The Night the Data Board Went Blank at My Dinh, and the Discipline of a Sports Writer

At minute 34, the tracking board in front of me in the press tribune at My Dinh National Stadium was blank. Six panels, six zeros: passes, distance covered, duels, pass completion, heat map, expected goals. The home side had touched the ball several hundred times by then. The board was empty because the feed from the data provider had dropped in the 12th minute, and nobody in the newsroom knew.

To my left, a young colleague had already opened three browser tabs and typed the opening paragraph of his analysis piece. To my right, a photographer was asleep on his backpack. I took out my worn leather notebook, folded a sheet of paper into six columns, and started counting with my eyes.

That night taught me something nineteen years of following teams had not fully taught me: a blank board is an event. How a writer handles it says more about the trade than any board full of numbers.

Vietnamese football has joined the data pipeline

A professional match today flows through at least five layers of collection. Wide-angle cameras covering the whole pitch. Goal-line and behind-goal cameras reconstructing incidents. Semi-automated offside systems at the biggest competitions. GPS vests on players' backs. And event-tagging software operated by people pressing buttons. Each layer returns a file. The files merge into the board I see in the tribune, then flow on into the club's analysis room, into the television feed, into my article.

Vietnamese football joined that pipeline not long ago. VAR arrived in V.League 1 in the 2026 season. Many V.League 1 clubs now use GPS vests and employ data analysts. The national team has had a dedicated department for this for several years. That is real progress, and I do not want anyone to dismiss it.

But every pipeline has a single point of dependency: one provider, one server, one connection. When that point breaks, nothing replaces it.

In 2026, at Camp des Loges, I learned to generate my own data. Over six months following Paris Saint-Germain, I noticed Neymar spent fifteen minutes before every session dribbling through six cones in an unchanging order, quite different from the head coach's plan at the time. I counted 127 repetitions in one month. When I asked him about it, he just smiled and said, in his own words, that it was a habit. From the speed at which he performed that drill, I could read his mental state before each matchday, something no statistics table contains.

Then, on the night of 5 January 2026, at Rajamangala, Vietnam beat Thailand 3-2 and won the ASEAN Championship 5-3 on aggregate over two legs, with leading striker Nguyen Xuan Son already injured in the first leg. An hour after the final whistle, my newsroom had fourteen drafts. Nine of them quoted the same dataset drawn from the same API source. If that source was wrong, all nine were wrong in the same way, and not one of those nine writers had counted a line himself.

That was when I realised the sports news industry had changed its founding question: from what I saw to what the system returned.

When the data file comes back empty

In analysis work there is a concept few outside the trade name out loud: the empty file. The file is structurally valid, it has section headings, it has data fields, but every field is blank. For me, it is a report with all its headings in place and every heading reading insufficient information to assess.

What makes an empty file more dangerous than an ordinary error is that it does not announce itself. A program that crashes is noticed. A program that finishes, reports complete, and returns a void is not. This failure mode passes through every checkpoint because it wears the costume of a valid result.

On that night at My Dinh, the board in front of me was exactly such a file. No red alert, no connection error line. Just six white panels, looking precisely like a match that had not started.

Had I sat still, I had three choices. One: wait for the feed to return and file late. Two: borrow numbers from the previous match and write in generalities. Three: count for myself.

I took the third path. Here is what I counted over the remaining fifty-six minutes: the home side's central midfielder received the ball with his back to his own goal 41 times; the right full-back pushed beyond the halfway line 12 times, and on 9 of those occasions nobody covered behind him; the goalkeeper played long balls beyond 40 metres 3 times, and on all 3 the home side lost possession within ten seconds. None of those six panels recorded any of it. Even when the feed came back, they would not appear, because no system tags the situation a full-back pushing forward with nobody covering behind him.

I filed at 22:51. I quoted no metric from the provider. At the foot of the piece I wrote plainly that the figures came from the reporter's own count at the stadium, and that the automated data board had been down from the 12th to the 78th minute.

That footnote mattered more than the article itself. It told the reader which parts were measured and which parts were judgment. My trade survives because readers trust that distinction.

I first learned this lesson from the opposite direction, at the 2026 World Cup. Olivier Giroud scored only one goal in the tournament, and the French press called him a striker who could not score. I was assigned to record the behaviour of players who rarely make the highlights. I counted 214 pressing actions and 38 final-third recoveries by Giroud across seven matches, the highest in the squad. I wrote a 2,400-word piece about contributions that never reach the scoreboard. Didier Deschamps read it and invited me into a private press room to say thank you.

That dataset exists because a person sat down and counted. Machines did not generate it. That is the difference between data and a record, and it is also the line between an analysis room and a newsroom.

I have also held a single source for three weeks before writing one line, and been mocked by colleagues for being slow. Those three weeks were a measuring method. Patience in this trade is a measuring tool, standing on the same shelf as the hardware rather than decorating it.

Since that night at My Dinh, I set myself a gate before writing anything: the piece must name a team, name a person, carry a timestamp, contain at least one direct observation of mine, and contain at least one line stating clearly what I did not measure. Without the first three, a report does not exist. It is noise with a headline attached.

The trap is that writers fear blank space

Today's sports content industry rewards addition. Every optimisation guide circles around added value, around the reader having to learn something new. I do not argue against that. But most of the added value in an article without data is the added value of noise: one more judgment nobody can verify, one more prediction nobody pays for when it fails.

The other side of the argument is strong too. Readers pay for interpretation. A writer who can only say insufficient data is useless, and I agree with that. This trade lives on making calls, not on refusing to make them. I have to pick a side in a tactical story and say out loud what will happen in the second half.

The line sits at the label. A decent analysis piece can hold forty per cent conjecture, as long as the reader knows which forty per cent. The death of this trade does not come from a shortage of data. It comes from filling blank space in a confident voice and letting readers assume everything was measured.

A wrong table can be corrected. A blank table filled with feeling, then quoted, then quoted again, can no longer be traced by anyone. That holds for match statistics, and it holds for transfer stories nobody bothers to source-check.

My trade will not die of a data shortage. It will die of a fear of blank space.

The next signal is not a goal

A major tournament season is approaching, and newsrooms will roll in again with three trucks of equipment. What I am waiting for is not a finish or a signing. I am waiting to see who, among the clubs and broadcasters in Vietnam, publishes the first data incident log after a match: stating which minute the system went down, who was affected, and which metrics had to be discarded.

Whoever does it first will lose a few weeks to colleagues' jokes. Afterwards they will win the most expensive thing in this trade, the one no server can issue: the right to be believed.