The Silence Between Two Points: What Tennis Data Cannot Measure

The file arrived at 7:12 in the morning, before I had time to brew my second pot of tea. The file name was unambiguous: Stage-Two Deep Analysis — Tennis Desk. I opened it. Nine sections. Tables fully built, columns aligned, cells neatly framed. And in every cell, instead of a figure, the same repeated sentence: insufficient information to assess. Analysis subject: insufficient information. Playing style: insufficient information. First-serve points won: insufficient information. Nine sections, hundreds of cells, one sentence.

I sat still for a long while. Outside, the city was waking up. In here, a machine had just refused to work the way humans rarely dare to: it admitted it did not know.

The strange thing is that I felt no irritation. I felt familiarity. Across nearly thirty years of writing inside sport, I have many times faced a blank page in the literal sense — after a corrupted tape, after a source went silent, after a match was postponed so long that people forgot it had ever been scheduled. But today's blank page was produced by an algorithm, and it was blank with discipline. No inference. No guessing. No embellishment. I asked myself: if that system were applied to most of the tennis copy I read each week, how many cells would be forced to read "insufficient information"?

Tennis has been bound to numbers for far longer than anyone has called it the data era. From the 1970s, the Grand Slams hired people to count double faults, winners and unforced errors by hand. In 2026, Wimbledon and the US Open installed Hawk-Eye, the camera-based ball-tracking system, and for the first time spectators saw a ball's trajectory reconstructed on screen to the millimetre. Eleven years later, in 2026, player-tracking systems began appearing at major events, measuring distance covered, rotation speed, and return position.

Since then, every match has become a mine. First-serve percentage. Points won on first serve. Points won on second serve. Return points won. Break-point conversion. Winner-to-unforced-error ratio. Distance run. Net approaches. Average forehand speed. Broadcasters buy data packages, analytics platforms sell subscriptions, bookmakers build models, junior academies hang charts in their corridors.

That industry runs on a simple faith: what can be measured can be managed, and what can be managed can be improved. The faith is not wrong. It is only incomplete. And that incomplete space is exactly where I still sit, after nearly thirty years, with a paper notebook and a pencil.

There was a pandemic season when the My Dinh stadium held no event at all for 214 days. I was emotionally spent, so I left Hanoi for Hai Phong, closed my door, reread my master's thesis in sociology, and started a newsletter called "The Empty Track", one legendary race each week tied to its social context. By year's end it had 3,200 subscribers, mostly coaches who had lost their training grounds. I learned something that still holds: the empty track is where I hear my own footsteps most clearly. When the stands, the scoreboard and the data are gone, whatever remains is the core.

So what does tennis data see, and what does it miss?

It sees very clearly what happens after the ball leaves the racket. Whether a first serve lands, where the ball travels, which stroke ends the point, who touched it last. It can measure points won on first serve — the metric coaches call the foundation of every tactic on a hard court. It can count return points won, the figure that lifted Novak Djokovic onto a tier of his own in this sport's history: across his career, the Serbian has won roughly a third of points when his opponent serves, a level most peers only touch at peak form.

But the data does not see the instant Djokovic steps half a shoe-length inside the baseline, lowers his centre of gravity, and accepts the risk of taking a 200 km/h serve on the half-volley. It records the outcome of that decision, not the decision. And at the top of tennis, the decision is what separates a great player from a very good one.

I once wrote about this in a profile after a trip to Russia in 2026. During the World Cup semi-final between Croatia and England, I mispronounced Luka Modric's name three times in the first half and was savaged on social media. I retreated to my hotel room and cried for 48 hours, cut off all contact, but still rewatched all five of Croatia's matches. My resulting profile of Modric's "invisible work" — more than 90 km covered across the tournament, 14 chances created from passes no standard stat sheet logged as assists — was shared by Croatia's Sportske Novosti.

Two days in Moscow are enough to understand that football is not only the glare of the stage lights. Tennis is the same, and perhaps harsher, because here there is no teammate to shield you. A player walks out alone, with one racket and a string of decisions stretching across three hours.

Take a metric that looks utterly objective: distance covered. Tracking systems log the total metres a player runs in a match. The figure looks handsome on a broadcast graphic. It is also nearly meaningless in isolation. A player who runs 3,500 metres may be someone dragged across the court by an opponent — or someone opening angles and forcing the opponent to chase. The same figure, two opposite stories. Raw data cannot tell hunter from hunted.

The same applies to speed: you can measure forehand velocity, but not the difficulty of that forehand when the player is off balance, after a long slide on clay, with breathing already broken. The ball still lands in. The point still counts. But its internal value differs entirely from an identical stroke in the tenth minute.

