The Empty Column: When Missing Data Gets Read as Good News

On my desk lies a tracking sheet, face up. One entire column reads the same three characters: N/A. That column is for break-point conversion. There is not a single line of data — no sample, no denominator, nothing to check against.

The person sitting across from me glances at it and says: "So he's fine."

I don't answer right away. After more than forty years of carrying a notebook, I have learned something that sounds far too simple: a blank page and a clean page are two different things. A clean page has been checked. A blank page has never been touched. A reader in a hurry always confuses the two, and in tennis that confusion carries a price.

Context

This sport has a structural problem football does not have: the samples are tiny. A player contesting 60 to 70 matches in a season is a good year. Of those, the number of tie-breaks usually falls somewhere between 20 and 30. Break points faced may run into the hundreds, but the number of times a player walks into a deciding game in the third set is far smaller.

Which means most of the labels we use to describe a player — steely, ice-cold, or shaky when it matters — are drawn from a sample small enough to worry about. Television needs labels. Audiences need labels. Rankings and sponsorships like labels too. But the one column that decides the outcome is the one left empty.

At this stage of the calendar, that is starker than ever. This is the hottest coaching market in several years: support teams changing hands, young players pushed onto front pages before they have the data to justify the position, and every new contract dragging a new story behind it. The story about a team always runs faster than the data about that team. That is not commercially wrong. It is only technically wrong, and technical errors usually take three months to surface.

The Core

The first thing I do with an empty column is go looking for the denominator. For tie-breaks the arithmetic is simple: if a player wins 50% of tie-breaks across a long career, then with 25 tie-breaks in a season the standard error is already around 10 percentage points. That means a player winning 58% and a player winning 46% in the same season may be standing in exactly the same place. One gets called steely. The other gets called mentally fragile. Both may simply be noise from a small sample.

The second thing is separating columns that survive time from columns that do not.

Second-serve points won is a column I write in ink, not pencil. It is stable, it reflects a player under genuine pressure, and it moves little from tournament to tournament. Return points won is stable too, but it must be read alongside the quality of the opponent's serve — returning against a 210 km/h delivery is a different job from returning against a heavy kicker.

And break-point conversion? That is the most animated column on television, and the one that regresses hardest toward the mean. Chances created are data; chances converted are narrative. A player who generates 12 break points in a three-set match has proved something about reading the opponent's serve. Converting only 3 of those 12 says more about the quality of the chances they created and about the player across the net than it says about nerve.

That is why I never open a piece with conversion rate. People watch the winner; I watch the space before the winner is struck. That space lives in the return position, in the direction of the return, in whether a player dares to step inside the baseline at 30-30.

The surface complicates the arithmetic further. On clay the serve is worth less, hold rates drop, and break points appear far more often. On grass the serve dominates, hold rates jump, and chances created shrink. A player converting at 40% on hard court and 40% on clay may be doing two fundamentally different things, even though the column looks identical.

Based on my experience watching matches across many seasons, I always add a line the broadcast graphic does not carry: what level the opponent was serving at that week. The same 40% column, placed beside two different opponent serve levels, tells two different stories.

Then there is the truly empty cell. A 19-year-old with eight tour-level matches behind him. The chart-keeper writes N/A into nearly every column: insufficient second serves under pressure, insufficient tie-breaks, insufficient grass-court matches. That empty cell can be read two ways. First: no weakness detected. Second: nothing is known at all. The first reading is more comfortable, and it is the reading that has pushed plenty of young talents into a spiral of misplaced expectation.

The forty-page notebook never lies. It only records what I record. The problem lies with whoever reads it.

The Contrarian Angle

Here is what I want to say plainly: unassessed does not mean low risk. This is the sentence most likely to be skimmed past in an entire report. When a column is left blank, people tend to fill it with a safe assumption — because an empty cell in a spreadsheet looks like a cell that has already been dealt with. In tennis, that empty cell is usually where everything hides: an undisclosed injury, an unstable technical change, a wrist problem that only shows up after the third set.

In the other direction, some players are measured far too much. Novak Djokovic, with 24 Grand Slam singles titles, or Iga Świątek, with four Roland Garros titles, are cases where the data is so thick that the label becomes private property with its own inertia. A label that was accurate three years ago is still quoted verbatim today, even though the technical conditions have changed. Plenty of data does not automatically mean data that is still right.

A practice court has no spectators, but every answer is there. Three in the afternoon, the third hour of a session, a young player's second serve toss sitting a couple of centimetres higher than in the first hour. That is data. Nobody puts it on a broadcast graphic, but it says more than any label about nerve.

Takeaway

An empty column is scarier than a column full of bad numbers, because it provokes no reaction. When everyone watches the ball, I only see the hand directing from the sideline — and that hand is usually holding a sheet with a few cells still blank.

The signal I will track through the coming hard-court swing: who fills that empty cell first, and with what. A young player entering a fourth consecutive tournament week with second-serve points won sliding week by week tells me more than any press conference. And if that column is still blank three months from now, the reader should ask: has nobody measured it, or has somebody measured it and chosen not to publish?