The Empty Report and the Limits of Table Tennis Analysis

There is a moment in this trade that sports writers rarely describe: the moment a data file opens and turns out to be empty.

Late at night in Guangzhou, I opened an extraction file for a table tennis analysis. Nine sections. Each one carried a single line: insufficient information. Source article title: none. Source: none. Information points list: empty, not a single row. Yet the additional notes field still held its instructions intact — “identify entities from the information points above,” “judge source quality from the source fields.” A template had been emitted and never filled. The template said more about itself than any content could have said from inside it.

The first reflex of anyone who has written sport is to fill the gap. I have done it. That is why I stayed with the empty file instead of closing it and going to sleep.

Professional table tennis analysis now runs on two tiers. Tier one is extraction: who played, which event, what score, at what date, at what source tier. Tier two is judgment: technique and tactics, player and head-to-head data, event systems and points, the landscape between the Chinese team and the rest, rules and governance, coaching staff and the talent pipeline, the risk surface, the media narrative, the industry transmission chain. Nine dimensions. Not one of them stands on its own if tier one is empty.

What matters here is that the framework was not broken. The structure was right, the order was right, the questions were right. The failure sat in ingestion. In systems work, this is the hardest failure to see, because it raises no error message — it produces a report that looks very much like a report.

In table tennis this kind of failure costs more than in many sports, because the sport enforces a strict clock. The ITTF and WTT ranking system in force since 2026 calculates points on a rolling 52-week window: the points from a title expire in the fifty-second week after the title was won. The ranking read today differs from the ranking read six months ago, not because anyone played better, but because a points column has fallen out of the window. From my own experience watching matches at WTT stops across Asia, most online arguments about a “form slump” are really arguments about an expiry calendar.

In the market where I work, most Vietnamese-language readers reach table tennis through translations from Chinese sources. That chain has a structural weakness: every translation is a compression, and every compression loses a timestamp. A ranking translated from an article written three months ago walks into a new piece without carrying its date. The reader ends up holding a fragment of data with no stamp on it.

When data is missing, writers reach for the most famous name in memory and attach a conclusion to it. I call that the fame filter. It is dangerous because it always produces a fluent sentence. The fame filter never returns an empty result; it only returns a wrong one. In table tennis the error is measurable: a player defending points in a given week is a completely different entity from the same player in a week with a clear calendar. Same win rate, two meanings.

In 2026, at forty-eight, I was writing about the World Cup. I criticised Roberto Martinez’s 3-4-3 because Belgium’s back line was unbalanced, and I wrote that Kevin De Bruyne running too much would break the team apart. De Bruyne averaged 11.2 kilometres per match, the highest in the tournament. Belgium beat Brazil 2-1 in the quarter-final using exactly that shape. Readers pushed back hard, and they were right on the conclusion. “Belgium 2026: a team of stars, missing the pieces.” I still keep that line, but as a note on method, not as a prophecy.

The lesson is not “stop criticising.” The lesson is that I criticised without a criteria set. So I built one for football: pressing index, distance covered, touches inside the opponent’s box. Table tennis needs its own, and I use exactly five: point-win rate on serve, efficiency across the first three shots, rally-length distribution, error rate on receive, and win rate from 9-9 onward. Without that set, every remark is just a feeling written in the past tense.

One more variable is usually skipped: the equipment adaptation window. When a player changes rubber or blade structure, coaching circles reckon on three to six months of fluctuation before the feel settles. Inside that window, results do not reflect ability. Reading results inside that window without naming the window is a structural error, not a matter of opinion. This is why I have added a separate section to every piece since 2026: the gap map — gaps on the table, gaps in the calendar, gaps in the adaptation cycle. “Dead space, living match.” Gaps are the third eye for reading real intent, but they are only trustworthy when match data proves they exist. A gap with no numbers behind it is just a handsome picture.

Another test I always apply: a fine rally in one match is a small sample; a run of form across a season is a large one. Table tennis moves in short rhythms, so the small-sample temptation is strong — four straight service winners are enough to become a headline. Four points say nothing about a 52-week season.

One detail in that empty file is worth more than a hundred lines of “insufficient information”: the notes field still held instructions rather than findings. That tells you the template was emitted before ingestion completed, or after it failed. For anyone who builds systems, this is a clearer signal than any error log: the system did not break in reasoning, it broke in fetching. Fixing the reasoning layer will never repair this.

From that empty file I drew an operating rule, and this is the part I want people in the trade to read closely: a report may only move into the judgment tier when the extraction tier clears a minimum threshold — at least one named entity, at least three information points, and a non-empty source field. Below that threshold, close the file with a single null paragraph instead of stacking nine “insufficient information” sections into a report. The difference between the two ways of closing a file is the difference between honesty and decoration.

The counter-intuitive angle sits here: in this trade, a formally complete report is always rated above a substantively correct one. Nine sections, a table in each, a few rows in each table — that creates a feeling of control. A paragraph saying “the source article contains no assessable competitive content” does not. But that feeling of control is the most expensive thing the reader pays for, because it moves risk from the writer to the reader.

There is a deeper layer. “Every tactical system is a confession.” So is a report. An empty report confesses that the writer does not know. A report filled from memory confesses that the writer did not verify. Readers are entitled to know which confession they are reading.

Cost says the same thing. A null paragraph costs a newsroom half a day and leaves no aftermath. A ranking wrong by a week costs a week of corrections, and a slice of trust that does not come back.

The risk is very concrete. If a ranking table is filled from memory instead of data, that value may already be weeks out of date under the 52-week calendar. Being weeks out in table tennis is a large error, because seeding and draw brackets shift week by week. And even with extraction complete, one more gate remains: source tiering. Every sensitive story — match-fixing suspicion, selection dispute, unverified injury news — needs its own source-tier pass before it is written. Skip that gate and the writer is no longer an analyst, but a rumour relay with better syntax.

What I want to verify at the next event is simple and needs no table of figures. Before writing any judgment about a player, I will ask myself: which point in this file expires in which week, and where did I read it. If I cannot answer, I will write a paragraph instead of nine sections. “Data does not lie, only the reader misreads it.” “The tactics board has no room for noise.” “An empty arena says more than thirty thousand spectators.” An empty data file — how many people in this trade would dare file it as it is?