Tennis's Empty Mirror: When the Data Disappears and What Remains

That night, on my computer screen in Liverpool, a data table appeared filled with nothing but zeros. Not the zeros of a match that had not yet begun, not the zeros of a player who had yet to score. It was the zeros of an analysis pipeline that had collapsed somewhere upstream, leaving a frame clean and empty to the point of insolence. I stared at it for a long while, my cup of tea slowly going cold at my elbow. Thirty-eight years in this trade, and it was the first time I had seen a tennis dataset with no player names, no tournament, no dates, not a single serve to analyse. Only the void. And inside that void I recognised something the profession rarely dares to say aloud: most of what we call analysis rests on the belief that data will arrive. We have never prepared for the day it does not.

I still remember the fateful evening at Anfield in 2026, when a seventeen-year-old boy named Rhian Brewster kept me at my desk until late. The xG model I ran for the U23 side produced a strange figure: his touches in the box were thirty percent below average, yet each shot carried an expected-goals value of 0.42. That is the kind of number that makes you stop. I took it to the coaching staff, was called a theorist, and then in a friendly against Tranmere Rovers he scored twice from three shots — exactly as the model had whispered. That night I believed data could tell stories the eye could not see. But tonight, staring at the empty frame, I finally understood the other side of that belief. A broken data pipeline is not like an empty inbox. It is more like a mirror polished so clean it reflects the void, and anyone who wants to believe in themselves can look into it and see whatever they wish.

What modern tennis analysis has not fully confronted is the boundary between the absence of data and the absence of truth — two different things, routinely conflated, and that conflation is becoming the largest blind spot in the entire industry.

To understand why, one must look at how far tennis has come in two decades. The Hawk-Eye era turned every ball into a coordinate. ATP and WTA statistics systems turned every point into a queryable row. Hawk-Eye, in-court sensor systems, data companies such as Tennis Data Innovations and TennisViz, and in Britain the growth of deep analysis platforms for academies. At Grand Slam level, every match at Wimbledon or Roland Garros now generates millions of data points: serve speed, spin rate, return position, net-points-won rate, shot interval, even the heart rates of spectators. The trend is not limited to the majors. Challengers, the lowest tiers of ITF events, are being digitised too. Academies in Spain, France, and the Nordic countries all hire data specialists. I know a friend working for a youth training centre in Croatia who sends a twenty-page report every week for an under-14 student.

Behind that elaborate infrastructure sits an assumption never formally stated, yet present everywhere: that data always exists, is always complete, always clean, always trustworthy. On this assumption we build machine-learning models. On this assumption we write pre-tournament briefings. On this assumption we sign transfers and advise on tactics. But an assumption is not a guarantee. And when it collapses — as it did that night on my Liverpool screen — people discover they had been building a house on a foundation no one had ever checked for depth.

I have spent years watching how tennis data is collected in the harshest environments. It is an unglamorous job, full of dead ends, and it almost never appears in headlines. There was a Challenger in South America I followed in 2026, where the in-court sensor system failed midway through because of extreme humidity. The organisers still published full statistics after each match, but those numbers were not measured — they were estimated by eye by a volunteer who had worked fourteen hours straight. No one recorded that the data had been estimated. No one flagged that those figures carried a different level of confidence. They flowed into the shared database and sat there like true numbers, waiting to be pulled into someone's report in London or Munich, someone who would never know they were using a tired volunteer's estimate.

This is where I want to pause for a beat. When I write about a player, I always remind myself that every number in my hand has passed through a long chain of human hands. It was recorded, checked, transmitted, stored, retrieved. Every step can introduce error. Every step can turn truth into something close to truth but not quite it. And while our trade has become exquisitely sophisticated at analysing data, almost no one is paid to doubt the data itself.

That is why an empty data pipeline is more dangerous than a faulty one. When data is wrong, you still have a player, a match, a context against which to check it — the truth of the match is the anchor. When data is empty, you lose the anchor entirely. And an analyst who has lost the anchor will tend to drive his own stake into the ground and call it insight.

I once witnessed this in another form. In the summer of Russia 2026, I was in Moscow as an analyst for a sports outlet. I spent an entire evening reviewing Russia versus Croatia, noting that the Russians had run a total of 148 kilometres, twelve above their group-stage average, and I predicted they would collapse in extra time because such physical effort was unsustainable. I wrote a long, careful analysis, full of figures I was proud of. It got twenty-three reads. A colleague writing about Russian fighting spirit was shared thousands of times. That night I sat alone in my hotel room, looking out the window, wondering whether I was too dry. But later I understood something else: the audience did not reject data. They rejected data without a face. Every number needs a story to reach the heart. And my job is to find that story, not to pile up figures.

