Basketball, Data and an Empty Report: The Discipline of the Storyteller

A computer screen in a small room beside a training court lit up with a report that was flawless in form: headings, tables, proper order, even a closing line of conclusions. The content itself was empty. No team name, no player name, not a single metric filled in. The analyst stared at that blank space longer than necessary, then shut the machine down.

Anyone who has followed basketball seriously knows the feeling. It matches the moment a coach opens the box score at halftime and discovers the stat crew stopped recording in the first quarter. The arena remembers the final score. Nobody remembers why the score became what it was.

An empire does not collapse with thunder; it collapses with a slip in the final seconds. A sports story collapses the same way: it falls because a data cell was left blank, and because a writer decided to fill that blank with his own imagination.

The Vietnam Basketball Association (VBA) played its first season in 2026, according to the organisers. Nearly a decade later, the league has gone through several changes in team count, fluctuating between seven and eight clubs, while keeping limits on the number of foreign imports and overseas Vietnamese players each roster may register. That mechanism creates a very specific labour market: the quality of local players sets a team's ceiling, the quality of imports sets its floor.

The data infrastructure of Vietnamese basketball has grown alongside the league. Live box scores, shooting percentages, turnovers and contested possessions are updated possession by possession. Some teams hire video analysts, cut clips, and count how often opponents run a set play in the fourth quarter. On a few benches, an assistant sits beside a laptop all game, typing continuously.

Most fans reach basketball by another route: short clips. A dunk, a three at the buzzer, a block that flies out of bounds. Those clips spread fast and generate emotion brilliantly. They are far weaker at explaining.

Vietnamese basketball also has a second layer of competition beyond the domestic league: national team windows and regional multi-sport games. At that level, preparation time is short, the number of games is small, and every error in player evaluation costs far more than in a routine league fixture.

The gap between those two approaches is where a sportswriter is tested. From my own experience tracking games across several seasons, the hardest part has never been describing what happened. The hardest part is refusing to describe what I did not witness, and refusing to conclude what I have no data to support.

The biggest risk in basketball analysis is not missing data. It is producing a conclusion out of the empty space where data should be.

There is a paradox in how basketball metrics are read. The basic box score, points, rebounds and assists, is the easiest thing to read and the easiest thing to be misled by. A player who scores 24 points in a 20-point loss may have played very well, or may have consumed nearly thirty possessions to reach that number. Look at true shooting percentage and usage rate, and the picture splits in two. The same 24 points can represent two entirely different levels of contribution.

In Vietnamese basketball, this distortion usually appears in the import slot. A team at the bottom of the standings is forced to hand the ball to its import on almost every possession, because that is the only option that produces points. The result is a stat line that looks spectacular: 27 points, 12 rebounds, 4 assists a night. Placed next to the team's standing, the shooting percentage of the surrounding players, and the number of possessions that player finished, the real value is far smaller than the number. This is where writers fall into the trap most easily: praising a player when what is actually being praised is that player's circumstances.

The opposite direction exists and receives less attention. Local players who contribute through ball movement, off-ball cutting and well-placed screens are nearly invisible in a basic box score. A lead guard who scores 8 points but controls tempo and reduces the whole team's turnover count rarely appears in any headline. Impact metrics such as plus-minus while on the floor are the appropriate tool. In the VBA those metrics exist, but they are not yet widely used in public discussion.

There is a comparison I use fairly often. A basketball game runs like a competitive video game that has just received a new patch. Game pace is map tempo, space inside the paint is vision control, and a two-man action is a coordinated play between two lanes. When an opponent changes how it defends, the patch has changed the rules, and all previously accumulated data becomes reference material only. A team that fails to update will watch its numbers fall apart within a quarter.

The biggest difference between basketball and a video game lies in control over the data. In a game, numbers are perfectly accurate because the system records them automatically. In basketball, especially in leagues with limited resources, data is a human product: someone types, someone checks, someone decides which pass counts as an assist. Errors sit scattered everywhere. That does not strip data of value; it makes the discipline of reading data mandatory.

I always apply three layers of checks when evaluating a player or a team: the denominator behind each metric, the context of the team, and the gap between the regular season and the playoffs. The denominator determines meaning; a 42 percent shooting mark over four attempts differs completely from 42 percent over two hundred attempts. Context determines cause; the same metric means something different on a team fighting for a playoff berth than on a last-place team. The playoffs compress everything: fewer possessions, slower pace, half-court defence taking up most of the game, and players who live on fast breaks suddenly losing ground.

At team level, tiers must be read through data rather than through feeling. Title contenders, playoff teams, play-off chasers and rebuilding rosters each need a different set of criteria. Judging a rebuilding team by the standards of a champion is the most common error in commentary, and it usually stems from missing context rather than missing statistics.

A decent scouting report needs three minimum items: player name, season and team, plus at least one performance metric. Remove one of the three and the report becomes a personal introduction rather than a decision tool. The problem is not that teams lack data. The problem is that people still make decisions before the data arrives.

The most vulnerable part of a sportswriter is empathy. Anyone in this trade long enough has written a sympathetic piece about a losing team, sometimes better than the piece about the winner. That feeling is legitimate, and it carries a trap: sometimes the losing team simply played worse that day. Not every defeat is a romantic tragedy. Painting a loss in rosy colours is an act of dishonesty, and it insults the very people who lost, because they know exactly where they made mistakes.

Another use of data is just as dangerous: turning metrics into truth that cannot be challenged. Metrics have denominators, context, error margins and human recorders. When a stat table states no time frame, no opponent and no possession count, it is presenting a conclusion without its conditions of application. A basketball report that is complete in form but empty in content is a miniature version of that problem: when there is no data, the only correct choice is to say there is no data.

A report does not collapse under an accusation; it collapses under a blank cell. A writer collapses in exactly the same way, and often never notices, because nobody checks the empty parts of his work.

Vietnamese basketball is at a stage where every season brings more data, more analysts, and more fans willing to sit down after the game and read the box score. The value of this stage depends on one small habit: before writing a concluding sentence, ask what supports it. If nothing does, the blank space must be left blank.