When the Match Data Sheet Returns Zero

10:40 p.m. The data sheet for the match that had just ended came back at zero. Not one xG value. Not one PPDA figure. Not a single touch coordinate logged across the entire time the ball was in play. I had twenty minutes before the piece had to go live, and nothing in my hands to count. Ten years in the job had taught me to live with data that arrives late, arrives skewed, or arrives missing a column. A completely blank sheet is something else. It forces a choice that cannot be postponed: write by eye, or do not write. I chose a third route — rebuilding the sequence by hand from the match recording, and stating plainly in the piece that all positional data had been collected manually rather than through an automated system. V.League is not short of event-layer data. Every matchday, the statistics provider still logs pass counts, ball recoveries, pass completion rates. What runs thin sits at the positional layer. A tracking system that meets standard needs ten to twenty cameras, a frame rate of 25 per second, and a tagging crew working in real time. Running that for a single match in Europe costs somewhere in the low thousands of euros. For a league with a modest broadcast rights value, that line item is usually the first one cut when budgets tighten. The consequence is that a data journalist in Vietnam works with a tiered dataset. The solid tier holds the scoreline, cards, minutes played, substitutions. The middle tier holds pass counts and completion rates. The thin tier holds xG, PPDA, sprint distance, pressures — metrics that exist only when there are enough cameras and enough taggers. In 2026 I was the only female reporter in the post-match press room after SHB Da Nang against Ha Noi FC. When I asked head coach Le Huynh Duc about his side's xG of 0.4 in a 1-0 win, a male reporter cut in loudly: what would a woman know about football. I did not argue. I logged the tracking data of all 22 players and published a three-thousand-word analysis that night. When the press room laughs at xG, I know I am reading the right book, the one they have not opened. xG, put simply, is the probability that a shot becomes a goal. It is calculated from location, angle, stronger foot, the number of defenders within blocking range, goalkeeper position, then compared against tens of thousands of similar shots in the past. PPDA is the number of opponent passes allowed before the defending side makes a defensive action — the lower the figure, the higher the pressure. The value of those two metrics lies in separating process from outcome. A team can win 1-0 with an xG of 0.4 while the opponent loses with an xG of 2.1. The scoreline goes into the record books. xG goes into the capability file, and the capability file is what forecasts the next matchday. The paradox of Vietnamese football sits right here: because positional data is scarce, the habit of treating results as truth is harder to break. A team winning three straight is described as hitting form, when those three games may simply be three occasions on which opponents hit the post. In the summer of 2026 I analysed all 64 qualifying matches of the World Cup teams and noted that Croatia carried a PPDA of 8.2 — the highest pressing figure in Europe — alongside a final-third pass completion rate inside the top three. I published a prediction that Croatia would reach the final. Most of the response I received was ridicule. Croatia did reach the final, losing 2-4 to France. In 2026, when competitions had to be played in empty stadiums, I reviewed 156 V.League matches and found the home win rate had fallen from 46 percent to 38 percent — a shift never previously recorded in the league's data. An empty stadium does not erase the truth. It only strips away the haze that 40,000 voices used to create. But I have to state clearly what many articles skipped: 38 percent does not prove that crowds were the sole cause. The 2026 period also brought a compressed schedule, restricted travel, and an increased number of substitutions. A sound model has to declare those variables too, otherwise it is merely a correlation dressed up as causation. The biggest risk in data journalism sits somewhere else. It is the void filled with a confident tone. A piece with no figures can still read persuasively if the author is skilled enough with words like spirit, character, hunger. Those words cannot be audited, and precisely because they cannot be audited, they are never wrong. The same holds in reverse. Someone who has just learned to read xG tends to turn it into a religion, converting every defeat into statistical injustice. A single number can lie, but a model validated across 10,000 matches has no reason to pretend. What deserves suspicion is how a metric gets labelled once it enters an article. For me, data is a map, not the territory. It is useful only when it carries a clear pedigree: which source, collected how, what sample size, how wide the confidence interval. Without a pedigree, a handsome metric is just a quotation with nobody accountable for it. In the coming matchdays, readers will receive more analysis than ever before, most of it written very fast. When reading a number, a reader should ask who counted it, and what the counter stands to lose if the number is wrong. I still keep a habit from 2026: before publishing any metric, I check it against a second source. On nights when the data sheet returns zero, I choose to print that zero, rather than paint over it with an adjective.