When Data Falls Silent: The Discipline of a Report With Nothing to Say

Two in the morning in Liverpool. On the screen sits a nine-part analysis template, with room reserved for first-serve percentage, return points won, break-point conversion, the structure of ranking points being defended, a form curve, injury risk. Every cell is empty. No player name, no tournament, no number to hold on to. I sat there a long while, hands on the keyboard, and the only honest line available was: insufficient information to assess.

Twenty years ago I would have filled that template in fifteen minutes. Not with data, but with confident prose. That is the old temptation of the trade: a gap always looks like an invitation.

An empty sheet is still a result

The process I run for tennis analysis has two tiers. Tier one deconstructs the source article into information points: player, tournament, surface, result, timestamps, source reliability. Tier two applies nine analytical dimensions to those points, from technique and form data through tournament systems, the professional landscape, rules and governance, team management, risk, media narrative and the flow of the wider industry.

When tier one returns nothing, tier two cannot manufacture a conclusion. The nine-part grid still appears, complete, but every cell carries one sentence: insufficient information. To an outsider that framework looks like a failure. To me it is a valid result, and usually the correct one. A null report does not say that no risk exists. It says that nobody has yet identified one, and the writer is not permitted to invent it.

The market does not pay for emptiness. Newsrooms need headlines, forecasts, a name to hang the story on. That pressure has pushed more than a few of my colleagues, myself included, into writing first and checking later.

In 2026 the season was cut in half. Wimbledon was cancelled for the first time since the Second World War. The ATP and WTA froze their rankings from March to August, which meant that for nearly half a year every forecasting model built on the 52-week points cycle lost its footing. I had to write across that void, and there was no decent way to do it while pretending the data was intact.

Four times I learned the value of waiting

In 2026, while working as a data consultant for Liverpool, I ran an xG model across the under-23 squad and hit an anomaly. A young forward had a touch count more than 30 percent below average but an xG per shot of 0.42. That was Rhian Brewster, seventeen, just back from injury. I recommended he train with the first team and was challenged on the grounds that my numbers were too theoretical. In a friendly against Tranmere Rovers he scored twice from three shots. That night at Anfield I stopped counting numbers in order to listen to the ghosts whisper.

But that was a case where the data was already thick. Three other times it was the opposite.

In Russia in the summer of 2026, in the quarter-final between the hosts and Croatia, I counted 148 kilometres run by the Russian team, twelve above their own group-stage average. I wrote a long piece predicting they would collapse in extra time, having burned their fuel. It was read 23 times. A colleague wrote about fighting spirit and was shared thousands of times. Russia taught me that silence is also the deepest layer of data.

In June 2026 a Championship club asked me for a report on football without crowds. I gathered 500 matches and found two signals: home advantage lost only about 0.18 expected goals per match, but teams trailing at half-time played long balls roughly seven minutes earlier than usual. The coaching staff adjusted their pressing around the report and took eight points from twelve that month. This time the data spoke, because someone was listening.

Then Qatar 2026. Japan beat both Germany and Spain by pushing their defensive line 1.2 metres higher after the break. I missed it, not for lack of scouting data from their pre-tournament friendlies, but because I had told myself it did not matter. I am too old to believe in miracles, but young enough to know which miracles can be measured.

The temptation to fill a gap

There is a paradox in this work. When the data is complete, I have to be careful not to conclude too much. When the data is missing, I have to be twice as careful, because every sentence can become a fabrication that nobody will ever check.

In tonight's report, the first line of the technical and tactical section states that the playing style cannot be classified because there is no subject to analyse. The first line of the data section states that no form curve can be built without match results. The risk section states that no category can be scored, and its closing line is explicit: this reflects missing input, not a finding that no risk exists. That distinction sounds pedantic. It is not. An entire sports media economy lives by blurring it.

What does a hurried writer do with an empty template? He picks a rising player, assigns him a plausible-sounding serve metric, adds an anecdote about a five-in-the-morning practice session, and closes with a prediction bold enough to provoke and vague enough never to be falsified. Readers will finish it. Nobody will check. Invented numbers enjoy the advantage of never contradicting themselves.

I have walked that road. My early work read more smoothly than my work does now, and I know exactly why.

Correlation is not causation, and yet a data gap is easily mistaken for an empty reality. A player with no clay-court data is not a bad clay-court player. He is simply a player who has never been measured. There are things data never touches, like the way a stadium breathes, and admitting that limit does not weaken a report. It makes the report real.

What I might be getting wrong

The figure of 0.18 expected goals per match comes from a sample of 500 matches spanning several competitions and levels. Pooling them was my technical decision, and it may hide real differences between football cultures. Expected-goals models are not a single yardstick either: different data providers define a chance differently, and the same shot can produce two different numbers depending on the system. I presented these findings as though they stand alone, when in fact they only stand inside the frame of reference I chose.

As for tonight's empty template, what I might be getting wrong is my attitude. Perhaps I have turned caution into a moral position, and a moral position is as easily a defensive habit as any other.

A signal for the next round

Starting this season I keep an extra file on my machine called the gap log. There are no metrics in it, only a list of things I do not know and the reason I do not know them: too small a sample, no source, or simply that I refused to look. Whenever a season turns the page, I open that file before I open the league table.

Every dataset is a garden: the farmer plants questions, and the harvest comes back as contracts. A fallow field lies to nobody. It simply waits for someone to plant, or waits for someone brave enough to walk past without planting anything at all.