Nine Layers of Table Tennis Data — and the Lesson of an Empty Sheet

Nine Layers of Table Tennis Data — and the Lesson of an Empty Sheet The empty sheet at 3:47 a.m. It was 3:47 a.m. in Shenzhen when I reopened the spreadsheet I had left running overnight. It returned exactly one populated field: the domain label, table tennis. The nine analytical layers beneath it were blank. No player name. No match date. No single indicator. Just a label hanging over empty space. I sat and looked at it for about twenty minutes. In this profession, twenty minutes spent staring at an empty sheet is an investment, even if nobody pays for it. My editor messaged to ask whether I could file, about five thousand words, deadline in two days. I knew the shortest route exactly: open my memory, pull out a few familiar names, weave a plausible story, insert some figures I believe I still remember correctly, and close with a decisive conclusion. The reader would be satisfied. The editor would be satisfied. Only one person would not be satisfied, and that person was sitting in front of a screen at nearly four in the morning. In 2026, aged twenty, I was an intern at a sports newsroom in Shenzhen. At the post-match press conference I asked the head coach about the home side's shifting formation. A senior male reporter sitting beside me cut me off mid-sentence: girl, just note the goals, leave tactics to us. I did not argue. I went back, tabulated all thirty matches of that season, and found the team had lost eight of nine games whenever it surrendered control of the middle third. Two weeks later, the editor-in-chief handed me the data analysis column. In 2026, the press conference door closed in my face. Today, I read it through data. And tonight, the data is telling me something hard to hear: there is nothing to read. The nine-layer framework: a net for catching data Any serious table tennis analysis has to pass through nine layers: technique, tactics and equipment; player data and head-to-head records; the event system and points rules; the competitive landscape between China and the rest of the world; rules and governance; coaching staff and the talent pipeline; the risk surface; public narrative and expectation; and finally the industry transmission chain. I built this framework after years of noticing something uncomfortable. Table tennis is a sport in which a single number lies very convincingly. A ranking position conceals the participation schedule behind it. A service-point win rate conceals the level of the opponent. A head-to-head record conceals an entire equipment era. Three players with identical 68 percent win rates can be playing three different sports, if nobody bothers to read the footnotes. Equipment divides table tennis into sharper eras than most sports. The 38-millimetre celluloid ball was replaced by a 40-millimetre ball in October 2026. Speed glue was banned in September 2026, immediately after the Beijing Olympics. The plastic ball replaced celluloid from 2026 and completed the transition in 2026. Each time, the flight path, the spin and the rhythm of a rally changed. A head-to-head record running from 2026 to 2026 is comparing two different sports under one name. Then there is time. The international federation's world ranking operates on a rolling 52-week mechanism. The points a player holds today are a direct consequence of which events that player entered exactly one year ago. Old points expire on schedule. Points-defence pressure is therefore not distributed evenly across the year; it clusters in the months that coincide with the events where that player previously peaked. The result is a structural constraint, not a stylistic preference: without dates, there is no analysis. Not because I am a perfectionist, but because every conclusion is a function of the calendar, and a function without a time variable cannot be computed. Back to the spreadsheet at nearly four in the morning. Our pipeline runs in two stages. Stage one reads raw text and extracts structured information points: entities, events, dates, sources. Stage two takes those points and applies the nine-layer framework. Tonight, stage one returned an empty array. Only the domain label was populated, along with a self-aware note that time sensitivity had not been assessed. Everything else was whitespace. I had two options. The first was to write an analysis. That option required me to generate data from memory and then present that self-generated data in the tone of someone who had verified it. The second was to write a report about the whitespace itself, and use it to re-examine how we read table tennis through numbers. I chose the second. But to do that usefully, I had to state precisely what was missing, and what should have been there at each layer. Technique, tactics and equipment Tactics are what people draw on a whiteboard. Data is what they draw on reality. But table tennis reality is only measurable if you know what you are measuring. A complete technical profile has four groups. The first is playing style: two-winged attack, right-side attack, defensive chopping, close-to-table or away-from-table play. The second is specific technical elements: serve, receive, the first three strokes, the long rally, short-ball handling. The third is the physical profile tied to that style, because a game requiring constant lateral movement