Table Tennis: The Discipline of an Empty Data Sheet and the Value of Saying 'Not Yet

Two in the morning. The wall clock in the Shanghai office gives one dry click before the hour turns. An intern sends me a nine-tab spreadsheet. Tab one: technique and tactics. Tab two: player data and head-to-head records. Tab three: event system and points rules. Tab four: competitive landscape. Tab five: rules and governance. Tab six: coaching staff and youth pipeline. Tab seven: risk surface. Tab eight: public narrative. Tab nine: industry transmission. All nine tab headings are filled in, correctly, in the right font. Every cell underneath is empty.

His message reads: "I'll just fill in last tournament's numbers so it looks complete, the deadline is tomorrow morning."

I closed the laptop halfway, poured more water, and spent forty minutes answering with one word: No.

In table tennis reporting, a deadline is a physical force. It always pushes the writer toward the nearest number, never the correct one. And a nine-tab sheet with full headings and an empty body is more dangerous than a blank page, because it looks finished. Nobody discovers the emptiness until the piece has already gone out.

This article is about those forty minutes. About what an empty data sheet taught me about table tennis analysis, about the WTT points cycle, and about the traps that sit between a serve and a headline.

Context: the sport everyone thinks is easy to write with numbers

Table tennis looks like a sport that can be summarised by a single figure. World ranking, matches won, titles taken. But the thing called a "world ranking" is a moving balance sheet: the WTT ranking system, run by the body that operates the global professional tour, uses a rolling 52-week window. Points from each event expire exactly one year after that event ends, and every week that passes is a week in which someone loses points without losing a match.

The consequence is concrete. A player can hold the same ranking position while their actual level has moved far away, simply because this same week last year they reached a final. And a rising player can drop after a good month, because old points fall out faster than new ones come in. The ranking is not a hall of fame. It is a 52-week statement.

Above that system sit the three events treated as the true measure: the Olympic Games, the World Championships and the World Cup. Every argument about who is better, if it does not reference these three, is an argument about memory.

Below them is a tier ladder with explicit weights: a Grand Smash awards 2,000 points to the champion, a Champions event 1,000, a Star Contender 600, a Contender 400, plus continental and domestic events. Each tier differs not only in prize money but in structure: entry quotas, protected seeds, and mandatory participation obligations. For a player in the leading group, a year's calendar is not a competition schedule. It is a points-defence schedule.

At club level, the table tennis transfer market runs to its own rhythm: the Chinese national league, the German Bundesliga, Japan's T.League, France's Pro A. Players change teams, contracts are short, and every move reshapes both the points picture and the reporting picture. Fans follow their player across three or four different league systems with three or four different sets of standards, and usually nobody tells them the numbers are not in the same unit.

In 2026, twenty-six years old and fresh out of a statistics degree, I interviewed at a sports data company in Shanghai. A senior male director looked at me and asked whether I really understood football, or only liked looking at handsome players. I did not argue. I opened my laptop and presented a model predicting the results of a club's last ten matches using xG, expected goals, and PPDA, passes allowed per defensive action. Error: 1.2 matches. I was hired, at fifteen per cent less than a male colleague in the same role.

Since that day I have kept one habit: every article opens with a specific number or index, and every claim carries a statistical source note. People cannot argue with my voice. They can only argue with a table, and a table defends me better than any rebuttal.

Tab one: what a technical judgement needs in order to stand

When we say a player has "improved", the sheet has to answer four questions. First, their point-win rate on their own serve, broken down by spin type. Second, their point-win rate on return, split by whether the opponent served short or long. Third, the rally-length distribution, meaning the share of points ending inside the first three shots versus those extended into rallies. Fourth, point-win rate by phase: the first three shots, the medium exchange, the long rally.

Those four questions make a map. Without a map, the word "improvement" is an emotional state written in the present tense. A technical judgement without a phase-by-phase point distribution is just a feeling presented in a confident voice.

The same rule applies to equipment. A player who changes rubber, sponge thickness or pimples enters an adaptation window that typically runs three to six months, during which results fluctuate and cannot be used as evidence for any conclusion about form. The ITTF, the sport's world federation, caps the total thickness of a racket covering at 4mm, and for pimpled rubber the pimples themselves at 2mm. Those are checkable figures with documents and effective dates. But if an article names no equipment change at all, then any judgement about an adaptation window is a judgement about something that does not exist in the article.

That is exactly what the nine-tab sheet told me at two in the morning: not one line on playing style, not one line on a technical element, not one line on coaching deployment. So tab one stays empty. Not because the writer was lazy. Because there was genuinely nothing to write.

Tab two: a points total is not a ranking

There is a distinction that table tennis readers are rarely told. A player's ranking is the sum of scores with different expiry dates. To read it properly you must decompose it: how much comes from the three majors, how much from Grand Smashes, how much from continental events, how much expires within the next eight weeks. Points-defence pressure is the gap between expiring points and the minimum a player can realistically replace in the same window.

A points total is not a ranking. It is a 52-week payment history, and the reader is entitled to know which invoices are about to fall due.

