Nine Layers of a Table Tennis Match Analysis — and the Cost of the Empty Ones

Game seven, 9-9. The server tosses the ball above an open palm, and in the roughly four-tenths of a second that follow, a decision is made that nobody in the arena can read: topspin or backspin, long or short, into the middle or into the left corner. The ball touches the table, the opponent miscues the return, applause erupts like a pressure valve. The match ends. Twelve hours later, four hundred articles appear on sports sites. Three hundred and eighty recount the score. Twenty talk about nerve and big-match experience. None talks about the choice of serve at 9-9.

The applause fades, but I hear the athlete's breathing more clearly — and the lie of the scoreboard.

I stay behind. Not to write another celebration. I stay behind for a more uncomfortable reason: after forty-two years of watching sport, I have built myself a nine-layer analytical frame, and when I applied it to table tennis, eight of the nine layers returned empty. Empty not because I was lazy. Empty because the public data of this sport does not exist at the layers I need.

That is the real story of this article.

The nine-layer frame

I learned to take a match apart in football. In 2026, at forty-nine, I wrote three thousand two hundred words on Manchester United's 1-1 draw with Liverpool at Old Trafford, arguing that Mourinho's defensive 4-2-3-1 was a deliberate space-denial system built on forty-seven recoveries in the middle third. I was attacked hard. I did not retreat — I live-streamed for two hours, redrew nine tactical situations with a simulation tool, and challenged readers to rebut me. Four days of argument, twelve thousand views. From then on I understood: to argue, you need evidence; to have evidence, you need a data layer.

In 2026, when Germany went out in the World Cup group stage in Kazan, I sat for six hours in a rented room in Busan re-watching the full ninety minutes and found that Germany's back line pushed up an average of sixty-one metres but recovered at only four point two metres per second — one point eight metres per second slower than against Sweden. I wrote Germany died of pride, not of Son Heung-min. The Korean football blogging community was furious. But I had the number.

In 2026, when global football froze under COVID-19, I built a series called Football in the Laboratory, using data from fourteen Bundesliga matches played in empty stadiums to test a hypothesis about crowd pressure. I found home sides' pressing success rate fell twelve point seven percent. A K League data analyst called it over-reach from a small sample. He was partly right. I kept the lesson and added a Limitations of the Analysis section to every piece I have written since.

Those three episodes taught me one thing: every tactical conclusion must be anchored to a countable unit of information. Without that unit, what remains is only tone of voice.

So when I moved into table tennis reporting for the Korean market, I built a nine-layer frame: technique and equipment; player data and head-to-head; event system and points rules; competitive landscape; rules and governance; coaching staff and talent pipeline; risk surface; public narrative and expectation; industry transmission. Nine layers, each a different question asked of the same forty-millimetre ball.

Then I ran the frame against a real match. The result is why this article exists.

Layer one: technique — the only layer with meat on it

This is the one layer where amateur viewers and specialists both see something. Table tennis has a very clear technical architecture: serve, receive, the first three shots, then the long rally. Each of those four phases can be measured by its own point-win rate. A player who wins sixty-five percent of points in the first three shots but only forty percent from the fifth shot onward is a player with a system — and that system says precisely what he is afraid of.

Equipment is a sub-layer but not a detail. Pimpled rubber outwards is capped at two millimetres; sandwich rubber at four millimetres including sponge — ITTF figures, and not trivia. A quarter-millimetre more sponge changes trajectory, changes spin, changes the dwell time of the ball on the rubber. From September 2026, the ITTF banned volatile-organic-compound glues, ending the era of speed-gluing before every match. Then the thirty-eight-millimetre celluloid ball gave way to the forty-millimetre plastic ball, and later to forty-plus. With each change, an entire technical generation had to be rewritten.

I always attach a three-to-six-month window to any equipment change. Inside that window, results fluctuate without reflecting true ability. Any journalist who reads results inside that window as truth is reading noise.

But here is where layer one breaks: point-win rates by phase are not published. There is no table. There is no open database. Scores per game are published, sometimes point totals, and almost never point structure. I can watch and count by hand — and I have, many matches, with pencil and paper — but hand-counting one match does not produce a sample. One match is an anecdote. Three hundred matches is data.

The bus is parked in front of the table — and I start questioning both the man holding the bat and the man holding the camera.

Layer two: player data — the numbers exist, but they do not speak

The WTT ranking runs on a fifty-two-week rolling window: only the best results of the past year count, and when an old result expires, its points vanish from the total. This creates what I call points-defence pressure. A player holding last season's title points walks into the same event this year with two options: defend or lose. There is no third option.

