Japan's PPDA 6.8 and the Data Gap That Broke My World Cup 2026 Model
Core answer: Mô hình dự đoán World Cup 2022 thất bại khi xếp Đức vượt qua vòng bảng vì bỏ sót chỉ số PPDA 6.8 của Nhật Bản. Nhật Bản pressing cực mạnh trong hai trận gặp Đức và Tây Ban Nha, biến áp lực phòng ngự thành lợi thế mà mô hình dựa trên xG không đo được. Key facts: - Ngày 1 tháng 12 năm 2022: Đức bị loại ở vòng bảng World Cup 2022 dù có xG tích lũy cao nhất bảng E. - Nhật Bản đạt PPDA 6.8 trong hai trận gặp Đức và Tây Ban Nha, mức áp sát khắc nghiệt nhất giải. - Năm 2018, Croatia vào chung kết nhờ chạy trung bình 112 km mỗi trận và giữ PPDA 8.2. - Năm 2020, tỉ lệ thắng sân nhà tại Bundesliga giảm từ 54% xuống 48.7% khi sân không khán giả. - Năm 2017, CLB Hà Nội thắng Quảng Nam 1-0 tại V.League với xG 2.87 so với 0.45. Source attribution: Phân tích độc lập của Bùi Cường, tổng hợp từ dữ liệu công khai World Cup 2022, V.League và Bundesliga | Cross-checked: VuaBong.vn Related Q&A: Q: PPDA là gì? A: PPDA là số đường chuyền đối thủ được phép trên mỗi hành động phòng ngự; chỉ số càng thấp thì pressing càng quyết liệt. Q: Vì sao xG không dự đoán được kết quả của Đức? A: xG chỉ đo chất lượng cơ hội, không đo áp lực phòng ngự tạo ra hoặc triệt tiêu cơ hội. Q: Bài học rút ra cho phân tích dữ liệu thể thao là gì? A: Mọi phân tích cần mục “rủi ro và khoảng trống”, vì ô dữ liệu trống nguy hiểm hơn cả số liệu sai.
On 1 December 2026, when the final whistle blew at Al Bayt Stadium, my model still had Germany in the round of 16. I was sitting in my office in Hanoi with three windows open: the live group table, the accumulated xG sheet, and the notebook I kept throughout the tournament. Germany had the highest accumulated xG in Group E. Germany dominated possession. Germany out-shot every other team in the group. Every column I collected said the same thing: this team deserved to advance.
Then Germany were eliminated. And the thing that eliminated them was not in any of my columns.
I have watched football through data for more than two decades. My job is to rebuild the truth of a match from numbers rather than from the memory of spectators. The prediction model I used at the 2026 World Cup was built on four data groups: goals, xG, possession-control metrics, and the fixture list. It was the framework I had used for years, and it had been right before.
In 2026, while working as a data editor in Hanoi, I wrote that Hanoi FC deserved to win 3-1 rather than scrape a lucky 1-0 against Quang Nam in the V.League. The evidence was in the full match: xG 2.87 against 0.45, 68 percent possession, and 14 shots inside the box. I was mocked because “football is not mathematics.” A week later, coach Chu Dinh Nghiem admitted he had rewatched the tape and adjusted his tactics based on that analysis. That night, the media called his team soulless. xG said the opposite, and I chose to trust xG.
In 2026, I travelled to Russia to report. While most of my colleagues picked Brazil or Germany, I wrote that Croatia would reach the final, because their midfield ran an average of 112 kilometres per match — the highest in the tournament — and the trio of Modrić, Rakitić and Brozović held a PPDA of 8.2, a suffocating level of pressure. Croatia did not reach the final because of luck. They reached it because their legs did not know how to stop.
Then came 2026. The pandemic closed the stadiums, and I threw myself into testing a hypothesis about home advantage. I bet that the home win rate would fall from 54 percent to below 50 percent. The Bundesliga restarted, and that rate dropped to 48.7 percent; Borussia Dortmund won only 3 of their remaining 8 home games. But my recovery model failed badly, because it did not anticipate differences in training-ground quality or in team psychology. When the stands were empty, my model collapsed. I knew I had forgotten the human factor.
That is why the gap in Qatar took me three weeks to get over. The problem was not a number I calculated wrongly, but a number I never collected: Japan's PPDA.
