The Annual Season and the Forgotten Signals: When PPDA Leads the Table

There is a moment in the annual league season when the table becomes meaningless to an analyst. It is not when the league leaders pull far ahead of second place, nor when the relegation battle is settled. It is the period — usually falling in the middle of the season — when a team still sits in the top four, still wins regularly, still draws media praise, but whose PPDA has been quietly rising for seven consecutive rounds. No one notices. The points column still looks fine. But the data has been speaking long before the first newspaper asks a question.

I have tracked this kind of signal since 2026, when I was a 19-year-old journalism student with a World Cup prediction model built on xG and xA from five European leagues across three consecutive seasons. The model gave Germany a 78% chance of reaching the semifinals. Germany crashed out in the group stage after losing to South Korea. Since then, I have learned something that no textbook teaches: the table is the result of the past, but pressing metrics are the forecast of the future.

The annual league season has a rhythm that cup competitions never have. It is long, repetitive, and allows variables to accumulate to the point of exposing their true nature. A team can fool viewers over three or four matches. But over thirty matches, the data has nowhere to hide. And during this window, there are three types of signals the media routinely ignores: the decline of PPDA, the volatility of running distance, and the quiet collapse of home advantage. Each signal is a story. And each story shares one conclusion: when the model fails, the data starts telling the truth.

To understand why PPDA matters so much, we need to rebuild the context. PPDA — Passes Per Defensive Action — measures the number of opponent passes before the defending team intervenes. The lower the PPDA, the more aggressively a team presses, the more actively it cuts the opponent's rhythm. The higher the PPDA, the more a team waits, the more it cedes territory. This is not an outcome metric. It is an intention metric. And precisely because it measures intention, it reflects physical, mental, and structural states in ways goals scored never can.

In an annual league season, a team's PPDA rarely stays still. It fluctuates with fixtures, injuries, and psychological pressure. But it has one rule: a team that presses well in the early season will begin to raise its PPDA from around round fifteen, unless it has enough squad depth to rotate. This is a purely physical variable. Germany 2026 is the clearest example — not because they suddenly lost form, but because my model at the time never assigned a physical variable to the equation. I dismissed it because I believed it could not be measured. Wrong. Running distance can be measured. Minutes played can be measured. Recovery time between matches can be measured. I simply did not want to believe that a non-data variable could decide.

A similar story repeats at club level. Take a typical team of the annual season: they start with an average PPDA of 8.5 over the first ten rounds. By round twenty, that number has risen to 10.2. By round thirty, it reaches 11.8. On the table, this team might still be sitting fifth. But on the pitch, they have already become a different team — slower, less controlled, more reliant on counterattacks. Their xG begins to fall. Their points per match begin to slide. And by the time the media notices, they have dropped to twelfth.

What is interesting is that this collapse is uneven. Some teams maintain a low PPDA all season, but those are usually teams with two or three alternatives for every pressing position. Some teams are forced to raise their PPDA but still maintain results, because they shift to an effective low block. And some teams raise their PPDA without changing their defensive structure — those are the teams that will collapse. PPDA is the signature, running distance is the confession.

Running distance provides the evidence that PPDA only hints at. Two teams with the same PPDA can have completely different physical profiles. Team A presses by moving together, running distance distributed evenly, no individual overloaded. Team B presses by forcing individuals to cover, running distance concentrated in two or three players. Team A survives the season. Team B breaks in the decisive stretch. This is why I always split a team's total running distance into three layers: distance run without the ball, distance run with the ball, and high-speed running distance. Only by looking at all three together can an analyst see the real story.

In the annual league season, running distance reveals something the stat sheet never shows: psychological fatigue. A team that runs more than its opponent in the second half is not always the team that wants it more. Sometimes it is the team that has lost control of the match and must compensate with unstructured effort. A spike in second-half running distance, especially for a favorite, is often the sign of a team flailing — not of a team fighting. The media calls it "fighting spirit." I call it "measured bewilderment."

But if PPDA and running distance are signals of fitness and structure, home advantage is a signal of context. And this is where the story becomes most interesting, because no signal is more abused than home advantage in everyday football analysis.

In 2026, when stadiums were empty due to the pandemic, I collected data from nine rounds of the Bundesliga after football returned. Home win rate fell from 44.2% in the 2026-19 season to 36.7%. Average goals per match fell from 3.1 to 2.8. This was a natural experiment no analysis department could stage: remove the crowd, hold every other variable constant, and measure where the home advantage actually comes from. The result showed that home advantage is not sacred ground, but a frozen variable — and when the context changes, the variable melts with it.

Imagine this in the current annual season. Every prediction model — from betting companies to television pundits — assigns a fixed home advantage coefficient to matches. This coefficient typically ranges from 0.3 to 0.5 goals. But that coefficient was built on data from seasons with full stadiums. When a team moves to a temporary venue, when a team loses its crowd because tickets are expensive or results are poor, that coefficient becomes outdated immediately. The problem is that models do not update match by match. They update season by season. And that is the blind spot.

In the annual league season, home advantage needs to be separated into at least four variables: crowd (number and intensity of support), pitch (familiar or not), travel (opponent coming far or near), and refereeing (measurable home bias or not). When these four are collapsed into one, the analyst has created a magic number with no predictive value. Worse, they have created a belief. And that belief affects how teams take the pitch — favorites sometimes play more cautiously away from home because they believe in their "advantage," while underdogs play more freely because they have nothing to lose.