spends knees and hips in a completely different way from a blocking game played close to the table. The fourth is measurement. Four indicators matter more than the rest. The share of points finished within the first three strokes tells you whether a player lives on the serve or on the rally. Service-point win rate tells you about the quality of the start. The average rally length in won points versus lost points is the indicator I use most, the way football analysts use pressing metrics; it shows exactly which state a player wants to drag the opponent into. And win rate from nine points onward, when each rally is worth three times a first-game point. Equipment is the most misunderstood part of popular writing. Rubber hardness, a five-ply or seven-ply blade, short or long pimples — these are not decorative details. A rubber change can cost two weeks of lost feel, and in the current dense calendar, two weeks is two tournaments. So every equipment file needs three columns: what changed, whether it has been mastered, and which group of opponents that change creates an advantage against. This layer returned zero. No style named. No technical element named. No equipment named. Nothing to assess. I still give it space, because if a table tennis file has no such layer, the reader deserves to know they are reading an incomplete file. Player data and head-to-head records Players leave the arena, spectators leave the stands, but data never leaves the game. The problem is that data only answers when you call it by the right name. A player profile has four basic fields. World ranking and its trend. Current points composition and the upcoming expiry schedule. Age and position on the career curve. And head-to-head record, split into three layers: overall, last two years, and majors only. The third layer is the one that matters. In table tennis, overall head-to-head records are polluted by three things. Small sample size, because a specialist may meet a specific opponent only a handful of times per season. The time gap between meetings, combined with different ball eras, which makes aggregation meaningless. And tournament bias, because a qualifying-round match at a minor event is not the same object as a semifinal at a top-tier event. I often use recent history to illustrate how ranking diverges from true strength. Ma Long won men's singles gold at the Rio 2026 Olympics and the Tokyo Olympics held in 2026, and took the world singles title in 2026, 2026 and 2026 — a collection spanning four event eras, three ball generations and two service-rule changes. Fan Zhendong won the men's singles at the Paris 2026 Olympics and held world number one for very long stretches before that. Truls Moregard, a Swedish player, eliminated Wang Chuqin in the first round at Paris and reached the final before losing to Fan Zhendong. One tournament, one draw, three entirely different fates. Read only the ranking, and Moregard is not a threat to the world number one. Read only the Paris result, and Moregard is a top-tier threat. Both conclusions are correct simultaneously, because they measure different things: one measures annual output, the other measures the ability to peak in one specific week. That is precisely the tension every player profile must handle — and precisely the tension an empty spreadsheet cannot handle. Across more than a decade of watching tournaments, I have drawn one uncomfortable rule: most shocks at major events do not come from a weaker player beating a stronger one. They come from a player peaking in the exact week of the event while the opponent is in a points-expiry stretch and under pressure to defend a position. Annual data cannot see the week. Only weekly data sees the week. This layer returned zero, and here the blank does the most damage. No name means no ranking, which means no expiry schedule, which means nothing. Our framework has a strict rule: when information is missing, record that it is missing; do not infer. That rule exists because of this layer. A fabricated name pulls in a fabricated ranking, then a fabricated expiry schedule, then a fabricated forecast, and by the end of the chain nobody remembers the starting point was fiction. Event system and points rules This is the most date-dependent of the nine layers. The international competitive structure changed fundamentally in 2026, when a new commercial series replaced the old tour. It tiers events by points value and prize money: a top group running only a few times a year, a middle group running more often, and a lower group where younger players accumulate points. Three questions define this layer. Which tier is the event in, and how many points does the champion receive. Where does it sit in the four-year Olympic cycle. And what effect does it have on entry slots for players from a country with high internal competition density. The third question is least discussed and decides the most. For federations with dozens of players good enough for major events, slot allocation is an internal problem, and that problem runs on points. Points do not only determine seeding. Points determine who travels, who stays home, and therefore who gets the chance to accumulate more points. A self-reinforcing loop. The draw is the fourth component. Draw difficulty, the chance of meeting a nemesis early, and the organiser's separation of players from the