That lesson — data needs the clothing of a story — followed me for years. At Qatar 2026, when Japan beat Germany and Spain with a defensive line 1.2 metres higher than their opponents in the second half, I frantically re-checked my own data to find why I had missed it. The answer was not in the data. It was in my pre-tournament bias. I had focused so heavily on the big teams that I had ignored Japan's warm-up friendlies, sessions that, had I watched closely, would have shown the traces of a defensive system built specifically for taller opponents. I promised myself never to let prejudice cloud my data eye. And in recent years I added another rule: never trust a report unless I know where it was born.

Back to the empty mirror on my Liverpool screen. If I took that emptiness and continued, what would happen? There are three scenarios, and all three are frightening.

Scenario one: I fill the void with assumption. My experience gives me enough to guess a player, a match, a context that fits the analytical frame. I could write a diffident but confident piece, based on patterns I have seen thousands of times. It would read very persuasively. And it would be entirely fiction. This is what I call the most dangerous professional crime: inventing truth from an empty dataset and letting the fluency of the prose conceal the groundlessness of the content.

Scenario two: I stay silent. I tell the editor the input was empty, that I cannot analyse what does not exist. This is the most honest scenario, but also the most undervalued in the industry. Because in a world where speed always wins, the silent one is considered slow. No one praises an analyst who refused to write. No one pays a fee for discretion.

Scenario three — and I think this is the most common in today's tennis data world — I push the problem aside, temporarily ignore the gap, and continue producing content based on the most recent, or oldest, or otherwise sourced data, without telling anyone that the original source failed. Gradually, that film of dust accumulates. Weeks later, no one remembers that a dataset was once empty. No one remembers that there was a match we did not truly observe.

This third scenario is the most troubling, because it is nearly invisible. It is not a clear lie. It is a chain of small, unmarked silences.

What is notable is that other sports encountered this problem before tennis, and they handled it in very different ways. English football went through a minor crisis in the early 2010s when it emerged that possession and distance-covered figures in some lower-division matches had been crudely estimated yet published as official. Professional basketball leagues are far stricter, with cross-checking systems across multiple independent data providers. But tennis has a peculiarity that makes the problem harder: tournaments are scattered worldwide, with different data providers on each continent, inconsistent standards, and a ranking system computed from that data — points on which a player's career can depend.

Do you see the gravity? Ranking points determine which players enter the main draw directly, which must play qualifying, which are seeded, which face hard opponents in the first round. A small distortion in point calculation at a distant Challenger can silently change an entire Grand Slam draw months later. These are things rarely mentioned, because they are not glamorous, have no highlights, no big names. But they are precisely where the integrity of this sport is decided.

Once the industry loses the ability to doubt its own data, it will gradually come to believe the scoreboard is objective truth. But the scoreboard is always a human-made construct, with human-set limits and human-unacknowledged gaps.

At this point I want to return to something that seems unrelated but is directly relevant. In 2026, when European football was paralysed by the pandemic, a Championship club contacted me to produce a special report on performance in empty-stadium conditions. They worried about team morale. I analysed five hundred matches and found something surprising: home teams lost only 0.18 expected goals per match without fans, but teams trailing had a tendency to play long balls seven minutes earlier than usual. Seven minutes. That was a number I was not looking for; it emerged as I peeled the data back layer by layer, the way one leafs through an old diary. The coaching staff adjusted their pressing tactics based on that finding and took eight of twelve points in June. I tell this story not to boast. I tell it to say that even in crisis, data can illuminate a path forward — provided one is humble enough to listen to it and clear-eyed enough to distinguish its voice from one's own echo.

There is something data can never touch — the way a stadium breathes. I have stood in arenas large enough to know that atmosphere cannot be digitised. The roar of the crowd, the tension before a decisive serve, the moment a player looks down at the court and takes a deep breath — those are not in any data column. But they shape the match in ways data can only faintly reflect. And this is where pure analysis fails: it can measure consequences, but it struggles to reach causes.

I think about this as I look at the empty mirror on the screen. That emptiness, in the end, is itself a truth. It is the truth of a system that broke somewhere. A severed connection. A hung script. A lost data feed. I can treat that gap as meaningless, or I can treat it as a signal. And in my trade, every signal, however small, has value — as long as I let it speak.

