The Empty Grid and the Data Discipline of a Table Tennis Writer

At 4:40 in the morning I opened the spreadsheet and found nine empty rows. The first row covered technique, tactics and equipment. The second covered player data and head-to-head records. The third covered the event system and points rules. The fourth covered the competitive landscape between table tennis nations. The fifth covered rules and governance. The sixth covered coaching staff and the talent pipeline. The seventh covered the risk surface. The eighth covered public narrative and expectation. The ninth covered the transmission of an entire industry.

At 4:52 the twenty-eight-year-old editor messaged me: Have you got a line yet? I wrote back: Nothing. Three seconds later, a question mark. I understood that question mark. Twelve years ago I would have sent exactly the same message, and I would have assumed the person answering me was lazy.

A day when the data source returned nothing. And the table tennis writer who got it right that day was the one who wrote nothing.

Seven minutes of video, seven years of re-watching. Every time I see a different match. I still use that line when I have to explain why I refuse to conclude after a first viewing. That night was different: I had nothing to re-watch. No footage. No scorecard. No event name. No date. A document arrived, and it was empty in exactly the places that matter.

Back in March 2026 I told an editor that three minutes of video could not explain a derby. Four days later I cut a seven-minute video myself, encoding every pass in the match between SIPG and Shenhua, with not one wasted word of voiceover. It reached 2.1 million views and moved me into a different job. The lesson was not to make short videos. The lesson was that the only thing keeping a viewer is information density, and information density only exists when the writer agrees to spend four days on seven minutes.

Tonight I had four days. I did not have the raw material.

The framework I use for table tennis did not come out of a meeting. It came out of the hardest question in the trade: what is your evidence?

The process has two steps. The extraction step reads the source text and pulls out the title, the source, the article type, the core viewpoints, the list of information points, the entities named, the time sensitivity and the source quality. The deep-analysis step examines nine dimensions, like that spreadsheet.

Between those two steps sits a single binding rule: every conclusion in the analysis step must be anchored to a specific information point from the extraction step. No information point, no conclusion. No entity, no subject. No time sensitivity, no judgement. The rule sounds rigid, and it is deliberately rigid, because in eighteen years of work I have seen far too much table tennis analysis built backwards: the conclusion written first, and the evidence then recruited to defend it.

That night, extraction returned nothing. No title. No source. Type unclassified. Core viewpoints blank in every sub-field: the one-sentence summary empty, the author stance undetermined, the purpose undetermined. The list of information points entirely empty. Entities not derivable. Time sensitivity not assessed. Source quality not assessed.

And because step one was empty, the nine dimensions behind it were empty too. I want to be explicit about that before walking through each dimension, because the first instinct of most people in this trade is to fill the gaps.

Based on my experience tracking matches, a piece written from memory always reads more smoothly than a piece written from data, and it is always wrong in precisely the place that matters. Memory remembers the winning shot and forgets the forty-five seconds before it. Memory remembers who won and forgets the structure that let them win.

Now let us walk the dimensions and be clear about the evidence each one demands.

Technique, tactics and equipment. A cell like this needs a style category: two-winged attack, pimpled rubber, two-winged blocking, away-from-the-table chopping. It needs the point-win rate on serve, the effectiveness of the third-ball attack, the average rally length, and things as mundane as height, footwork and age. On equipment, it needs to know whether the player has just changed rubber, changed blade, or only changed glue. A rubber change is not a story about a bat. It is a story about eighteen months of adaptation, and for eighteen months every performance metric is distorted in ways the scoreboard never records. Without that data, a remark that a player has slowed down is a guess wearing the clothes of analysis.

Player data and head-to-head records. This cell asks for world ranking, points to defend, and how closely the ranking matches true strength. Then it asks about head-to-head: overall, last two years, the three majors, and the nemesis question. Here I have a personal rule, learned after being punished by data more than once. Head-to-head is the most abused data type in table tennis, because it compresses fifteen years of two careers into a single percentage. A player who has won seven of ten can have lost the last three, and in a sport with dozens of events a year, the last three are the next three.

If tomorrow I want to write about a Vietnamese player such as Nguyen Anh Tu or Mai Hoang My Trang at an international event, this cell first demands a ranking, points to defend, major-event record and away form. It does not let me write that she is finding form just because I watched one match.

Event system and points rules. This is the dimension readers skip and writers undervalue. The cell asks for the champion ranking points, the prize money, the strength of the entry field, and the event position in the Olympic cycle. A small event in November can matter more than a big event in March, if you look at it from the points-to-defend column rather than from the advertising board. For Vietnamese fans, that means a qualifying event a domestic player enters can decide a major-tournament place more than a televised friendly ever will.

Competitive landscape and the China-versus-the-rest balance. This cell needs the number of top-10 seats, the number of titles at the last five editions of the three majors, and under-21 depth. Under-21 depth is a slow indicator. It does not say who wins this year; it says who will still be there in seven years. Writing about the present without a slow indicator produces a piece with a shelf life of forty-eight hours.

