has become the favoured tool of analysts, myself included. The reason is clear: the final result of a football match is heavily influenced by luck, while progress data tries to measure the quality of chances created and prevented. Over time, teams with good progress metrics usually achieve better results, even if short-term results differ.

But there is a trap I have seen many analysts fall into: treating progress data as supreme truth and ignoring what the match actually shows. This is a form of intellectual arrogance. When a team keeps winning matches with a low xG, there are two explanations: either they are lucky and will soon be punished, or they possess a quality the xG model does not capture. The second explanation is often ignored, but sometimes it is the truth.

I followed such a team through a season, and what I realised was that they had a special ability to control match tempo — they did not try to create many chances, but tried to control when and how the match unfolded. They accepted letting opponents hold the ball in harmless areas, they waited for mistakes, and they maximised the decisive moments. The xG model does not capture this strategy, because it measures chance quality, not match control.

This is why I always combine data analysis with watching footage. Highlights make idols, but consistency makes legends. No model replaces sitting down to watch a match and understanding why things happened as they did. Data tells you what happened; footage tells you why.

I must also acknowledge a limit of my own. In 2026, when FIFA expanded the Club World Cup to thirty-two teams and staged it in the United States, I publicly doubted the new format, arguing it diluted the competition's quality. When the newsroom assigned me to cover the event, I rigidly applied my old data model and failed to predict the group-stage results. I had not anticipated that five substitutions per match would completely change the tempo and tactical approach.

After a match in which a big club lost to a lower-rated opponent, I sat down with a young colleague and asked him to explain how xG algorithms weighted by playing time worked. It was a humbling lesson. I updated my system, added squad management and depth factors to my analyses, and wrote a series on the fatigue of stars in densely scheduled competitions. Since then, I always add a "data limits" section at the end of every piece and actively collaborate with young analysts to renew my methods.

What truly creates a difference in the run-in

Back to the current regular season. When all clubs already have data, already have analysts, already know how to optimise metrics, what still creates a difference in the run-in? From my observation, three factors.

First, adaptability. The clubs that succeed in the run-in are not those with the best system, but those able to adjust their system to circumstances. When injuries strike, when the schedule thickens, when opponents change their approach, the flexible club survives. I have seen rigid clubs with a single system collapse when conditions changed, and flexible clubs overcome difficulty.

Second, squad depth. Over a long season, no club can field its strongest eleven in every match. Clubs with good depth can rotate without losing quality, while clubs dependent on a few stars struggle when those stars tire or get injured. This is a factor data models often underrate, because it concerns less-noticed players.

Third, and perhaps most importantly, the ability to read opponents in decisive moments. This is the hardest factor to measure and the hardest to copy. It lies in no tactical diagram, appears in no metric. It is the understanding of an opponent's weaknesses, of their psychology in tense moments, of how they react when trailing or leading. Tactics do not live on the diagram; they live in how you read your opponent.

I followed the matches of a title-contending club and noticed that their coach frequently changed his approach based on the specific opponent, sometimes within the first fifteen minutes of a match. He observed how the opponent set up, identified the weakness, and adjusted immediately. This is a skill that cannot be programmed, and it is often the decisive factor between success and failure in the big matches.

Takeaway: The variables for the road ahead

As the regular season enters its decisive phase, I will watch three variables. The first is the adaptability of the leading clubs as pressure rises — whether they are flexible enough to change when needed, or will rigidly cling to what brought them success. The second is the fitness of key players, especially those over thirty, amid a dense schedule. The third is the mid-table clubs, those with nothing to lose, who can produce shocks for title contenders.

I do not know what will happen. No one does. But I know that what decides the final outcome will not lie in the flashiest numbers, but in the overlooked details — a well-timed defensive action, a sensible substitution, an invisible player doing his job without anyone noticing. Football is a game of moments, and the decisive moments usually come from places the floodlights do not reach.

If you are following this season and trying to predict the outcome, try once to look at what does not appear on the scoreboard. Try to count the counter-attacks cut off at the root, the dangerous passes read in advance, the moments a midfielder drops deep to hold the structure. That is where the truth lies. And that is where the trophies are decided.