Digging Into the Empty Layer: Table Tennis, Data, and the Cost of a Broken Analysis Chain

Digging Into the Empty Layer: Table Tennis, Data, and the Cost of a Broken Analysis Chain

Three in the Morning, and a Blank Page

At three in the morning in Shenzhen, I opened a file a colleague on the data desk had sent over. The filename was precise: "Stage-Two Deep Analysis — Table Tennis Domain." I had been waiting two days for it, after filing a long report on a regional youth qualifier. I opened it, and in the blue light of the screen I saw a table with nine rows. Each row held an empty box. Beside each empty box was a small line of text: "insufficient information."

I sat still for a while. Not out of anger. I sat still because I had seen this exact scene before, only with different skin on it. It was not a technical error. It was a habit.

In my trade, there is a temptation larger than fabricating numbers. It is presenting a complete analytical skeleton with nothing inside it. A beautiful frame, split into nine layers, with tables, arrows, risk ratings, and a "confidence: high" field — and at its centre, not a single player's name. Someone reading that report without checking will believe they just read an analysis. In truth they read an empty shelf.

I am not writing this to tell the story of a broken file. I am writing because that empty shelf is sprouting everywhere in the sports industry, especially in table tennis — a sport where everything is measured to a tenth of a second, a spin rotation, a centimetre. And the more that can be measured, the easier it becomes to forget that what to measure is the harder question.

Context: The Information Ecosystem of Modern Table Tennis

Table tennis holds a structural paradox. It is small enough that you can play it in an apartment. It is also complex enough that a single point can contain twenty technical decisions in under four seconds. That compression makes the sport's data both extraordinarily rich and extraordinarily easy to misread.

Picture the information chain of an international event. At the lowest layer sits raw data from the table: scores, rally lengths, rally durations, service-point conversion. A little higher sits technical data: spin type on serve, placement, footwork direction, ball speed off the racket. Higher still sits tactical data: win rate when leading, win rate at decisive points, tendency in shot selection when the opponent changes rhythm. And at the very top sits the hardest thing to measure: mental state, physical durability across seven days of competition, and what I call a person's "maturity curve."

Every layer has its own workers. Every layer has people selling reports. And because every layer can be presented independently, people started separating them. The raw-data desk sends a table. The technical desk sends a chart. The tactical desk sends a paragraph. The synthesizer — sometimes a news editor like me, sometimes an automated model — receives it all and is expected to produce a conclusion.

The problem starts there. When any link in the chain is empty, what emerges at the end is not a weak conclusion. It is a fake conclusion, delivered in the voice of a real one.

I have tracked youth table tennis for twenty-three years. I have stood in arena corridors waiting for a fifteen-year-old to finish practice, only to ask one question about how it feels to be down a set. I have sat through three full days at an U19 event to see whether a kid's form after two good matches was a geological layer or just a layer of dust. I mention this not to boast. I mention it to state something simple: my job is digging, and a digger has to know where he is digging. A machine can hold a shovel with correct technique, but it cannot tell fresh fill from foundation.

Layer One: Technique, Tactics, and Equipment

Modern table tennis has two broad technical streams running in parallel. The first is the two-winged looping game, built on topspin, pushing the opponent back to mid-distance and finishing with a forehand loop from mid-range. The second is the close-to-table speed game, built on early contact and redirection, breaking rhythm before the opponent can set up. Each stream has its own set of metrics, and this is where empty analyses fall into their first trap: they apply one metric set to two different technical streams.

A concrete example. Ball-placement accuracy — the share of shots that land where intended — does not mean the same thing for the two streams. For the mid-distance looper, a high rate is a condition for survival, because every miss is a ball the opponent attacks from above. For the close-to-table speed player, ten percent of misses may be a fair price for twenty percent of direct winners. Put both rates in one table and compare them without classifying the playing style, and you are comparing the temperature of two cities with one thermometer while forgetting that one is indoors and one is outdoors.

In my own notebook, each player is filed into one of four style groups before any metric is examined. Group one: two-winged topspin, mid and long distance. Group two: close-table speed, placement first. Group three: defensive counter-attacking, chopping and blocking. Group four: all-round, no fixed stream, opponent-dependent. Only after classification do I open the metric table. This is the step automated pipelines tend to skip, because classification requires watching video, not reading numbers.

