The Blank File: Why Tennis Injury Data Refuses to Speak

The Blank File: Why Tennis Injury Data Refuses to Speak

Melbourne, mid-January. On Court 3 at Melbourne Park, in the seventh game of the second set, after a hard push to his left to chase a cross-court ball, the player stops, turns toward the umpire's chair and raises a hand. Three minutes later the tournament physio arrives. He sits on the sideline, removes his shoe, and the right ankle is already swollen. Tape goes on in two layers, running along the arch and looping over the lateral malleolus. The stands go quiet. The match lasts four more games and then ends.

The next morning I open the ATP results page. The note beside his name fits into one word: Retired. No injury site. No diagnosis. No estimated return window. In a tour that moves hundreds of millions of dollars a year, a body has just sent an urgent signal, and the only thing recorded is a single word.

That scene is a composite, stitched together from many matches I have watched from the stands at Melbourne Park over the years. The word Retired is not a composite. It is the standard format, and it is the entire public record of a medical event.

I keep a separate column in my spreadsheet. I call it the blank column. It is where the scoreboard says nothing.

Why tennis has no real injury database

In 2026, while I was an international communication student in Melbourne, I spent more than four months rebuilding a dataset of 314 injuries across three A-League seasons. That work left me with a number I have carried through my entire career: players who returned to the pitch before day 14 had a recurrence rate 41 percent higher than those who returned after that mark. That number exists only because football agrees to keep records. There is a date of injury. A type of injury. Minutes played. A return date.

Tennis gives me no equivalent, and the reason is not laziness on anyone's part.

A Grand Slam organising committee holds complete medical files. But those files sit inside a doctor-patient relationship, belonging to the player and his team. The media only touches the shell: a withdrawal, a walkover, a medical timeout, a retirement mid-match. The ATP and WTA do publish seasonal injury surveillance research, but it usually arrives aggregated: distribution by body region, by surface, by gender, by round. Not per player, and almost never with load metrics attached.

Sports medicine calls the hardest part of this the definition problem. The most common counting method in research is time-loss: a case only counts if it costs an athlete at least one training session or one match. That method is consistent, comparable, and was standardised through international methodological consensus in the mid-2000s. It also carries a fatal blind spot: every ache not severe enough to make someone leave the court evaporates from history.

I call the tennis injury record a map with holes punched through it. The coastline is drawn beautifully. The small bays are missing.

Four layers of blank space

Definition comes first. Under time-loss counting, a player serving hundreds of times per match while his wrist aches for six straight weeks will not appear in any table, as long as he never misses a match. The injury existed. The case did not. The threshold at which pain gets recorded is not a biological threshold; it is an administrative one.

Privacy is the hardest layer to challenge, because it is legitimate. No player owes a journalist in Melbourne his medical file. But there is a gap between not publishing a diagnosis and not publishing any structural indicator at all. A tournament knows exactly how many minutes each player moved, how many sprints he produced, how many serves he hit at maximum in a single set, because the electronic line-calling system captures all of it. Those numbers could be released anonymously and in aggregate without violating anything. They are not released.

Load is where tennis falls furthest behind football. Football has GPS units in shirts, sprint distances, weekly training volume. Tennis has something far harder to standardise: mechanical load at the shoulder, elbow and wrist, generated by the repeated service motion itself. A player serving hundreds of times in a tournament week is ordinary at the top level. But that figure is recorded nowhere, so nobody can compare it with last week, last month, or last season.

The calendar closes the loop. Professional tennis runs nearly year-round, switches surfaces several times a season, and most players never get an off-season long enough to rebuild a physical base. In football, a returning player gets weeks of individual work, then a few reserve matches, then first-team minutes. In tennis, a returning player usually walks straight into the first round of an ATP 250 against an opponent in real match rhythm. No buffer period is recorded, so no buffer period is ever verified.

Three files, three forecasts off target

Roger Federer underwent three operations on his right knee over roughly two years, from early 2026 to mid-2026. The public version of that story is one line: knee surgery. He returned at Wimbledon in 2026, reached the quarter-finals, lost to Hubert Hurkacz in straight sets, and never played professional singles again. He retired at the Laver Cup in September 2026.

What I want to know, and cannot know, is not the diagnosis. I want to know his knee flexion range in week three after the second surgery. I want to know his daily serve volume in the fortnight before Wimbledon. Those numbers exist, inside his team's records. They do not belong to the public. So when a 39-year-old body surrenders in a quarter-final, the only vocabulary left is age, knee, bad luck.

Andy Murray is the same kind of file pointing the other way. He had hip surgery in January 2026, then again in January 2026. The second was a hip resurfacing, an intervention most of the profession treats as the end of a singles career. He said publicly that his goal had become simpler: to be pain-free enough to play with his children. Then he came back, and played for several more seasons.

Alexander Zverev fell in the Roland Garros semi-final in June 2026, against Rafael Nadal, tearing several ligaments in his right ankle. He returned after more than seven months. Dominic Thiem injured his right wrist in June 2026, his absence ran longer than projected, and although he returned to competition, he never recovered the standing he once held.

Three cases, three completely different endings, and not one line of public data that could have separated them beforehand. In all three we have only the name of the injury, a label, never the process that led to it.

Based on my experience of watching these matches, the common denominator is not the injury type. It is that all three players entered the decisive phase of a career with no reference table to compare themselves against their own previous season.

Two symmetrical mistakes

I have been wrong in both directions, and both times because of the blank space.

In 2026, at the World Cup in Russia, I tracked Neymar playing just 50 days after surgery on his fifth metatarsal. I recorded that his dribbling volume rose while his sprint speed dropped, and I wrote a series predicting reinjury risk. That prediction did not play out as written. My error was not in the data. It was in filling the blank with a conclusion stronger than the evidence allowed.

The symmetrical error is just as common and more dangerous, because it attracts less criticism. It is treating silence as safety. A player who withdraws from no tournament for six months gets described as healthy. But if nobody measures, healthy is a polite way of saying there is no evidence to the contrary.

An empty field in an injury record is data that was never collected, not an indicator of safety. I have to remind myself of that every time I sit down to write. Data does not lie, but the body always knows how to hide its illness. My job is to stand in the dip between those two statements.

Then there is culture, which I see most clearly because I sit between two of them. In Vietnam, the reflex in sport is to endure the pain. In Australia, the default reflex is to measure before the pain arrives. The first produces athletes who tolerate a great deal and injuries that surface late. The second produces athletes who are over-managed and layoffs that last longer than necessary. The blended answer I believe in sits in between: keep the Vietnamese will, but never take your eyes off the numbers. For that to work, the numbers have to exist first.

The notable thing is that the technical barrier to this is close to zero. Hawk-Eye, serve-speed sensors, ball-tracking data: all of it is already in place at every major event. The obstacle is the will to publish, not the capacity to measure.

What I would like to see

An anonymised injury ledger at tournament level, updated weekly, with four fields: body region, date recorded, days lost, and the player's load index over the preceding 14 days. No names. No detailed diagnosis. Just four fields.

Those four fields would be enough to do something nobody can do right now: detect injury clusters before they become headlines. Enough to answer the question every tennis analyst dodges: are ankle ligament tears rising as the calendar compresses, or are we simply paying more attention.

A meniscus does not tear from one collision. It tears because for two seasons a body has been quietly writing a leave request. I do not believe in accidents; I believe in risks that were never tabulated. Tennis's problem, at this moment, is that it has a great many bodies and very few tables.

If the major tournaments agreed to publish those four fields, the data would still not lie. But at least it would start to speak.