same federation into opposite halves — all shape the final result in ways the scoreboard never records. An easy draw can produce a final that looks very different from the true level of the two players. No event is named in the input. No tier. No date. No draw. The time-sensitivity field is marked unassessed. For a sport where everything depends on the calendar, the absence of dates is not a minor gap. It is the reason this layer cannot run. Competitive landscape: China and the rest Our framework separates men's and women's events, and further separates doubles and mixed doubles, because the openness of competition differs sharply between these groups. Analysing them as one block produces a wrong conclusion in every group. At the macro level, recent Olympic cycles show one stable pattern. In Paris in 2026, China took all five table tennis gold medals. Three years earlier, in Tokyo, they lost exactly one, in mixed doubles, when a Japanese pair took gold. The difference between four and five does not tell us whether the gap narrowed or widened. It tells us that in one specific event, on one specific evening, one specific pair played exactly as they needed to. In men's singles, the chasing group is now clearer than ever. Sweden has Moregard, who proved in Paris that a European player can beat the world number one at a major. France has Felix Lebrun, a pimples player who presents a completely different spin problem to opponents. Brazil has Hugo Calderano, who has gone deep at many events and represents an entire continent. Japan has Tomokazu Harimoto, who appeared at the top level while still very young. This picture does not show the collapse of a hegemony. It shows the number of countries capable of producing a player good enough to reach a major semifinal is rising. In women's singles, the picture is far flatter. Chen Meng won the women's singles at both Tokyo and Paris. Sun Yingsha holds world number one and won mixed doubles gold in Paris. The depth of China's women's squad is such that a player ranked third domestically is still a medal contender at any international event. This asymmetry between the two fields is one of the most writable aspects of modern table tennis, and it is visible only when the two groups are separated. That is why the framework demands separation. This layer also returned whitespace. No federation named, no event line named. It cannot be assessed. Rules and governance Every rule reform in table tennis is sold with a beautiful justification: more spectacle, shorter duration, a wider audience. But history shows each change redistributes advantage in a specific direction, and that direction can be predicted in advance. Increasing the ball diameter from 38 to 40 millimetres in 2026 reduced speed and spin, lengthened rallies, and shifted advantage towards players with physical endurance and a rallying game. Cutting games from 21 points to 11 in 2026 increased variance, gave each point more weight, and made upsets more frequent. Banning the hidden serve in 2026 removed a source of free points for players who lived on their serve. Banning speed glue in 2026 reduced the speed and spin of loop drives and forced a group of players to rebuild their game mid-career. The switch to the plastic ball from 2026 changed bounce and flight path in ways many defensive players needed years to absorb. The lesson is not whether reform is good or bad. It is that when reading any reform proposal, the right question is not whether it makes the sport more attractive, but which group of players it moves advantage away from, which group it moves advantage towards, and over what timeframe. On internal governance, table tennis has an under-discussed peculiarity: selection criteria for major events in countries with high competition density usually run on quantified points, but always leave room for coaching-staff discretion. That room is where every controversy is born. A quantified criterion can be argued with numbers. A discretionary decision cannot. Here I must state my professional view clearly, and I will state it through an example closer to Vietnamese readers. In football, the subjective judgement space in video-assisted refereeing is larger than people think, because the concept of a clear and obvious error is itself vague. Table tennis has an equivalent, differing only in name: the edge ball, and the service fault. Both depend on the human eye, at moments when a millimetre and a tenth of a second decide the point. When a rulebook creates a vague concept, every interested party can cite that rulebook to defend the conclusion it wants. This layer has no input data. No rule named. No personnel decision named. It cannot run. Coaching staff and talent pipeline This layer decides results over the next one to ten years, and it is the most ignored layer in daily coverage. A pipeline profile must answer four questions. What is the age structure of the current senior squad, and how many pillars will pass their peak within three years. What is the conversion rate from youth to senior level, measured as how many players per cohort reach the top group. How stable is the coaching staff, measured by the number of leadership changes in a four-year cycle. And what is the internal power structure: who decides entry slots, and who decides who trains with whom. At the international comparison level, three