This is where I want to speak about the value of placing data in context. A number divorced from the match, the timing, the lineup, and the fitness state is just noise. I once concluded that Italy would control the game against Belgium in the Euro 2026 quarterfinal, based on Italy's average PPDA of 8.2 and Belgium running 17% less than in previous matches. Italy won 2-1. But I do not tell this story to praise myself. I tell it to point out that the success came from combining three data sources — injuries, fixtures, advanced metrics — not from a single number. If I had looked only at PPDA, I could have been wrong. Data does not get emotional, but it remembers everything the press forgets — and the analyst's job is to remind us of what has been forgotten.

In the annual league season, there is one phenomenon I track very closely: the cumulative effect of a congested calendar. Teams in European competitions often play three matches in seven days in the middle of the season. Teams with good squad depth can rotate. Those without must choose: play the strongest lineup and accept fatigue, or rotate and accept dropped points. This is where fitness data and results data diverge most clearly. A team can win four out of five matches after a congested run — but if their PPDA rises from 8.5 to 10.5 during that stretch, those four wins are the result of weaker opponents, not of a resilient team.

This is where we enter the contrarian section. Because everything I have just said can be reversed if the reader misunderstands the relationship between data and results.

Correlation is not causation. This is the mantra every analyst must carve into memory. A team with rising PPDA can collapse. But another team with rising PPDA can win the title — because they deliberately changed tactics to preserve energy for the knockout rounds, or because opponents grew used to their style and they had to adapt. In both cases, rising PPDA is not the cause of collapse. It is a symptom of a broader process the analyst must decode.

This is why I always tell colleagues that I believe in variance more than I believe in champions. A champion is whoever wins at the end. Variance is the story of how they got there, and of how they might not have gotten there if a few variables had shifted. A champion with high xG and stable PPDA is a sustainable champion. A champion with high xG but wild PPDA swings is a lucky champion — and the following season they will pay for it.

There is a paradox I have observed many times in my career: the longer I track data, the less I trust definitive conclusions. In 2026, at 23, I was responsible for tracking Enzo Fernández's transfer from Benfica to Chelsea for 121 million euros. I used World Cup data — 82% passing accuracy, 14 successful tackles — to build a valuation report. But that transfer also depended on agents, payment terms, and Chelsea's urgency. Data does not reflect those. Data explains the past, it does not predict the future. This is the limit any honest analyst must acknowledge.

And it is precisely this acknowledgment of limits that separates a data analyst from a prediction salesman. The prediction salesman tells you they know what will happen. The data analyst tells you they know what happened, in which context, and which variables could change the outcome. The difference sounds small. But it is the difference between science and superstition.

In the annual league season, superstition takes many forms. There is superstition about "big-club mentality." There is superstition about "head-to-head tradition." There is superstition about a "destined manager." Each superstition has a kernel of truth — big-club mentality exists in some cases, head-to-head tradition exists in some periods, the fit between a coach and a club exists in some structures. But when these kernels are abstracted into universal laws, they become noise. And noise in football analysis harms not only the analyst. It harms the fans, who make decisions on bad information.

I return to the Bundesliga 2026 story once more because it is the perfect example of how old data becomes meaningless when context changes. During those nine rounds, it was not only the home win rate that fell. Yellow cards for away teams fell too. Corners for home teams fell too. Controversial referee decisions fell too. This shows that home advantage has a component coming from the crowd — not from the home team being familiar with the pitch, but from the crowd pressuring referees and creating atmosphere for players. When the crowd vanished, that component vanished with it. And the models did not update in time.

In the current annual season, we are witnessing a similar but subtler phenomenon. Some clubs have moved to temporary venues while their own grounds are renovated. Some clubs still have crowds but no longer generate the same pressure because tickets are expensive or results are poor. Some clubs play in neutral cities for logistical reasons. Each case weakens the home-advantage coefficient, yet the models still assign it mechanically. And bettors, analysts, fans — all are led by a number that is already dead.

This is why I write. Not to provide predictions. But to remind that every number needs a context, every conclusion needs a limit, and every model needs a confession about what it does not know. Germany 2026 was a gift, because it proved that a model also needs to fail in order to grow. If Germany had won that year, I might have kept believing in my model and ignoring the non-data variables. That failure — bitter as it was — taught me the most important lesson in this profession.

So what is the signal for the next round? In the annual league season, there are three things I will watch closely in the coming weeks. First, the PPDA divergence between teams in the top six and teams outside it. If the PPDA gap between these two groups widens, that is a sign that the strong teams are beginning to separate physically, not just on points. Second, the minutes played by key players at clubs in European competitions. If a team has three players past the 2,500-minute mark before round twenty-five, that is a red alert. Third, the home win rate in matches between top-eight teams. If this rate is unusually low, that is a signal that home advantage no longer plays a decisive role at the elite level — and favorites need to adjust their approach.

But I do not want to end with a list. I want to end with a thought. In eleven years of observing this industry, I have seen many models rise and fall. Models based on xG. Models based on pressing. Models based on passing networks. Models based on machine learning with millions of parameters. Each model promised a deeper look into the game. And each model failed at some point — against a team that defied the rules, against an anomalous season, against a variable that had not yet been named.

What I have learned is not which model is right. It is which model is honest. An honest model knows its limits, knows that data is a foundation rather than a truth, knows that context can overturn every conclusion. Such a model does not give you answers. It gives you questions. And in football — as in life — the right question is always worth more than the wrong answer.

In this annual season, when you look at the table and see your favorite team sitting third after twenty rounds, ask yourself: where is their PPDA heading? Where is their running distance concentrated? And does their home advantage still truly exist? Those three questions will not give you a prediction. But they will give you a way of seeing. And in a long season, the right way of seeing is worth more than any number.

I still watch matches with a notebook and a spreadsheet beside me. Not to record results — I remember results. But to record what the table does not say. Because after all, data does not lie. Only the person reading it can lie to themselves.