On equipment, this is the layer mainstream coverage is almost always lazy about. A racket has two parts: the wooden blade and the rubber. Rubber comes in two main families — smooth, for generating spin, and pimpled, for killing spin or reversing it. When a player switches from smooth to pimpled rubber on one side, their entire skill chain changes over six to ten months. During that window every metric worsens, but not because they got weaker. They are relearning a new language on the same arm.

I once watched this happen at a youth club in the south. A sixteen-year-old with a win rate above seventy percent dropped below forty for two consecutive quarters after switching rubber on his left side. Four domestic reports called it "decline." Nobody checked whether he had changed equipment. Nobody asked the coach. The analysis rested on an empty field labelled "equipment factor" and filled it with two words: "poor form."

That is why I always put the equipment question first in any interview, before tactics. Which blade, which rubber, what thickness, what hardness, changed when. If the answer is uncertain, I record "undetermined" and refuse to use any metric from that period to judge ability. A metric without equipment context is a stone without stratigraphy. You can hold it, but you cannot tell how old it is.

At the purely technical layer, one metric matters most to me and is used least: the conversion rate from serve to the third ball. In table tennis, the serve is the only shot you fully control. The third ball — the first shot after the opponent's return — is where tactics are exposed. If a player wins on serve but loses on the third ball, they are serving beautifully without a plan. If they win the third ball at a high rate, they are playing a structured game.

I have counted this by hand for several young players across the last three events. The method is simple: a notebook and a screen. For each player I keep three columns: direct service winners, third-ball wins, third-ball losses. After one match I have a clearer picture than any percentage table. The gap between column one and column two tells me whether the player is living on luck or on planning. The gap between column two and column three tells me whether they have the patience to keep playing after the serve.

Layer Two: Players, Rankings, and Head-to-Head

The world ranking in table tennis is one of the most misread points systems in sport. A ranking point does not measure current ability. It measures accumulated results over a fixed window, then expires gradually. A player can sit fifth in the world while performing at a fifteenth-place level, simply because they defended old points while others lost theirs.

I call this "points-defence pressure." Each player holds a bag of points. Every event they enter, an old result is withdrawn and a new one added. If they do not enter, the bag is still drawn down on schedule. So in the three to six months after a major event, a group of players is forced to compete continuously not to rise, but to avoid falling. This group is the most injury-prone, the most form-fragile, and the group most often labelled by media as "in crisis."

In my notebook, every player has a column called "points state." I mark three colours: green if the bag is growing, yellow if holding, red if draining. A player performing well with a red bag is playing with a stone tied to the ankle. Their results cannot be used to judge ability. A player performing poorly with a green bag is someone with time. Their results cannot be used to conclude the opposite either.

On head-to-head records, this is where empty analyses do the most damage. A head-to-head table can look scientific with three columns: total meetings, last two years, and majors. But if you read only total meetings, you are reading history without reading time. Table tennis moves so fast that a match-up from three years ago has almost no predictive value. The blade changed. The rubber changed. The coach changed. The fitness changed. The psychology changed.

I have a rule I set myself years ago: count only match-ups within eighteen months, and separate major-event meetings. The reason is simple. Pressure at an annual tour event and pressure at an Olympic Games or a World Championship differ in kind, not just degree. A point in qualifying and a point in a semifinal carry the same score on the board, but not the same weight on a player's spine.

When assessing a player, I always split opponents into three groups: same nation, same continent, different continent. The win rate against each group tells a very different story. Winning heavily against compatriots but losing to foreign opponents signals a player who understands teammates but cannot read strangers. Winning heavily against foreign opponents but losing to compatriots signals a player locked by the people he trains with every day. These are two entirely different problems requiring two entirely different fixes.

In table tennis, one metric strikes me as the true measure of nerve: win rate at decisive points — the tenth point or the final point of a game. This metric is less technical than you would think. Two players can share technique, fitness, and tactics, yet at a decisive point one chooses a safe serve and the other chooses a risky one. That difference is not in any textbook. It lives in how a person has made peace with their losses.

