Home Advantage Thins Out and the Pressing-Metric Race in Modern Football

In May 2026, the Bundesliga returned after nearly two months of suspension, and the stands at every stadium were empty. I sat in front of a screen in Hai Phong, reopened the data from the first 26 matchdays — the ones with crowds — and placed it beside the 9 matchdays that followed, when the grounds held only the sound of the ball and plastic seats knocking together. Same league, largely the same people, two different pictures. The home win rate fell from 55% to 43%. Yellow cards rose 22%. The away teams' PPDA — the average number of passes a side allows its opponent before winning the ball back — dropped from 11.4 to 9.8. A night in Hai Phong taught me one thing: people look at the price board, I look at the movement board.

PPDA needs explaining before it gets used as a label. The metric counts the passes a team allows its opponent inside the 60% of the pitch in front of its own goal, divided by its defensive actions. The lower the PPDA, the earlier and more aggressively that team presses. Alongside xG — expected goals, based on chance quality — PPDA is one of the two most familiar axes for reading a team's tactical intent. Newcomers to data-driven football often skip it, because xG is easier to picture: how many goals a shot from inside the box is worth. PPDA is more abstract, but it answers a question xG cannot: how long is this team willing to let its opponent hold the ball before acting?

The before-and-after comparison is how I test every conclusion. When the Bundesliga returned, I did not compare one team with another; I compared a team with itself under two different conditions. That approach removes most of the variation in squad quality, because the subject stays the same. What remains — what changes — is the environmental variable: the crowd. That is why I trust the 55%-to-43% figure more than I trust a composite league table.

Before 2026, I read football mainly through xG. That habit has a trap: a team that creates good chances can still lose, and a team that presses poorly can still win through a single moment of individual quality. In the summer of 2026, at the World Cup in Russia, I placed my faith in Germany, with 67% average possession, an xG of 2.1 and 91% passing accuracy. Germany lost their opener to Mexico and were eliminated by South Korea on 27 June. My spreadsheet was right about the numbers and wrong about the people: pitch temperature, Mexico's high press and the psychology of a defending champion sat in no column at all. That was the first time I understood that a model can be right about the past and still wrong about the present.

That shock sent me back to an old question: if xG tells only the attacking story, where does the defensive half live? PPDA is part of the answer. When away teams pressed harder in the 2026 Bundesliga — PPDA falling from 11.4 to 9.8 — they won the ball back sooner, forcing home sides into long balls and out of shape. With no stands to roar on every home counter, the home side lost part of a psychological edge that never appears in any statistics table. Germany's leading clubs were forced to adjust their squads: favouring midfielders who can escape a press and centre-backs who can pass long accurately, rather than simply picking the best duellists.

The 22% rise in yellow cards is a fact worth pondering. The referees did not change, the laws did not change, yet cards rose when the stands emptied. The most plausible explanation is that crowd noise creates pressure that makes referees more cautious toward the away team. When that pressure disappears, decisions become more even, and football becomes fairer in a strange way — but also less dramatic.

Euro 2026 was the next lesson. I predicted Belgium would win because they had the tournament's highest total xG. Italy, under Roberto Mancini, won, and the reason did not lie in attack. Italy's PPDA was just 8.7 — the lowest of the 24 teams. It meant opponents were allowed an average of only 8.7 passes before Italy closed them down. The foundation for that number was an experienced defence, with Giorgio Chiellini and Leonardo Bonucci at its centre, and goalkeeper Gianluigi Donnarumma — later voted the tournament's best player. I missed the PPDA figure because I was staring at xG. After the final, I spent three weeks rebuilding a pressing dataset across 14 major leagues and found a pattern: every European champion from 2026 to 2026 had a PPDA below 10.

What matters is the direction of the number: pressing has shifted from a tactical choice into a near-mandatory condition for winning a title. A team can win a single match on high xG, but to go the distance in a tournament it needs to recover the ball fast enough before its opponent can organise. Once, a team could sit deep, concede territory and wait. Now even sides that prefer to defend must press in an organised way for certain spells, because opponents are good enough to punish a passive defensive block.

I do not want to turn PPDA into a new dogma in place of xG. My numbers do not need applause. They need to be right — time is the referee. A metric is only worth something when it survives the test of a large sample and of context. Nine crowdless matchdays in the Bundesliga is a small sample, and I know it. A league can be shaped by a congested calendar, by injuries, by one player in form. A low PPDA can reflect a disciplined defence, or it can simply be the by-product of an opponent choosing to play long and handing over control voluntarily.

But even the before-and-after comparison has holes. The nine crowdless matchdays came at the end of the season, when many teams had nothing left to play for, when the calendar was compressed by the pandemic, and when squads had to change because of positive cases. Those three factors together can produce an effect identical to losing the crowd, without any crowd being lost. I noted this in the margin of the spreadsheet, because a tidy conclusion is usually one that has left something out.

This is where the data and the story part ways. Based on my experience following matches, a team that presses well usually has a low PPDA, but a team with a low PPDA is not necessarily pressing well. Some teams press poorly yet keep a low PPDA because they defend in a low block and opponents choose to pass sideways. Correlation is not causation. If I look at a single metric, I repeat the mistake of the summer of 2026, only with a renamed column. Same number, two readings, and the wrong reading always comes from forgetting to ask under what circumstances the data was produced.

A chart does not lie, but it does not tell the whole story. I look for the missing part. And the missing part here is emotion. With the stands empty, I realised I had failed to count one variable: emotion does not sit in a spreadsheet. The roar of a crowd does not create xG and does not lower PPDA, but it changes how a defender steps into a challenge in the 88th minute. When the Bundesliga returned to empty stands, the first thing to vanish was not technique but the invisible pressure pressing down on referees and on the legs of away players.

In the V.League, which I have followed for years, that variable is even clearer. A packed ground can turn an average team into one that is hard to beat in the first half. But when I look back over recent seasons, what carries a team through a whole campaign is not home inspiration but structure: the ability to win the ball back and reorganise after losing it. That is why I began reading V.League matches along the same two axes I use for European leagues, even though the data quality here is far thinner. That thinness is itself a fact: it reminds me that not every league lets you measure things the same way.

If the coming season runs at its current density, I will keep an eye on the PPDA of teams competing for continental places, because the side that holds a figure below 10 through the run-in is usually the side that goes far. Running alongside that is the gap between home and away win rates: when the gap narrows, a league is shifting from reliance on the crowd to reliance on the system. That is a healthy signal, even if it makes home nights a little less loud.

Data gives me a map, but the map is not the territory. I will still open the spreadsheet every morning, still update PPDA and xG, but I keep an empty row at the bottom of every table — a row for what I have not yet measured. This season will probably teach me one more variable, and I will have to rewrite myself again. Three in the morning, the market asleep. That is when the numbers are most awake.