Rafael Nadal is the classic case for this blind spot. The stat sheet records his very high second-serve points won, but it cannot record the feeling of a heavily spun second serve kicking up into an opponent's left shoulder. A returner waiting to receive is not only calculating placement; they are calculating fear. And fear does not appear in any data set I have ever read.

I have sat in newsroom meetings many times, listening to editors argue over which number should drive the headline. A player wins 78 percent of first-serve points — that sounds impressive. But if that 78 percent was produced against a world No. 120, on a preferred surface, in the first round, then the figure tells us very little about what that player will do against the second seed in the quarter-finals. Some data does not need to be loud; it only needs someone patient enough to read it. But some data is so loud that people forget the context that produced it.

This is where I remember Kuala Lumpur, 2026. I was 35, the only female journalist in the track-and-field press area at the 29th SEA Games. I found that Nguyen Thi Oanh won the women's 1,500m through negative splitting — her first 800m slower than her last 700m by 2.3 seconds. When I pitched the tactical analysis to a male editor, he laughed and said women do not understand pacing. I did not argue. I quietly spent three weeks reviewing all the footage and self-published on my personal blog. The piece drew 50,000 views in 48 hours and was shared by the national team's head coach.

Rebellion does not have to be loud; sometimes it is quietly rearranging the numbers. From that day I stopped asking male editors for permission before writing. Every subsequent piece came with a data chart I drew myself. But from that day too, I understood that a chart only has value when the person drawing it knows where to look — and knows what they are leaving out.

Back to tennis. There are three layers of data I distinguish in daily work.

The first is what the scoreboard tells you: scores, ratios, percentages. This is the layer every fan can reach, and the layer most abused. A newspaper that merely transcribes the official stat sheet is not doing journalism; it is performing a transcription function.

The second is what tracking systems record but scoreboards never display: average return position, share of balls landing in the back third, average stroke depth, recovery time between points. This is the layer where coaches actually work, but it demands a patience most newsrooms do not have.

The third is what nobody records: a player's breathing after a love-game loss, the way they look at their towel before a decisive serve, how fast they leave the chair after being broken back. This is the layer I call soulless — not because it lacks life, but because it lacks a chronicler. What is not measured is usually not remembered.

Throughout my career I have always tried to descend to the third layer. That is why I stay in the stands until the ground is swept, listening to rubber soles squeak on the surface. That is why I note how long each player takes to stand up after a lost point.

Once, at a domestic tournament, I followed Ly Hoang Nam from the first round to the final. Vietnam's No. 1 — a SEA Games gold medallist in both singles and doubles — played a match that a stat sheet simply cannot convey. He won in three sets with a first-serve percentage of only about 55, a figure any automated analytics engine would label concerning. But if you sat long enough, you saw him save his best first serves for the right moments: at the tenth game of the deciding set, just after his opponent had broken back and momentum was rising.

The machine does not see that intent. The machine sees only the percentage.

This is where my position formed over many years. I believe most tennis data is used as decoration rather than as a tool. A few numbers are pasted onto a pre-match graphic and called analysis. A record is cited and called depth. A head-to-head tally is counted and called context. But data truly has power only when it answers a question the viewer had never thought to ask: why did the seventh game of the second set turn the match?

Inside the tennis industry, data is squeezed between two opposing pressures. On one side is genuine professional pressure, from coaches and sports scientists who need to know where their player is losing points. On the other is commercial pressure, from broadcasters, sponsors, streaming platforms and bookmakers who need a tidy number to sell. Those two pressures pull data in different directions, and in most cases the commercial pressure wins.

That is why I observe the sports rights market with growing caution. Streaming platforms bid astronomically for rights packages, absorb losses for years, then compensate by simplifying content to court mass audiences. Tennis has one of the most fragmented rights structures on the planet: Grand Slam rights sit with each organising body, ATP and WTA rights sit in separate bundles, and in many markets a fan must pay for three or four services to watch a full season. As costs rise and audiences fragment, content is forced to sacrifice depth to widen its reach.

The result is increasingly identical copy: a few numbers quoted, a few beautiful rallies retold, wrapped in a sensational headline. The hardest part — explaining why a player changed rhythm, why a 180 km/h serve is more effective than a 210 km/h one, why a player accepts retreating half a metre in exchange for time — is cut because it costs time and generates no clicks.

This is the counter-intuitive point I want to make clearly. When the input data file is empty and the machine returns nine sections of "insufficient information", I think the machine is being more honest than most humans. We humans, facing a gap, tend to fill it with story. A player loses and we assign them psychological injury. A player wins unexpectedly and we assign generational talent. A run of three wins and we call it a breakthrough. We do this because gaps make audiences uneasy, and unease does not sell advertising.

The machine does not do that. It says: I have nothing to conclude. And if you ask me, that is the most honest sentence anyone in this industry can utter.

