The Truth of the Empty Spreadsheet: When Cricket Data Refused to Answer

That morning eight tabs were open on my screen. Each tab was an analytical dimension: format and match, player technique and numbers, team landscape and rankings, league and commercial ecosystem, rules and governance, risk, public narrative and expectation, industry transmission. Eight dimensions. And everywhere a number should have sat, one word stood: N/A. Zero factual anchors. Only one field in the entire sheet carried any signal — a label, cricket_asia.

I have spent twenty-seven years on newspaper desks, where empty space is never left empty — it gets filled with "according to sources." The data desk taught me the reverse. Here empty space stays empty. If it fills, you first have to prove that what filled it was measured, not guessed. I left print to follow the numbers, and they led me here — to a sheet with nothing to write.

This was 2026. During the IPL I launched The Numbers Desk. I modelled death-over economy — overs 16 to 20 — across all eight franchises. Economy rate simply means runs conceded per over; in T20 a figure below 7 is generally excellent. The finding: sides conceding under 8.5 runs per over in that phase reached the playoffs five times in six. A senior television analyst dismissed it on air that evening as "spreadsheet cricket." The next morning I published the methodology footnote and kept the column running weekly.

Since that day I have had one rule: no claim without a sample size attached. Every column carries a methodology box — data source, sample, and one line on what the model cannot see. It made my copy slower to produce and far harder to wave away.

In 2026 I convinced my outlet to send me to the Russia World Cup as its data correspondent — one of the few Indian women in that press corps. I built an xG model across all 64 matches and showed that France won the knockouts with lower xG than their opponents, winning on efficiency rather than dominance. A European analytics blog cited the piece. Back home I began translating xG logic into cricket: expected runs per ball, wicket probability by phase.

The architecture of cricket coverage has changed since. The data desk is no longer the commentary box's assistant; it is its rival. But every pipeline has three parts: extraction, analysis, publication. Where the first part dies quietly, the second and third keep showing themselves to be alive. This article is the post-mortem of one such quiet death.

The sheet in front of me carried all eight dimensions in full template form — risk matrix, ranking table, transmission map, everything. Only the data was missing. And that is the most dangerous shape of all, because an empty template looks filled.

Finding one: the process stopped before completing its own checklist.

Two fields in that sheet were telling me themselves that they should have been populated. The time-sensitivity cell read "not assessed." The source-quality cell read "judge from the source fields" — when the source field was itself N/A. This is as if a reporter came back from the ground with only a date in the notebook and told the desk to pull a headline out of the material. This is not thin content — it is the extraction stage shutting down early.

Sports journalism knows this failure well. I reported from grounds for years, and what I learned there is that the most expensive error is never the wrong number — it is the empty number that gets treated as filled. A wrong number gets caught. An empty number never gets caught, because it has no opponent.

Finding two: mistaking a label for content is the easiest trap.

Only one field in the whole sheet carried a signal — cricket_asia. That is a routing label, not a fact. Its job is to say which branch the article should be sent to, not what the article says. But when everything else is empty, the label becomes the story. "Asia-region cricket news" sounds like information; it is a postal address.

I have seen this trap at my own desk. In IPL auction reporting the easiest numbers are the broadcast figure and the purse, and the hardest number is how much that purse strengthened the game. Commercial scale and competitive health are different things; a bigger purse is not a stronger game. When content is empty and only the label survives, we do exactly what auction coverage does — treat the easiest number as the story.

Finding three: with zero anchors, every downstream claim will be invented, and the ramp to invention is the real danger.

If I were to declare a single player name, a single team ranking, or a single format off this sheet, it would be invention in the plainest sense. And that is where the ethics of data journalism break. Because invention, dressed in the right scaffolding — "according to sources," "recent trends suggest" — reads like analysis.

Here I have to restate something basic about my trade that plenty of impressive-looking people do not actually know. Cricket numbers are not memoir; they rest on three things — format, role and sample. Format is fixed first, because Test, ODI and T20 tables are not comparable. In Tests, bowling average — runs conceded per wicket — is the primary measure; in T20 it is economy and strike rate. Strike rate itself means runs per hundred balls for a batter and balls per wicket for a bowler — confusing the two is a standard error. Role is the next step: opener, anchor, finisher, seamer, spinner, all-rounder, keeper — the choice of metric follows the role. And sample is the last door. One over in one match cannot support any claim.

In this sheet, none of the three was fixed. Format unknown, role unknown, sample zero. Under those conditions DRS and its umpire's-call debate, or DLS and its rain-revision arithmetic, apply to nothing. NOC and RTM conditions, the WTC points table, the FTP schedule, ACU monitoring — all listed in the template, none attached to a fact. This is the vocabulary trap: precision matters, and precision drifts into exclusion. A reader who watches only ICC events needs every term defined on first use — and when there is no content at all, vocabulary is only smoke.

If I had to decide when this sheet could be moved forward, my minimum information set is six items: article title and publication, format and match nature, at least one named entity, three or more discrete information points with attributable sourcing, a calendar anchor in the form of a date or series name, and, where available, any statistic or quote that can be format-tagged. Without the first four, no dimension can be opened.

So what is the correct professional outcome? Not that I fill eight dimensions by some means. The correct outcome is a null result, reported as a null result. The spreadsheet did not settle the argument; it gave us a better question — why our pipeline returns an empty result labelled "success."

Now the part where I have to fight my own habit. My identity is that of an analyst who produces the counter-intuitive finding, and here the counter-intuitive finding is very easy to produce: "See, a null result is also a result, a win for data journalism." I will not say that. The plain thing first: the extraction failed. That is a failure; do not convert it into an achievement.

The reversal holds in exactly one place. Those who dismiss data journalism as "spreadsheet cricket" will read this sheet as their victory — "look, the machine came back empty-handed." I do not agree with them, but their point is real. The job of a model is not to state the truth but to sharpen the question. And this sheet sharpened a question that gets buried inside any filled-in analysis: is there a gate in our system that can stop an empty sheet? In this sheet the gate worked, because every cell said N/A. The danger lies in the sheet where that gate is absent and every cell looks full.

When the narrative outruns the data, I slow the article down. This piece is the product of that slowing.

For the road ahead I will propose one concrete fix: install a completeness gate in the extraction stage. If the information-points list is empty, the one-sentence summary is blank, and time sensitivity reads "not assessed," the stage must not return "success." Format declaration should also be a mandatory field, and source grading — official board, authoritative cricket outlet, general media, self-media — should be settled at extraction time.

And one question I would put to every analysis this season: how many of them were built on a sheet that looked like this one — and nobody checked?