The Empty Cells in Tennis Analytics

The 2026 Wimbledon final lasted four hours and fifty-seven minutes. Roger Federer won 218 points in that match. Novak Djokovic won 204. Djokovic lifted the trophy. The first fifth-set tiebreak in the history of a Wimbledon final, triggered when the score reached 12-12, ended in a place the aggregate scoreboard never touched: the two championship points Federer held at 40-15 while serving at 8-7, and the way Djokovic sent both of them back, not beautifully, only deep.

That night I sat in the seventh row of the main press room, watching the number 218 blink on the screen and then get swallowed by the ticker. Nobody in the room asked why the man with more points was the man who lost. That question had no seat in the news. It lived somewhere else: in the data cells left empty, empty because nobody wrote them down in time, and empty because no instrument existed to measure them.

Since that night, whenever I open a tennis analytics sheet, I look at the empty cells first. Empty cells do not wait to be filled. They mark places where there has never been a measure.

A modern Grand Slam singles match generates several thousand data points. Hawk-Eye records where the ball lands to the millimetre. The serve clock counts every second. Serve speed and spin, rally length, distance covered, net approaches, second-serve points won, all of it lands on a server before the spectators have settled into their seats. But that system thins out fast as you travel down the tiers. On a Challenger court, the person keeping stats may still be a volunteer with a notebook and a pencil. At an ITF event, sometimes the only remaining tool is a phone recording the court so someone can watch it back.

Tennis analytics always has two floors. The ground floor collects: who served, where the ball landed, how long the rally ran. The upper floor interprets: what the collected material means. When the ground floor returns empty, no shot logged and no player identified, the upper floor can do nothing except admit it can do nothing. The industry calls that a null result.

A null result has a dangerous charm. It reads a great deal like calm. "No anomalies detected" sounds far gentler than "never measured", even though the two sentences describe two different conditions. During Grand Slam season, when the volume of numbers multiplies and the pressure to have an opinion within fifteen minutes of the final serve peaks, the gap between those two sentences is erased fastest. A writer needs a conclusion. An empty cell gives nobody a conclusion.

The first thing an aggregate scoreboard will not tell you is that total points measure a different match from the one that hands out the trophy. Federer won 218 points and lost. The number is correct, and it measures his average strength across nearly five hours. But titles are not awarded on averages. They are awarded across forty seconds. Total points measure a player's average strength; the trophy measures his strength in the exact instant the scoreboard was not watching. Between those two measurements lies a gap, and that gap is where titles are decided.

My own experience watching matches in person over more than twenty years gives me one rule: never read a metric before you know where it was born. Distance covered is the clearest example. A player who runs twelve kilometres across five sets is usually described as durable. But running a lot also means being pushed a lot. In his peak years, Rafael Nadal was always among the tournament's heaviest runners, and most of that distance he ran to his left, chasing balls his opponents believed they had already won. When a player's distance suddenly drops, the easiest reading is that he is conserving energy. The other reading is that he can no longer get there. One number, two opposite meanings, and neither one is written in the cell.

First-serve percentage behaves the same way. A high number can signal dominance, or it can signal caution: the player lowering his target toward the middle of the box to get the ball in, trading away a second serve he no longer trusts. A low number can be a technical fault, or a tactical choice aimed at a narrow corner. The sheet presents all three possibilities in exactly one cell and lets the reader choose.

The fourth thing, and for me the most painful, is the ranking. A tennis ranking is a memory book, not a measurement. It accumulates results over 52 weeks and then subtracts the oldest ones. Which means a player ranked 30th may be performing like the 12th while still carrying the points of the 45th from last season. That position is true about the past and false about the present, and it is false in exactly the way nobody can fix, because fixing it would break the tournament system.

In Vietnam, the ground floor of that architecture is thinner still. At domestic events, the stats keeper is a coach with a notebook, and the list of what goes unrecorded is longer than the list of what gets written down. Ly Hoang Nam once reached the ATP top 300; Nguyen Thuy Linh once broke into the WTA top 100. Both milestones are real, and both were built almost entirely outside any professional measurement system. When a Vietnamese player loses in the first round of an ITF event, there is no analytics sheet to read. There is only someone sitting beside the court, remembering.

In 2026, when the pandemic froze sport, I built a documentary project filming empty venues. On a deserted tennis court, I heard a ball bounce in the vaulted corridor, an umpire's chair creaking in the wind, birds. An empty court, it turns out, has its own sound of longing. The pandemic froze sport, but it could not freeze what we tell each other. During those two months, when every analytics sheet was blank, I understood that my job had never been to fill the cells. It was to point out which cells were empty, and why.

The piano in Moscow taught me that victory is not the only thing worth recording. That lesson arrived in 2026, in the middle of a World Cup, when I was writing about a Croatian midfielder and realised that most of what made him who he was sat outside every statistical column. Years later, when I moved to covering tennis for the American market, the lesson was still intact.

A null result does not mean low risk. That is the sentence I would like to write in bold and tape to the wall of every sports analytics room. In daily practice the two states get merged into one: "we could not measure it" and "we measured it and it looked normal" both produce an empty cell, and that cell then gets read in a tone of relief. But the most dangerous empty cells in tennis are always the rows with no column: an undisclosed injury, a drained motivation, a cracked relationship with a coach, a fear of losing that has grown larger than the desire to win. No Grand Slam issues a data sheet for any of those.

Attached to this is a reading I consider lazy: the concept of the "unforced error". The name assigns all responsibility to the player striking the ball, when most balls labelled that way were created by a previous shot from the opponent, half a metre deeper, a little heavier, landing exactly where the player least wanted it. Calling it an unforced error writes the opponent out of the story, and out of his own data sheet. It is one of the empty cells that has been legalised.

The players who are tracked least are usually the ones who most need tracking. Outside the top 100 there is no Hawk-Eye, no analytics sheet, no one writing. Before they were contracts, they were children carrying a dream in search of a home. They play in front of stands where I have stood listening to the sound of rackets, and when they leave, no data cell records it.

This Grand Slam season will close, and someone will lift a trophy with fewer points than his opponent. Somewhere on your analytics sheet, empty cells will remain on exactly the most important rows. The question then is not what you missed. The question is whether you dare to write that you never measured it.