Layer Three: The Event System and Points

The professional table tennis calendar now runs as a year-round series across tiers, plus the symbolic majors: the World Championships, the World Cup, and the Olympic Games. At the bottom sit small events with little points and little prize money, used for accumulation and experimentation. In the middle sit events with meaningful points, enough to shift ranking positions. At the top sit events where a single match can shape an entire four-year cycle.

One thing rarely said: in table tennis, a major event is not only a stage for performance, it is a mechanism for allocating resources. A good result at a major brings more than a medal. It brings training-camp slots, sponsorship, and a place in the squad for the next cycle. So when I write about an event, I try to place it on two axes: the points axis and the resource axis. The first is computable. The second is not, and precisely because it is not computable, it is often ignored.

A complete event analysis must answer specific questions. How many points does the champion earn, what is the prize money, and most importantly: where does this event sit in the four-year cycle. An event six months before the Olympics has entirely different value from one right after it. Before the Olympics, strong teams hide their cards. After the Olympics, they restructure. Same event, same entry list, two contexts, two readings.

I have been wrong because I ignored this time axis. Years ago, I underrated a player because of poor results at an event immediately after an Olympics. I read it as decline. In fact he was rebuilding technique after changing his playing style, and that event was a test recorded in the record books as a failure. Three months later he won three consecutive events.

On draws, this is the most mystified part. A bracket called "hard" or "easy" is usually based on feeling, not data. The serious approach assigns each potential opponent a specific nuisance index — not ranking, but stylistic fit. A low-ranked opponent whose style counters you can be more troublesome than a high-ranked opponent whose style suits you. I call this the "real difficulty" of a bracket, distinct from the "nominal difficulty" computed from rankings.

After years of watching, I have found that most upsets at majors do not come from a good player exiting early. They come from a good player meeting exactly the opponent their style has no answer for. The draw does not create results. It only opens a road to a result already sitting in the file.

Layer Four: The Competitive Landscape and China versus the Rest

World table tennis over the past two decades has had a clear structure: a near-absolute leading tier, an unstable chasing group, an emerging group, and the rest. The leading tier for years was China, with a depth so deep that a player ranked sixth domestically could be a medal contender internationally.

But that structure now shows small cracks, and this is where I want to spend attention. A few names from outside China have proven the gap is no longer a wall. A Brazilian player has spent years inside the world's top ten, something once considered impossible for a country without a table tennis tradition. A Swedish player reached a World Championship final and won Olympic silver, breaking the assumption that Europeans could only defend. A French player using the traditional penhold grip won an Olympic medal, proving a supposedly outdated technique can live if its footwork is modernised.

Japan is the most instructive case. It built a structured youth development system with regional training centres and an early international competition programme for teenagers. The result was a generation arriving three to four years earlier than the standard timeline. One player won an international event at fourteen, something the Chinese system, with its strict standards, rarely permits.

My point is not that China is weakening. My point is that a monopoly model is becoming a multipolar model, and that shift demands a new way of reading. In the old model you compared one team with the rest. In the new model you compare multiple development systems, and the question becomes harder: which system produces the most top-level players in the shortest time, with the lowest attrition.

I have never seen an analysis table answer the attrition question. People count successful players. Very few count the ones left behind. In any development system, every successful athlete usually comes with a large number of departures nobody can name. Look only at the visible part and the system seems efficient. Dig into the submerged part and a different price appears.

Layer Five: Rules and Governance

Table tennis has undergone several rule changes in three decades, and each redistributed advantage between player groups. When the ball was enlarged, speed and spin dropped, helping the physical game and hurting the maximum-spin game. When scoring shifted from twenty-one to eleven points, decisive points multiplied, helping mentally stable players and hurting slow starters.

After each change I hear two kinds of reaction. The first is technical: who gains, who loses, how long to adapt. The second is political: who proposed it, why, and who is most affected. The second is often dismissed as speculation, but ignoring it is self-deception. Every rule change has a proposer, and proposers have interests. That does not make the change wrong. It means a full analysis must include both faces.