That is an elegant piece of design, and it is one of the few transparent things in this sport. But it answers only one question: who is where. It does not answer the more important question: why.

The ranking is a promise the athlete will break; I merely witness the betrayal.

What I need at layer two is the win rate against players from other associations, broken down by opponent type, split between majors and minor events. I need performance in deciding games and at clutch points. I need to know what percentage a player wins when trailing in the seventh game. Those numbers exist in coaches' heads, in team spreadsheets, in closed meetings. They do not exist anywhere I can cite.

So what happens? Fans get a substitute: raw head-to-head. Who beat whom how many times. That answers a meaningless question — who won more — and skips the meaningful one: by what method, under what conditions, and are those conditions repeatable.

Layer three: the event system — where the rules rewrite the story

Table tennis has three majors: the Olympic Games, the World Championships, and the World Cup. These carry the heaviest weight in both ranking points and legacy. Below them sits the WTT structure — Grand Smash, Champions, Star Contender, Contender, Feeder, plus the WTT Finals. Each tier has its own points gradient, and that gradient is not administrative trivia. It is a behavioural steering tool.

A twenty-five-year-old at the threshold of the top ten chooses a completely different calendar from a thirty-two-year-old defending a top-five position. The first needs points, so he plays a lot, including small events, accepting early exits at big ones to protect his body. The second needs selectivity, so he withdraws — and every withdrawal becomes a question about motivation in the press.

The same act, read against two different players, produces two opposite stories. And there is no public data to tell them apart.

Draw structure works the same way. Seed distribution, the chance of landing in the same quarter as a nemesis, the organiser's separation of same-association players in the early rounds — all of it creates a map the viewer cannot see but the player lives inside. I can redraw that map once the draw is published. I cannot redraw the pressure it creates before the ball is in play.

Layer four: the landscape — the map exists, the coordinates do not

World table tennis has a clear tier structure anyone who follows it long enough can feel. China sits at the dominant tier. Japan, Korea, Germany and a European cluster form the chasing group. Emerging programmes such as Brazil and France produce individuals capable of breaking the structure at the level of a single match.

I can describe that structure in words. I cannot quantify it, because no table publishes the title rate at the three majors over the last five editions, the share of the world top ten by association, or the depth of the under-21 cohort by country.

And here I have to be blunt with myself: any article about the table tennis landscape without those three numbers is an article built on memory. My memory is good. It may also be two seasons out of date.

The Germans do not redraw the tactical map; they burn the old one and call it illumination. I learned that. But the Germans burned the map after they had measured. Burning the map without measuring is burning the house.

Layer five: rules and governance — the grey zone has an owner

The ITTF service rule is very specific: the ball must be tossed at least sixteen centimetres from an open palm, above the level of the playing surface, behind the end line, and the free hand must be removed before contact. Every detail exists to prevent a specific behaviour: hiding the ball, serving from inside the body, serving with a low toss.

But enforcement depends on the umpire, and the umpire is human. That is a grey zone with an owner. A player serving at the edge of legality benefits in matches without a dedicated service judge and suffers in matches with one. Same motion, two outcomes, two different seasons.

I have no data on the frequency of service faults called by umpire, by event, by player. Nobody publishes it. And because nobody publishes it, every umpiring argument in table tennis ends in sentiment.

At a higher level, national-team selection is the most consequential and least visible governance variable. What percentage of places is decided by published points and what percentage by coaching discretion determines the entire ecosystem downstream. I do not have that ratio for any association.

Layer six: the talent pipeline — easiest to talk about, easiest to get wrong

The career age curve in table tennis is steeper than in most sports. Female players peak roughly one to two years earlier than male players. Under twenty-two is the rising band, twenty-two to twenty-eight the peak band, twenty-eight and beyond the veteran band where load management begins. Those markers are not absolute, but they are stable enough to read against.

The pipeline problem is not at the top. It is in the gap between twenty-three and twenty-six. That is the band where a player has left the junior ranks, has not broken into the senior line-up, and starts losing patience. When a team has too many veterans at the top and too few people in that band, the consequence does not appear immediately. It appears three years later, when two veterans leave in the same season.

I can point at that gap by feel. I cannot point at it with numbers, because nobody publishes the age distribution of the senior squad, the junior-to-senior conversion rate, or the number of international entries allocated to the under-21 group.

One thing I remind myself constantly: a generational gap is a slow phenomenon. It does not make news. It only makes consequences.

Layer seven: the risk surface

When I build a risk matrix for a player, I split it into six categories: competitive, selection, generational, governance and public opinion, systemic, opponent.