development models coexist. The first relies on local sports-school systems with enormous intake numbers, filtered down through multiple tiers. The second relies on dedicated national training centres, concentrating resources on a small group very early. The third relies on professional club systems in Europe, where players compete weekly in a national league and earn a living from it. These three models produce three kinds of player. The first produces depth, because large numbers always produce a few outstanding individuals. The second produces peaks, because resources are concentrated on few people. The third produces durability, because players compete continually under real win-or-lose pressure. No model wins outright, and that difference explains why the world competitive landscape moves in cycles rather than in a straight line. Here I allow myself to address Vietnamese table tennis, because it is the part domestic readers care about most and the part with the least public data. Vietnam's table tennis still stands outside the group of nations with permanent places in the main draws of the international event system. That does not mean we lack technically gifted players. It means we lack top-level matches per year for each player, and in table tennis, the number of top-level matches per year is a more important variable than the number of training hours. A player training two thousand hours a year but playing ten matches against peers will progress more slowly than a player training twelve hundred hours but playing forty matches. This layer has no data. No team, no coach, no cohort named. Risk surface This is the section I agonised over most, because it touches an area where public data is very often managed. Risk in table tennis breaks into six groups. Competitive risk covers injury and mid-cycle technical change. Selection risk covers points-defence pressure and internal slots. Generational risk covers a senior squad ageing faster than the handover rate. Governance and public-opinion risk covers selection controversies and pressure from fan communities. Systemic risk covers calendar density and the travel burden between continents. And opponent risk covers the emergence of a new player from a country that has never had one. Injury is the group I track most closely, and the group where I trust official announcements least. A player's return timeline is typically managed by the team's communications apparatus, and the phrase waiting until the weekend usually means the injury has not healed, only that it does not yet need to be announced. In table tennis, a more reliable signal than any press release is the decision to withdraw from an event already entered. A name disappearing from an entry list is hard data. An interview answer is not. This is where an empty sheet becomes valuable. Absence is also data. A player not entered for an event that history shows he always plays is a signal. A late withdrawal is a stronger signal. And a player who does compete but is placed in an unusually easy section of the draw may also be a signal about physical condition, because coaching staff often want to reduce the number of matches a player must play. This layer has no input data, and here I must add one thing about this analysis itself. The largest risk in tonight's document is not in any group above. It is that an analytical pipeline received an empty input and might keep running anyway. If I do not stop here, that null result will flow into another report, then another, and at some point someone will cite it as an expert conclusion. Public narrative and expectation Every moment in a season has one story louder than the others. The analyst's job is not to retell that story but to measure how much numerical pressure it can withstand. Three checks. Does the story rest on a real foundation, and is that foundation a long run of results or a single match. Is the sample size sufficient, measured in matches and in distinct opponents. And how long will the story survive, measured in weeks, before a contrary result appears. The second part of this layer is the expectation gap. What the market expects, what objective reality shows, and how wide the gap between them is. In table tennis that gap is widest at two moments: immediately after a major, when every conclusion is inflated, and immediately before a major, when every prediction is pushed to an extreme. Here lies a feature of modern table tennis that I consider more important than any other, both analytically and ethically: the emotional intensity of fan communities. A player can become the centre of a large volume of emotional social-media content, and when that happens, media heat detaches entirely from performance fundamentals. The ratio between the two is an indicator I track, and when it crosses a certain threshold, all data about that player becomes harder to read because too much noise surrounds it. A final point concerns sourcing. Analysing public narrative is only feasible if one can distinguish mainstream journalism from self-published content from advocacy produced by an interested party. The input had no source, no date, no article type. It cannot be analysed. Industry transmission The last layer describes how a small event in the arena spreads into a market change. The chain has three segments. Upstream is equipment, youth development and