On governance, this is the area where I write most carefully. Table tennis has three overlapping governance layers: international, continental, and national. Each has its own rules, and they do not always align. A player can meet international criteria but not national ones, or the reverse. Those grey zones are where selection disputes are born.

I once saw a case where selection criteria were published in one vague sentence: "based on results and form." Those five words are an empty shelf. Nobody defined which events, which tier, which time window counted as results. Nobody defined how form was measured. So when the list was published, nobody could verify its reasonableness. The entire debate that followed became an arena of emotion.

The lesson I took from such cases is clear: a standard that cannot be quantified is not a flexible standard. It is an empty standard, and an empty standard is always filled by power. I say this not to accuse anyone. I say it because over years of writing about youth sport, I have seen too many talents eliminated by a standard with no definition.

Layer Six: Coaching Staff and the Development Pipeline

A table tennis team has three staffing layers that directly affect results. First, the head coach, who sets philosophy and overall strategy. Second, the personal coach, who works with one or two players daily. Third, the analysis and fitness staff, who never appear on television but decide whether a player holds form across a long season.

I pay special attention to the second layer. In many table tennis nations, a personal coach stays with a player for years, sometimes from age twelve to twenty-five. That bond creates technical stability but also a risk: when the player has outgrown the teacher, they can be held inside an old model. I call this "the teacher's ceiling." It is nobody's fault. It is a natural law of long-term teacher-student relationships.

On the development pipeline, I picture it as a water pipe system. At the source are talent classes and sports schools. In the middle are youth squads and youth tournaments. At the end is the national team. What deserves study is not the far end. It is the share of water that makes it through each segment.

I once tried to measure this in a small system. I started with a talent class of thirty-two children. Three years later, eleven were still training. Five years later, four still competed. Seven years later, one still competed at national level. That number does not say the system failed. It says the system has a specific conversion rate, and that rate must enter every calculation about that nation's table tennis future.

On the internal ecology of a squad, one issue is rarely analysed properly: internal competition. In a squad with too many talents of the same generation, each must share international opportunities. That can be positive — but it can also be negative, when some players are buried too long in reserve roles until their development cycle is delayed by years. In a sport where the peak often arrives between twenty-two and twenty-six, a two-year delay can be a whole career.

Layer Seven: The Risk Surface

Risk in elite table tennis is not evenly distributed. Some risks are constant, others appear only at special moments. I split them into six groups.

First, injury risk. Wrist, shoulder, lower back, and knee are the four most common zones. For players under twenty, wrist risk is especially high because training volume rises faster than muscle coverage of the joint. For players over twenty-eight, risk shifts toward back and knee through accumulated hours and slower recovery.

Second, selection risk. This is the least discussed but most silent. An eligible player can lose a slot for reasons unrelated to ability. When that happens to a young player, they lose more than an event. They lose a year of development.

Third, generational-gap risk. When a generation of talents leaves the top together in a short window, a national team can enter a three-to-four-year gap even while the pipeline runs.

Fourth, governance and public-opinion risk. One misstep in communication can create pressure strong enough to shape competitive results.

Fifth, systemic risk. Funding, policy, and shifting priorities at governing bodies can change the fate of a development programme within one cycle.

Sixth, opponent risk. Another nation can produce a style your system has no answer for. This is the most measurable risk, and usually the latest detected.

Layer Eight: Public Narrative and Expectation

In table tennis, the public narrative runs a fairly clear cycle. It opens with discovery: a young player appears, wins a shocking match, and is given a grand name. Then amplification: every statistic is used to reinforce that name, every small detail becomes evidence of genius. Then comparison: the player is placed beside a legend. Finally disappointment: results do not come, and the public starts hunting reasons, usually inside the player.

I have a rule for young editors on my team: do not use the word "prodigy" in a player's first three hundred days under eighteen. Not out of strictness. Because I have watched how that label operates. It does not describe a fact. It creates an invisible contract with the public, and when the contract is broken, the player pays.

A complete public analysis must include a section I call the "expectation gap." Two columns: market expectation on one side, objective assessment on the other, and the difference between them. The larger the gap, the higher the probability of a psychological shock. For a young player expected to win a medal at a major who has never won a qualifying match, that gap is a forecastable risk. Yet we usually only recognise it after the shock has happened.

