The Rest of the Scoreboard: Four Times Data Spoke Before the Result

The Rest of the Scoreboard: Four Times Data Spoke Before the Result

Opening: Long An 2026 and the gap between two columns

In July 2026, the rented room in Binh Duong had no air conditioning, and I sat in front of a laptop screen with the ceiling fan humming. I spent three full weeks manually entering the data of the first 20 rounds of the V-League for Long An FC into a spreadsheet. The average xG column sat at 2.1. The actual goals column: 0.8. The gap between those two metrics was a question almost nobody in Vietnamese football was asking at the time, because public debate was consumed by fighting spirit, foreign players, and the coaching staff. I wrote the piece, titled it "Long An: bad luck or a finishing problem?", and concluded the club only needed to keep its coaching staff to survive relegation.

Four weeks later, the club's leadership sacked the coach. Long An was relegated with 21 points. The article was shared 2,000 times across Vietnamese football communities, mostly to criticize me. That was fine. It was the first time I understood something that later became the foundation of my entire writing career: data does not lie — the listener is simply not patient enough.

The rest of the scoreboard

My job is not to retell the score. The whole country has already watched the match. The job of a data journalist is to drag the cursor to columns most people never touch: resource differentials by time marker, objective priority order, combat efficiency normalized by position. From those columns, the picture of why the result was truly decided had already existed long before the referee blew the final whistle.

The crowd looks at the score; I look at the rest of the scoreboard. Many people ask me why I do not write about marquee matches, big clubs, rising stars. The answer lies in this: everyone watches the marquee matches, but very few stay behind after the final whistle to reopen the spreadsheets. It is precisely at that moment, when the screen has gone dark, that the numbers begin to speak.

From esports to football: how I learned to count

I entered the industry in 2026, starting as an esports athlete and tournament organizer, then moving into esports media. That background taught me something school never did: every decision in competition leaves a trace, and traces are measurable. In esports, people are already used to analyzing frame by frame, pick by pick, movement beat by movement beat. When I crossed over to football, I carried that habit with me and realized most fans were still reading matches by feel.

By 2026, while a second-year student, I began logging every phase of V-League play into a personal spreadsheet. No expensive software, no analytics team behind me. Just a laptop and patience. Each matchday, I spent about twelve hours rewatching footage and entering data by hand. It was slow, but slowness was an advantage: entering each row myself forced me to see every phase, including the ones television never replays.

Croatia and the PPDA metric

In 2026, thanks to the Long An article, a football outlet invited me to contribute during the World Cup in Russia. I chose Croatia as my subject, largely because nobody else chose them. Asian public opinion at the time was focused entirely on Brazil and France. I stayed back, broke down Croatia's first 5 matches, and the metric that made me pause was PPDA — the number of opponent passes completed before being pressured.

Croatia's average PPDA across those 5 matches: 9.2. That is a very low figure, meaning opponents had almost no time on the ball. Croatia did not need to control possession to control the match; they controlled by stripping away space from others. Luka Modric and Ivan Rakitic were not the ones running the most, but they were the ones dictating the tempo of every phase. In my model, this midfield pair played the role of a valve: when needed, they slowed the tempo to draw the opponent up; when the chance came, they accelerated with a single pass into the space behind the defensive line.

I published "Croatia reach the final without needing to control the ball". Many objected, arguing Croatia was merely lucky as they repeatedly went to extra time and penalties. That argument sounds reasonable, but it ignores a fact: to reach a penalty shootout, a team must hold its shape long enough not to be eliminated earlier. Luck does not hold shape through three consecutive knockout matches against three opponents with completely different styles.

In the semifinal, Croatia beat England 2-1. The article reached 8,000 views and was shared by a European editor. That moment established my belief in a principle: data allows a writer to move ahead of popular opinion, provided the writer is patient enough to read the entire spreadsheet. I shifted from descriptive writing to systematic tactical analysis. Every piece had to pass two tests before publication: is the data strong enough, and does the conclusion truly challenge old thinking?

Lingard and the misunderstood column

In 2026, global competitions were suspended. I had been working for 8 months when my salary was cut by 30%. Instead of waiting for football to return, I used my free time to analyze the motion data of Jesse Lingard at Manchester United.

Two metrics jumped out at me. Average distance covered: 11.2 km per match — a high figure for an attacking midfielder. Goals and direct assists: 0.2 per match. A player running that much while contributing that little directly is usually labeled poor. But if you place the two metrics side by side and ask "why", the answer lies in the system, not the player.

At Manchester United at the time, Lingard was asked to hug the flank and drop deep to press, while his ability to move into the space between the lines was constrained by the squad structure. He ran a lot, but ran into zones that produced no attacking value. This is a form of systemic wasted energy, and it can only be seen when distance covered is placed next to receiving positions, not next to raw goals.