People look at the rankings; I look at what the rankings hide. A player holding world No. 12 for eighteen consecutive months may be impressively consistent, or may be trapped in an equilibrium with no exit. The rankings do not distinguish the two. It is only a number updated every Monday.

To understand why, you must look at points structure. One player may be defending a huge haul from two events in March and April, meaning from May to year's end they can only move upward. Another may have just won a major, meaning for the next twelve months they must defend nearly all those points while rivals lose nothing. Same position, two entirely different pressures. Same number, two opposite seasons.

This is the kind of analysis rankings do not provide. And it is the kind most sports readers are never given, because it requires the writer to spend three weeks instead of three hours.

During transfer windows and pre-season, the noise grows louder still. Players change coaches, agents, schedules. Some change their training surfaces altogether. Social media floods with rumours about who will retire, who will return from injury at a small South American event, who has just signed with a new equipment brand. Most of that information has a very short shelf life.

My filtering method is simple: follow the money, follow the contracts, follow the parties' actual moves rather than their words. When a player appoints a coach with clay-court expertise, that is a structural signal, not a rumour. When a player reshuffles the calendar to gather four consecutive events on the same surface, that is a strategic statement. When a tournament announces a new prize-money structure, that is a signal about real cash flow, reflecting the organisers' expectations for their own future.

Those signals can be read. Rumours cannot. And in the quiet season, the ability to read signals is all that separates an analyst from a reporter.

Back to the story at the top. I have kept that empty file in my working folder, undeleted. Occasionally I open it again, not to laugh but to remind myself that whenever I take on an analytical assignment, I must ask two questions before writing the first word: What do I actually have? And what am I being asked to conclude that there is no basis to conclude?

Twenty-eight years of watching this industry have taught me that the greatest disaster in sports writing is not a lack of data. The greatest disaster is having just enough data to sound plausible without enough truth to actually be right. A nine-section analysis reading entirely "insufficient information" may be judged a technical failure, but it is also a lesson in integrity.

Once, younger and more confident than I am now, I wrote an analysis of a player based on the three most recent matches I had on tape. I constructed a tidy narrative of technical transformation. Three weeks later the player told the press he had been through a family upheaval and had never changed his technique at all. I never corrected that piece. I simply stopped writing that way.

Elite sport is the art of repetition — and of breaking repetition. A player serves a thousand times identically so as to serve one different ball at the exact moment it is needed. A player returns thousands of balls at the same rhythm so as to change rhythm at the exact instant. Data describes the repetition well. It describes the break poorly. But the break is what audiences pay to see.

There is one thought I always carry when holding a stat sheet: that sheet was produced by hundreds of points, and every point in it was once an unrepeatable moment. The number is the ash of the moment. Ash is useful for understanding the fire, but ash is not the fire.

In Vietnam we remain some distance from the world's tennis data industry. Most domestic tournaments have no player-tracking systems. Coaches usually take notes by hand, relying on memory and feel. Young players rarely possess a personal data record long enough to see their own development arc.

I do not think that is entirely a disadvantage. When tools are scarce, people are forced to observe with their eyes, remember with their bodies, and pass knowledge on by word of mouth. That is a form of knowledge that exists in no data file. But it has clear limits: it fractures easily between generations, easily degrades into unverifiable personal experience, easily becomes folklore.

The best approach I have witnessed combines both. A young coach in Da Nang once told me he had started filming all his student's training sessions, while also keeping a paper journal of his own impressions after each session. He said: the machine tells me how much my student ran, the paper tells me whether he was happy that day.

That method, to me, is the model of a decent data culture.

The legacy of a sport does not lie in the number of metrics it generates. It lies in whether the next generation can learn something from those metrics, and whether they dare say "I don't know" when the metrics are insufficient or absent.

I used to think the task of a sports journalist, growing with seniority, was to understand more, judge more, conclude more. Now I think differently. The task is to know precisely the boundary between what one knows and what one does not, and to state that boundary clearly to the reader. That is a form of respect for the audience no algorithm can replace.

If this piece has one thing to say to young writers entering sport, it is this: dare to return an empty analysis template when you truly have nothing to say. The crowd will not applaud immediately. But three years later, rereading your own work, you will be grateful you did not invent a story merely to fill the silence between two points.

Elite sport remains the common language of humanity, whatever a player's mother tongue, whatever the surface — clay, grass or hard court. But that language stays pure only when its interpreter does not add words the match never spoke.

I still keep the habit of arriving at the ground early, before the stands fill, when the surface still bears no shoe marks. That is when I write my most honest lines, because there is nothing yet to prove, no metric to show off, no conclusion to defend. Only a silent court and a writer waiting for the match to speak for itself.