The Counterintuitive Angle: When an Empty Map Is Presented as a Discovery

This is the part I want to dwell on most.

I have read many analyses in my life. The one that struck me hardest was not dense with data. It had nine sections, each with tables, charts, a risk framework, and confidence markers. And at the centre of each section the content read "insufficient information."

What is remarkable is that such a document can still read persuasively. It has rhythm. It has structure. It has arrows pointing left to right. It has a table of six risk groups, one row each, one empty box each, ending with a line reading "overall risk: cannot be assessed." A hurried reader will not notice there is nothing inside. A hurried reader remembers the frame, not the content, and days later believes they read a full analysis.

I call this the empty-map effect. An empty map is still a map. It has a scale. It has a compass. It has grid squares. It just lacks terrain. And in my trade, a map missing terrain can be more dangerous than a wrongly drawn map, because it raises no suspicion.

I am not writing this to attack any tool. I am writing because I see the same error in many places, at many levels. In some newsrooms, reporters are asked to submit analyses in a fixed template even when data is insufficient. The frame must be full. The boxes must have content. And when boxes must be full but data does not exist, people fill them with language. At that exact moment, analysis becomes literature.

There is another way. I learned it from a senior editor at a newsroom where I worked early in my career. He once told me something I carried for years: "If you have no data, file a page stating where you looked." Back then I thought he was talking about honesty. Now I understand he was talking about method.

A report stating where it looked and found nothing is a valuable report. It tells the next person where to start. A nine-section report full of words but without sources helps nobody. It shifts the burden of verification onto the reader, and the reader in most cases will not verify.

I once wrote a piece about a seventeen-year-old I rated highly after watching two matches. When I finished, I read it a first time, then a second, then a third. Only on the fourth reading did I trust my eyes. But when I reopened the fitness data, I found a detail that did not fit: his maximal oxygen uptake was below the average for his age group. A player with good technique but an insufficient fitness base is a player who can win three matches and collapse in the fourth. I rewrote the piece. I did not say he was weak. I said his schedule needed adjusting.

That piece was pushed down the page. Three months later he was injured and out for nearly eight months. I do not tell this to claim I was right. I tell it because it closes a circle in how I work: data is never the answer, but it is a reason to slow down. And in this trade, slowing down is a skill.

What I Keep

Once, at a youth event, I sat beside a coach who had led teams for more than twenty years. We watched an unremarkable match. Midway, I asked how he knew a child would go far. He was quiet for a long while, then said: "I don't watch the shot. I watch the loss."

I carried that into my notebook. Years later, I began adding a column for every young player, with no numbers. That column records the reaction after a lost point. Some turn away. Some look at the floor. Some look straight into the coach's eyes. Some smile. After a few years I realised that column predicted better than any technical metric I had ever collected.

Of course, I did not drop the data. Data shows only the surface; the submerged part must be dug by hand. Some gems lie too deep for machines to reach. And every generation of players is a geological layer. You must remove the soil before you find the fossil.

But I learned the reverse too. Three pandemic years taught me one thing: nothing is constant. A dataset that held true for a decade can become meaningless in three months. Empty stadiums changed human behaviour, and no table predicted it. Since then I prepare two scenarios for every forecast: one following historical patterns, one following the latest data. If they align, I feel safer. If they diverge, I write plainly that I am unsure.

The last thing I want to keep is not a conclusion about table tennis. It is a principle for working with data in any sport.

An analysis chain is only as strong as its weakest link. If the input is empty, the output cannot be full. If someone hands you a report that looks complete but contains not one name, one specific date, one verifiable source, do the simplest thing: ask where they looked.

The answer will tell you whether you are reading an analysis or a frame. And in table tennis — where everything is decided in under four seconds by a person standing a few metres away — that difference is everything.

I still keep that nine-blank-box report on my machine. I do not delete it. Every time I open it, it reminds me that a digger's job is not to decorate the hole. A digger's job is to find the right place, and to know when to say there is nothing here.

A contract is signed on paper, but decided from the bench. A talent is born in one practice session, but shaped over the next three years. And a table tennis nation is decided not by what it wins today, but by what it can keep after today's generation walks away.