Rules and governance. This is where associations, event governing bodies and selection rules appear. The cell asks who benefits, who loses, and what precedents exist. Every rule change has winners and losers. A report that names only the winners is not yet a report. It is also the dimension where I am most careful, because it is where public opinion is easiest to push in the wrong direction.

Coaching staff and the talent pipeline. This cell asks about the head coach authority and competence, the fit of the personal coach, the stability of the coaching team, the age structure of the main tier and the conversion rate of the junior tier. Changing a personal coach does not change technique in three months. It changes the order of priorities in three weeks. And the order of priorities, not technique, decides whether a player dares to serve long at nine-all.

The risk surface. This dimension has six groups: competitive, selection, generational gap, governance and public opinion, systemic, and opponent. The biggest risk does not sit in any of the six cells. It sits in all six being empty while the report is still stamped as reviewed.

Public narrative and expectation. This cell asks about the durability of the narrative, a sample-size check, and the gap between public expectation and reality. The most durable story is not the best story. It is the story with data behind it and a sample large enough not to collapse at the first defeat.

Transmission through the table tennis industry. The final dimension splits the industry into three layers: upstream equipment, youth development and training; midstream events, associations and clubs; downstream broadcasting, commerce and derivative markets. One player changing bat sponsors can move part of the equipment market for two quarters. One change to the service rule can move an entire generation of coaching for ten years. With no entity named, none of the three layers can be drawn.

Nine dimensions, nine empty cells.

And this is where I want to slow down, because the dangerous part of the story is not the nine empty cells. It is the human response to them.

A report that looks complete is more dangerous than a report that is empty.

That is the real blind spot of this trade. When a framework comes with twelve headings and three tables, the pressure to fill it is structural, not moral. Editors need copy. Readers need conclusions. Algorithms need structure. All three forces push in one direction: fill the blanks.

And the most dangerous error in the trade is a silent one. A cell marked insufficient information routinely gets read as checked, no problem. The result is an analysis with no technical faults, no formatting violations, and nothing to say. It gets published, shared and cited, and nobody in that chain knows the whole building stands on an empty file.

I have made the same class of error at a much larger scale.

I was wrong in 2026, when I trusted the name and not the structure. On final night I said on air that a team with thirty-nine per cent possession was playing negative football. The next morning, positional data showed that team had produced twelve shots with an expected-goals value of 3.1, against 2.4 for the opponent. I had confused style with effectiveness. I wrote two thousand words arguing against myself, and from then on I banned words like cowardly and ugly from any analysis I write.

The 2026 error and tonight situation share a root: trusting a complete-looking form instead of the structure inside it. They differ in one way. In 2026 I filled the gap with a prejudice about a team. Tonight, if I am careless, I will fill the gap with a prejudice about a tournament.

Two hundred and eighteen matches with no crowd. Nobody to blame, nobody to celebrate. Only tactics left. I bring that experience back because it is why I believe in data discipline to an uncomfortable degree. Between May and July 2026, when competitions returned to empty stadiums, I collected data from 218 matches. Average goals fell from 2.7 to 2.1. The rate at which away teams avoided defeat rose by twelve per cent. A Championship club asked for my fifteen-page report. Those figures do not say which team was better. They say the environment is a tactical variable, and an omitted variable produces a wrong conclusion in a way that is very hard to detect.

The same principle applies when I move to table tennis: a grid with no environment column, covering home, away, a congested calendar, time zone, table surface and arena noise, will always misread the defeats in which a player did not actually play badly.

In a transfer window, the pressure to fill blanks doubles. In domestic leagues, rumours of players moving clubs appear daily, with numbers and unattributed quotes attached. Noise exceeds signal. The work is not to add conclusions but to filter: release clauses, wage bills, injury timing, and the structural logic of a squad. Ranking rumours by evidence matters more than ranking rumours by appeal, and that is the entire difference between a report and a report that gets shared a thousand times.

Someone will say I am turning caution into an excuse for not working. I have heard it. The charge lands on people who use caution to dodge conclusions. It misses people who use it to separate I have not found it yet from there is nothing to find. Humility belongs in the admission of a data limit. Once that admission is made, the writer still has to conclude, still has to choose, still has to issue a judgement that can be proven wrong.

If I wrote insufficient information in every cell and stopped there, I would have turned care into an escape hatch. The editor question mark was a timely reminder: after saying there is no data, the next job is to say what data is needed, where to get it, and when it will arrive.

Three signals to track over the coming week.

The fill rate of information points in the extraction step. If it stays at zero, every downstream analysis is decoration. Reachability of the original source: an empty file is usually a transmission or scraping fault, and rarely a genuinely empty article. Schema-conformance rate over a rolling window: one failure is an incident, twenty failures in a week is a systemic defect, and systemic defects cannot be fixed by writing better.

I will leave the spreadsheet empty for one more day. When the data file arrives, I will reopen those nine rows and read from the top, and I will remember that the only thing that makes this piece worth reading is not the nine empty rows. It is that I refused to fill them.

When a table tennis analysis is published with no source, no date and no named entity, the work is to ask one question: what is this piece using to speak?