coaching methods. Midstream is events, federations and clubs. Downstream is media, commerce and derivative markets. An upstream example: when the ball changes material, sales of certain rubber and blade lines shift with it, because changed feel sends recreational players looking for new configurations. A midstream example: when a new event system launches and changes how points accumulate, players' schedules change, and therefore the travel, coaching and medical costs of an entire team change. A downstream example: a player who wins gold at a major lifts the commercial value of himself, his club, and the equipment brand he uses. There are contracts people laugh at, until the numbers tell their story. In the equipment industry, a player's value is measured not by the number of titles but by how many recreational players buy that player's exact blade line in the twelve months after a title. That indicator appears on no ranking list, and it is the indicator market managers like me track. This layer cannot run either, because no triggering event is named at any segment. A counter-intuitive angle: the value of a null result There is a professional temptation I have watched take down a great many good writers. It comes from a belief that a good analyst is one who always has a conclusion. That belief is wrong, but it is generously rewarded, because a piece with a firm conclusion always spreads faster than a piece saying the data is not enough. Tonight I have a rare chance to do the opposite. An empty spreadsheet is the most honest document in the entire production chain. Any number I could write from memory is a loan taken against someone else's credibility, and that loan comes due exactly when I can no longer repay it. When I write that a player has a high service-point win rate, readers cannot check it, but a coach can. My confidence interval is far narrower than I imagine, and narrowest precisely where I feel most certain. The interesting thing is that in table tennis, absence often speaks louder than presence. A win tells us what a player can do on one evening. A withdrawal tells us far more about true physical condition. A gap between two events tells us about the training programme. And a blank column in a statistics table can tell us that the person keeping the records simply did not work. That is the biggest difference between a data report and a commentary. A commentary is never empty, because it only needs an opinion. A data report must have a source field. I must also warn myself about another habit, the hunt for indicators that break every pattern. It is a form of professional addiction that data people fall into easily, because it is rewarded with attention. But before publishing a counter-intuitive indicator, I force myself to answer one question: if this indicator is correct, would the opposing coach act differently. If the answer is no, then that indicator serves only the writer's ego, not the understanding of table tennis. The empty stadiums of 2026 taught me that table tennis has parts that make no sound. Tonight's whitespace is the same. It is not evidence that table tennis has nothing worth writing. It is evidence that our pipeline dropped something, and what it dropped was the source. My prediction model has no heart, and that is why it never gets hurt. But a writer has a heart and does get hurt, and that hurt is the only thing that stops him from inventing data. The closing signal for the next cycle I will not end with a summary, because summaries belong to people who already have enough data. I will end with what needs tracking, and what I will check before opening any spreadsheet next time. First, the date field. From now on, a table tennis file without a publication date will not enter the analytical pipeline. This is the cheapest change with the highest return, because every layer downstream depends on it. A rolling 52-week ranking does not forgive temporal ambiguity. Second, the entity field. A pipeline is only valid when the entity field contains at least one name, and that name must pull in at least two concrete information points. Without a name, six of the nine analytical layers are blocked simultaneously. Third, the source field. Before analysing public narrative, the writer must record whether this is mainstream journalism, self-published content, or a document with interests behind it. Skipping that step turns any narrative analysis into disguised guesswork. Fourth, and this is what I want readers to carry away. When you read a table tennis analysis, look for the sourcing section before the conclusion. A piece that states clearly it does not know something is more useful than a piece certain of everything that cannot show where its figures came from. Numbers do not need to be acknowledged. They need to be read in the right place. As for the spreadsheet at 3:47 a.m., I left it exactly as it was. I added one line to the notes column: source not provided, analytical result null, recommend re-running extraction on the original text. Then I saved the file, closed the machine, and went to sleep. It may be the only piece of my career in which the most important part is the part that asserts nothing. I do not need a press conference to prove I understand table tennis. I have 64 matches in my laptop. But to read those 64 matches, I need to know what year they were played in.