I wrote "Lingard is suffocated by an overly rigid system", predicting that if given freedom at a mid-tier club, he would explode. In 2026, Lingard scored 9 goals in 16 matches on loan at West Ham and returned to the England squad. That explosion proved my analytical model worked well even during a suspended season. Crisis does not create phenomena. It merely exposes data that was forgotten.

Morocco and the irrefutable low block

By 2026, I was an established data writer in the newsroom. Before the knockout stage of the World Cup in Qatar, while the world dissected Spain and Portugal, I pivoted to Morocco.

Two metrics decided everything. Morocco's average xGA per match: 0.3 — the lowest in the tournament. Successful tackles in the central zone: 14.2 per match. Combine the two, and you have a defensive block built not on inspiration but on structure. Sofyan Amrabat swept the central zone; Achraf Hakimi sealed the right flank; Yassine Bounou stood behind with a high save rate. I wrote a series declaring that Spain, despite 78% possession, would be powerless against Morocco's low block.

My argument was simple: when a team defends in a low block, the opponent's possession rate becomes a meaningless metric unless it is accompanied by the number of chances actually created. Spain passed a lot, but most passes occurred in front of the defensive block, not through it. Spain's shots from dangerous zones over 120 minutes stayed very low.

Many colleagues thought I was reckless. The match ended 0-0 after 120 minutes, and Morocco won on penalties. The article was internationally recognized. From that point, I abandoned the habit of using statistics as decoration and shifted to building my own prediction models for each tournament. One number is an accident. A cluster of numbers is a confession.

Method: my tracking sheets

For an analysis to hold, I build tracking sheets in three layers. The first layer is the base layer: basic metrics on possession, shots, and pass accuracy. Everyone has this layer, and therefore it offers no differentiated value.

The second layer is the positional layer: heat maps, receiving zones, average distances between lines. This layer helps distinguish a team that passes a lot because it genuinely controls play from one that passes a lot because it does not know where to send the ball.

The third layer is the time layer: the match is divided into 15-minute blocks, and each metric is computed separately for each block. This is the layer I use most, because it shows when a match was broken. Some teams win in the second half but in reality lost from the 30th minute, when the midfield lost its ability to stop counterattacks.

When these three layers align, the conclusion almost emerges on its own. When they conflict, I stop. Conflict between data layers is usually a sign that the sample is not large enough, or that the human variable has not been accounted for.

Three traps the crowd falls into

There is one thing I constantly remind myself of, and it runs counter to how most people read a data analysis.

First, correlation is not causation. This is the most common trap, and also the one a data writer is most likely to fall into under the pressure of page views. A team with high xG but few goals does not automatically mean its attack is poor. The attack may be poor; the opposing goalkeeper may be outstanding; the team may be choosing the wrong moment to shoot. A data journalist has the responsibility to distinguish those three possibilities, not to merge them into a convenient conclusion.

Second, outlier numbers are not trends. One match does not make a trend. Three matches are suspicious. A tournament is enough to write about. I see many analyses built on the basis of a single match, and I always ask myself: if that match were played again, would the column hold? If the answer is "not sure", that data is not yet strong enough to be an argument.

Third, and the most sensitive trap of all, is dehumanizing people. A spreadsheet that is technically correct can still be wrong in human terms. I once nearly wrote a piece based only on numbers while ignoring ticket-office pressure, dressing-room communication signals, or a player's injured head. The data architecture collapses if it is missing a column named "human". Before criticizing a player, check your own database.

I do not write to be agreed with. I write to be verified.

Signals for the next round

The current transfer window is underway, and the noise is greater than ever. The transfer season is a chess game where most people only see the pawns. The metrics that truly matter are not in headlines about "wages" or "transfer fees", but in release-clause structures, the buying club's remaining wage bill, and the agent's moves. Those are the columns few people bother to drag the cursor to, yet they are where the fate of a deal is decided.

I often get asked: where should one read transfer news to avoid being led by the nose? My answer is not a single source, but a filter. News from the club or agent carries more weight than news from "sources close to". A number accompanied by contract structure carries more weight than a round number. And everything that cannot be verified by documents or official statements should be sorted into the "pending" group, not the "news" group.

Data does not automatically become truth just because it is printed. It becomes truth when we can verify the source, verify the sample, and verify even what it does not say. Football does not lack stories to tell; it lacks people willing to count again.

In the coming months, I will keep updating my tracking sheets, not to predict wins and losses, but to see what the data columns already said before the market reacted. If you are drowning in rumors, the task is not to read more, but to rebuild your own credibility filter. The data is already there. What remains is to sit down and count.