What I want to stress to those younger than me, who enter this field when everything is already digitised, is the ability to endure emptiness. In modern analytical culture, a gap is treated as failure. No data means no value. But some gaps are more honest than any data. A gap tells you something is missing. It forces you to ask questions. It does not let you deceive yourself with numbers that merely look good on the surface.

I am too old to believe in miracles, but young enough to know which miracles can be measured. And among measurable things, I distinguish sharply between what is truly measured and what is merely assigned the appearance of a number. That is the hardest discipline this trade has taught me, after thirty-eight years.

I no longer believe data can lie, in the active sense. But I believe people can create a counterfeit universe out of numbers. We can fill a gap with assumption. We can turn an estimate into a fact simply by not noting that it is an estimate. We can turn a small sample into a rule, a trend into a destiny, a good month of form into a great career. And in doing so, we naively imagine we are merely analysing.

There is a question I do not know anyone in the industry wants to answer: what happens when our models begin to be trained on unclean data, and those models are then used to generate predictions published to audiences as if they were fact? The question is not an abstract philosophical debate. It is a testable technical problem. But to test it, one must admit that data can be dirty. And admitting that has never been something the sports analytics industry wants to do.

That night in Liverpool, after staring at the empty mirror long enough, I did what I usually do when I meet something I cannot immediately understand. I stood up, made a fresh cup of tea, and began writing again from the start. Not an analysis piece, but a piece about the gap. About how a system can fall silent, and how we respond when it does. About what remains when the numbers vanish. It was a strange evening. No player, no match, no name to write about. Only a keyboard and a void made of bits, and a fifty-four-year-old man sitting between the two, trying to find something meaningful.

I think the meaningful thing I found is this. All my life I have hunted the ball, but what I was really seeking is the formula of longing. And longing is not in the scoreboard. It is not in the columns. It lives where data cannot reach: a deserted stand under stadium lights, the sound of a racket striking a ball on a late afternoon at some training centre, a look in a young player's eyes when he discovers he can win. Those things exist. They are real. But I can only reach them through story, not through a data column. And this is perhaps the final truth the trade has taught me: data is a means, not an end. People do not love tennis for the numbers. They love it for the emotion those numbers try, in one attempt, to capture.

I have no conclusion for that evening. I only have a new question, as almost always happens after I face a gap: if data has taught me to be humble, then is the emptiness of this dataset the greatest lesson in humility it has ever given me? And if the answer is yes, then perhaps I should not fear evenings like this. Perhaps I should welcome them, the way a farmer welcomes winter, knowing that rest is the condition for the next season's harvest. Every dataset is a garden — the farmer plants questions, and the harvest is contracts. And there are seasons when that garden must be left alone, not sown at all, simply to listen quietly to the earth breathing.

The next morning I sent a message to the colleague responsible for my data pipeline. I wrote briefly that the system had returned an empty result set overnight, and that I needed confirmation whether it was a technical incident or a condition of the source. I did not complain. I did not demand an immediate fix. I simply asked. That is the small ritual I have taught myself over many years: when meeting a gap, ask rather than fill. Let the question sit alone, wait, until it tells you what is truly missing.

And while waiting, I think of the players I have followed in my career. People usually talk about their numbers: titles, wins, Grand Slam championships. But what I remember most about them is not those numbers. It is the moments before the number appears: the moment a player stands at the baseline, takes a deep breath, and the whole world seems to hold its breath. That moment is in no data column. And it is what makes tennis one of the most beautiful sports humanity has ever created.

If there is one thing I want to leave to those who will carry on this trade after me, it is this: love data enough to know when it has nothing to say. Do not paint over a gap with pretty numbers. Do not let the fluency of prose fool you into thinking you understand. Because one day, like that night in Liverpool, an empty mirror will appear before you, and you will face a choice: fill it with assumption, or leave it empty and keep searching. I choose the second. Not because it is easier. But because it is the only way data can learn to sing.

It is now well past midnight. Outside the window, the city of Liverpool sleeps, and the streetlamps cast a yellow light on the empty street. I lift the tea that has gone cold a second time, take a sip, and place my hands on the keyboard. The data table on the screen is still empty. But in my head, hundreds of matches, thousands of players, hundreds of stories wait to be told. There is no column for them. No cell in the spreadsheet contains them. But they are still there, vivid and real, waiting for someone patient enough to tell them. Russia taught me that silence is also the deepest layer of data. And tonight, that empty mirror is teaching me that truth sometimes needs a gap in order to speak louder. When the stands are empty, the numbers begin to learn how to sing. And when the numbers have nothing left to sing, perhaps one learns that this sport, in the end, does not belong to the numbers.