With head-to-head records, a single aggregate number has no diagnostic value. You need windows: the last two years, and the three majors separately. A player may win eight of ten against an opponent at small events and lose all three meetings at majors. Those two facts tell entirely different stories, and only one of them is useful for an upcoming match at a major.

The most discriminating metric in the Chinese analytical frame is the foreign-match win rate, meaning the win rate against players from other associations. A player can dominate domestically and still be underrated, because the domestic circuit does not test the thing that needs testing. Conversely, some modestly ranked players carry very high foreign-match win rates. That group is usually overlooked by media, and usually decides team results.

On the age curve, the common grouping is under 22 rising, 22 to 28 peak, over 28 veteran, with the markers arriving roughly one to two years earlier for women. Each group carries its own risk pattern and its own question. But placing a player in the right group requires a date of birth and a career stage. Without those two data points, every sentence about a "golden cycle" is inference.

In 2026, when Germany were eliminated in the group stage after losing to South Korea, I wrote a piece showing that Germany generated only 0.48 xG while South Korea defended in a 5-4-1 block with an average PPDA of 6.2. In other words, Germany were not pressed into submission. They lost their own rhythm. The piece drew a wave of comments, including one I still remember verbatim: "women guessing." I did not argue. I posted the forty-page raw data file from Opta alongside it.

That Germany defeat taught me that accuracy can be a lonely place. It also taught me that accuracy is the only thing still standing after the readers' anger has faded. People can doubt my gender. They cannot doubt a data file with a timestamp column.

Tab three: the event system and the points rules

Each tier of the ladder is a mechanism for redistributing opportunity. A player who grinds Contenders for steady points holds a completely different points portfolio from one who only plays Grand Smashes and Champions events. Neither strategy is absolutely right; they pay in different currencies. One pays in matches and physical load, the other in the risk of an early exit at a major.

Mandatory participation turns rest into a cost. This is the point media usually skip when commenting on a star's withdrawal: withdrawing is not only losing a chance to win, it is losing points, losing a seeding slot, and possibly losing a bye into the main draw at the next event. Those three losses compound along a curve outsiders never see.

The draw is data, not decoration. Seed distribution, the likelihood of a "group of death", and the organiser's separation of players from the same association in the early rounds all bear directly on the probability of going deep. Without the draw in hand, any pre-tournament prediction is a prediction about a tournament with no draw.

Tab four: the landscape and the gap

The world table tennis map is drawn in four tiers: the dominant tier, the second group, the emerging forces, and the rest. Three indicators measure the gap between the dominant tier and everyone else: seats in the world top ten, titles at the last five editions of the three majors, and the depth of the under-21 cohort.

What matters is that these three indicators often tell different stories. An association can hold most of the top ten and still lose major finals. Another can hold few top-ten seats but field two players capable of beating anyone on a good day. Depth and peak are different quantities, and different table tennis nations choose to invest in one or the other.

Challenger models also differ in kind. One nation chooses a generational-skip model, pushing very young players onto the big stage early and accepting defeats. Another chooses an all-out attacking style, accepting high risk in exchange for pressure. Another leans on the European power-loop tradition. Another is rising on a new cohort trained to a data standard. And there is an overseas legion: players trained in one system but competing under another association's flag, usually the least predictable variable in team events.

I watch international matches a little differently from my colleagues: I record not the result but the way the result was produced. A 3-0 win can contain three sets in which the winner trailed midway through. Read only the score and you are reading the visible part of a much larger body of data.

Tab five: rules and governance

Four groups of rules routinely create movement that fans do not attribute to rules. The first is competition law: scoring, service alternation, hidden-serve and toss regulations. The second is event-system law: entry quotas, mandatory participation, points allocation. The third is selection law: the criteria for putting a player on a national team. The fourth is disciplinary law and equipment compliance.

Of the four, the third carries the most weight and is discussed the least. Whenever a national team announces a squad, the real question is not who was picked but what the ratio is between published quantitative criteria and the selector's discretionary room. If the criteria are published down to the number and applied mechanically, the list becomes predictable and the argument shrinks. If the criteria are vague in a few places, that discretionary room becomes the place where every suspicion pours in.

On equipment compliance, this is a field where a very small deviation can erase a result. Covering thickness, surface treatment, the legality of glue — all of it is regulated and all of it is inspected. An article about a shock defeat that does not check this family of causes has missed one of the few causes that can be verified with a measuring instrument.

Tab six: coaching and the pipeline

The health of a development system is measured by three things: the age structure of the main squad, the conversion efficiency from junior to senior level, and the gap in the middle age band. That gap is the earliest indicator and the most ignored.

When the 23-to-26 band thins out, the system faces a choice: extend the careers of the veterans, or push very young players up earlier than they should go. The second option has a heavy name in the trade: forcing seedlings to grow. It produces players who appear early, carry heavy pressure, and sometimes never reach the peak they should have reached at 24.

At coaching level, three variables need watching: the head coach's capability and authority, the fit between player and personal coach, and staff stability. The second is rarely mentioned and often decisive. A player who changes personal coach during a technical rebuild goes through two compounding disturbances, and results over those six months are close to unusable as an assessment of true ability.