The most dangerous category in table tennis is the first, and it has a specific form: a concealed injury inside a technical-overhaul window. When a player has just changed rubber, restructured the serve, and is carrying a mild shoulder or wrist issue, those three variables compound into a run of bad results that outsiders read as decline. Insiders read as transition.

I have seen this pattern many times in football — a player changes position, changes system, and is judged on three games. In table tennis it is harsher, because the margin of error is smaller: a one percent change in spin is enough to put the ball off the table.

Risk two is being decoded by video. Every player has a handful of serve patterns that repeat under pressure. An opponent with three weeks of preparation will find them. An opponent with three days will not. A congested calendar is an invisible shield, and it appears in no statistical table.

Layer eight: narrative and expectation

Every player lives inside a heat cycle. There are periods when he is the most talked-about name in the sport, and periods when he vanishes from the front page even though his level has not changed. That cycle runs on crowd emotion, and crowd emotion runs on small samples.

One win over a strong opponent at a mid-tier event generates a wave of narrative. Three ordinary results later, the wave recedes. Nothing has changed in ability. Only the denominator has changed.

I use one test on every article: is what I am praising a single event or a season-long pattern? If I cannot answer with data, I have to answer in prose — which means I have to admit I am speculating.

This is where table tennis data irritates me most. The ratio of heat to fundamentals cannot be computed, because the fundamentals are not published. People can count views. They cannot count quality.

Layer nine: industry transmission

A table tennis match does not end at the table. It flows into the equipment market, into the grassroots training base, into a player's commercial value, into policy and capital.

But that flow is measurable at only one end. At the equipment end, a major brand publishes quarterly sales, and those sales respond to an ambassador's image with roughly a two-quarter lag. At the player end, endorsement contracts are undisclosed. At the event end, broadcast rights fees are undisclosed. I have one third of the map.

I keep one rule at this layer: separate competitive value from commercial value, and never use the second to infer the first. A player who is talked about more is not a player playing better. In every sport those two curves are out of phase. In table tennis they are further out of phase, because match data is thinner.

The counterintuitive angle: opacity is the product, not the defect

The default reaction to eight empty layers is to blame the ITTF, the WTT, the national associations for not publishing data. That demand sounds reasonable. It is also naive.

Table tennis sells something very specific: the feeling of being unreadable. A serve the crowd cannot understand — and therefore cannot explain why the opponent miscued — is the core product of this sport. A rally a viewer has to watch three times to grasp is the entertainment value. If every serve were encoded, tagged, and published with a point-win rate by spin type, the sport would lose the thing it is selling.

Opacity is not an operational failure. It is product strategy.

Which gives the analyst his consequence: demanding complete data before writing is demanding that the sport destroy its own product. Nobody will comply. If I wait, I write nothing.

So the real skill is not filling the eight empty layers. It is building judgement inside the gap, and stating clearly where you are standing when you judge. An honest table tennis analysis should look like a map with regions explicitly marked as unsurveyed — not a map coloured in with fake paint.

The double-decker bus: the lower deck is what I know, the upper deck is what I infer. I keep the upper deck very light. If the upper deck outweighs the lower one, the bus tips over.

Limitations of the analysis

I have to state three limitations of this article itself.

First, it issues no conclusion about a specific player, a specific match, or a specific ranking. I do not have the information unit to do that, and I refuse to fill the gap with memory. Memory is a legitimate source for hypotheses. It is not a legitimate source for conclusions.

Second, the nine-layer frame is the product of a football man moving into table tennis. I carry the bias of my old trade: I believe in spatial structure, in measuring distance, in counting. Table tennis has dimensions that reach the limit of that method — spin, wrist tension, the instant of contact. I have no instrument for those. I have eyes and patience.

Third, and most importantly: eight empty layers do not mean this sport cannot be analysed. They mean the analyst must accept working with an incomplete dataset and must say so at every step. That admission once felt to me like a lack of confidence. Ten years on, I consider it the only reliable sign of critical thinking.

Data never lies, but it chooses who gets to tell the truth — I learned to become that person.

A test for the next event

I will not close with a summary. I will close with a test I intend to apply to myself at the next tournament, and which I suggest readers apply to every piece of table tennis commentary they read.

Before writing a single word about a match, write down three things you cannot know about it. Three, specific, no generalities. If you can list them, you are analysing. If you cannot, you are storytelling — and storytelling does not need a scoreboard, only a voice good enough that nobody asks a follow-up.

As for table tennis, I will keep counting by hand, with pencil, with paper, until somebody publishes a real dataset. And if nobody does, I will still stay behind after every match, in this room, with nine blank layers and one that has words on it.

That one layer is where I work. And it is why I am still writing.