PPDA stands for the number of passes an opponent is allowed per defensive action. The lower the figure, the more aggressively a team presses. A mid-level pressing team usually holds a PPDA around 10 to 12. Across the two matches against Germany and Spain, Japan recorded a PPDA of 6.8. That is among the most severe pressing levels of the tournament, and it sat entirely outside the dataset I had collected before the event.
I built my model to measure the quality of chances, not the pressure that creates them. xG answers the question of how many expected goals a shot is worth. It does not answer why a team is forced to shoot from that position. When Japan forced Germany to pass sideways in midfield and then surged at exactly the right moment, Germany's passing chains broke before the ball reached their attackers. Germany still generated xG. But that xG came from isolated moments, not from the game state they wanted. Dōan and Asano came off the bench and punished precisely that moment.
This is the point I had missed for years: my model measured the result of pressure, not pressure itself. I could see the shot, but I could not see the hand that squeezed the throat behind it.
I had touched this exact lesson without recognising it. Croatia in 2026 also won with a PPDA of 8.2 and legs that did not know how to stop. But I treated it as a special case of resilience rather than promoting it into a mandatory variable in every model. I praised pressing data, then went back to trusting plain xG when I walked into Qatar.
The mistake in Qatar was not a sampling error. It was a structural gap: I left an important data field empty, and that empty field stayed silent exactly as it always would. No warning. No exclamation mark. A blank cell in a spreadsheet.
I spent the next three months rebuilding the system, integrating pressing data, distance-run data, and even the non-traditional sources I had once dismissed as secondary. Rebuilding was not the hard part. Admitting that I had been missing something was.
When I calculate xG wrongly, I can check it and fix it. When I lack PPDA, I do not know I am missing it, so I become confident without grounds. The most dangerous number in an analysis is often the number that does not exist, because it leaves no trace to be suspicious of.
I do not believe in intuition. But I believe in what intuition confirms through data. The unease I felt looking at the Group E table was a signal; I simply had no metric to turn it into evidence. And I ignored it, because I believed my spreadsheet was complete.
That is also the trap for anyone who writes with data: after a few successful reversals of the crowd's judgement, you begin to believe you see more than others. But data is never complete. It is only complete within the scope you chose to collect. Between those two things lies an entire dark zone, and most wrong predictions are born there.
There are things data cannot measure, and I have to say this plainly. The psychology of a team on a big matchday. The silence of an empty stand. The feeling of a defender who knows he is being chased every step. None of that sits in any column I built, and all of it still decides results.
Since then, every analysis of mine carries one mandatory section: risks and gaps. I list what the model cannot see — injuries, psychology, fixture congestion, and the metrics I have not yet collected. Readers need to know my limits, not only my conclusions.
Next matchday, when an empty cell appears in your spreadsheet, read it before the numbers that are already filled. A number never needs us to defend it. On the contrary, we need numbers so that we do not deceive ourselves.
Cầu thủ liên quan
Bài đề xuất
Real Madrid's Rebuild Under Pedro Martínez: Jaime Pradilla and the Pace Problem2026-09-10
Elfrid Payton and Sacramento's Backup Point Guard Slot: An Autopsy of an Incomplete Data File2026-09-17
Porzingis's Absence and the Warriors' Structural Problem: When Steve Kerr Is Forced to Forget a Name2026-10-16
Atlanta Dream reach first WNBA Finals since 2026: an 85-83 win over New York and two technical fouls hanging over Angel Reese2026-10-11
When the Data Returns Zero: Lessons From a Basketball Analytics System That Went Silent2026-09-16
Bài đề xuất
Bài đề xuất
Sergio de Larrea teaches Spanish to his Dallas teammates: the locker-room story and the data gap behind it2026-10-20
Elfrid Payton and Sacramento's Backup Point Guard Slot: An Autopsy of an Incomplete Data File2026-09-17
Hapoel Tel Aviv erase a 13-point deficit against Virtus: the second-half glass decided it2026-10-22
NBA: Five Challengers to Dethrone the New York Knicks and the Salary-Cap Test No One Can Escape2026-10-01
EuroLeague upgrades referee roster: Adding Tsimbouris and the Wroclaw preseason clinic2026-09-03
NBA 2026-27 Opening Night: LeBron James and the Philadelphia 76ers Stun Champion New York Knicks on the Road2026-10-22