Tab seven: the risk surface and the public narrative

Six standard risk groups exist in table tennis analysis: competitive risk, selection risk, generational-gap risk, governance and public-opinion risk, systemic risk, and opponent risk. Each has its own trigger. A concealed injury inside a technical-rebuild window. A slump from equipment adaptation. A style decoded by video. A points-defence collapse. A trust crisis in foreign results. A middle-tier vacuum.

There is a seventh group that the nine-tab sheet generated by itself: analytical-integrity risk. The risk of issuing a very confident conclusion on a base of zero information. Of the seven, this was the only one rated high that night, and the only one that could be fixed immediately by stopping.

On the public narrative, there is one test I always run: the overhype test. Is a fine performance a single event or a season-long pattern? The answer lies in sample size, and sample size is the thing almost never stated in a headline.

In 2026, when global football stopped for the pandemic, I was at home and realised this was the only window in history to measure the effect of crowds on performance. I collected Bundesliga data when the league restarted in May, comparing 120 matches with crowds from the previous season against 98 matches without. Successful passes rose 7.3 per cent, sprints above 30 km/h fell 11 per cent, and goals from set pieces rose 14 per cent. I told nobody. I sent a 62-page report by internal mail at eleven at night.

There are evenings I sit with the numbers longer than with people, and I have never once felt alone.

That dataset left me one habit: before publishing any analysis, I ask a single question. Does this data change when the context changes? If yes, every conclusion carries a condition. If I do not know, it does not go out.

Tab eight: industry transmission

The chain runs from upstream equipment, youth development and coaching systems; through the middle layer of events, associations and clubs; down to broadcasting, commerce and derivative markets. Each layer has its own lag. A rule change in the middle layer takes one to two seasons to reach the end of the chain. An equipment change upstream can travel faster, but only shows up in results once the adaptation window closes.

There is one professional boundary I hold tightly: competitive value and commercial value are different quantities, and mixing them is the source of most bad analysis. A player with a large following does not thereby have a higher foreign-match win rate. A player rarely named is not thereby less important in a team tie.

On market data, we use it only as an objective expectation signal, never as a recommendation. A price is a crowd's opinion expressed as a number. It can be right, it can be wrong, and it never replaces analysis.

The contrarian angle: the empty report was the most valuable document of that night

I know this sounds backwards for a newsroom. But let me finish the story of those forty minutes.

I opened each tab and showed the intern that all nine were answering a different question from the one the article needed. They were answering: does this spreadsheet look like a spreadsheet. They were not answering: is there evidence for each sentence about to be written.

If he had pasted in last tournament's numbers, the piece would have existed in four hours. It would have had rankings, head-to-heads, percentages, verdicts. And it would have been wrong in a place nobody could check, because nobody would have a source to check it against. That kind of error does not produce a lawsuit. It only wears away the one thing that keeps this trade alive.

An honest empty report is worth more than a full report with a bad source, because the first shows you where the pipeline broke, while the second teaches you to trust the break.

There is a subtler temptation I meet constantly: filling the gap with model memory. The writer does not invent a number, but half-remembers that the player is ranked seventh, and writes it down. Under a rolling 52-week system, a six-month-old memory is enough to be wrong. Memory has no expiry date. Points do.

And there is a third temptation, the most dangerous because it gets praised: the fame filter. Once a name is famous, every number about that person is read favourably. Once a name is unknown, every number about that person is read the other way. The filter does not live in the data. It lives in the reader, and the writer has a duty to remove it before picking up the pen.

In 2026, while the media talked only about two familiar names in a major national team's campaign, I noticed a substitute. My data showed he covered 11.8 km per match and, more importantly, that 67 per cent of his runs went into the space behind the opposing back line, the highest in the tournament. I wrote about him. Two weeks later his assistant emailed to say the piece had helped him understand his own value and play with more confidence off the bench.

That is why I still do this job. Data is not made to decorate a conclusion you already had. It is made to find a person nobody bothered to look at. But for that to happen, the data has to exist first. When it does not exist, the only correct act is to say it does not exist.

Data does not lie. Only the reader has not been honest enough.

What to watch in the next cycle

Five signals, and I suggest readers watch them with me.

First, the fill rate of table tennis coverage: how many judgements actually carry a sourced number. Second, the presence of a source-tier field: does the piece say whether the information comes from an official announcement, from authoritative media, or from an unverified account. Third, entity resolvability: does the piece name the player, the event, the association, or only speak of an unnamed "champion". Fourth, time-sensitivity assessment: a number without a date is a number that cannot be used. Fifth, how a newsroom behaves when there is no data — whether it dares publish a short note saying the basis is insufficient.

The traveller does not need a compass if they have read enough data about the winds. But the wise traveller knows that, having read no wind at all, the first act is to stop and write in the log that the air is still. Not every blank space is a malfunction. Some blank spaces are data.

And if tomorrow someone asks me why a table tennis report can be only three lines long, I will answer with exactly those three lines. In an industry where everyone is afraid of being thought ignorant, the person willing to say "not enough data" is usually the only one who actually checked.