Zero Out of Eleven: The Cricket Report That Looked Complete and Was Empty

On Tuesday a two-page PDF landed on my desk. Title in its proper place, subheadings aligned, eight analytical dimensions, a table under each dimension, 'Analytical Conclusions' beneath each table, and 'Evidence' beneath the conclusions. The paper was immaculate. The problem was that every cell repeated the same line: 'N/A — insufficient information'.

I counted. Eleven expected fields. Eleven unusable. And the array on which all eight dimensions were supposed to stand — 'Information Points' — was entirely empty. No headline, no source, no format, no player, no match. What I was holding was not an analysis. It was the shell of one, neatly printed.

Forty years of turning over scorebooks and association files teach one habit: an incomplete record is a finding, and a complete-looking incomplete record is a lie. The difference is that a blank page in a scorebook announces itself immediately; a blank column in a dashboard starts to look tidy.

In the winter of 2026 I asked the KSCA for the keys to a storage room at Chinnaswamy. Fourteen seasons of under-16 and under-19 scorecards, sixty-two VHS tapes, two interns and seven months. I opened the cupboard and fourteen seasons blinked back at me. The dust was not dirt; it was a filing system, and nobody had been left to read it. That work taught me something that has only become more urgent in the data age: gathering evidence and performing evidence are two different professions.

The analytical scaffold around cricket grew fast afterwards. Every franchise acquired a 'head of data', the NCA grew a sports-science wing, boards began commissioning 'comprehensive assessments'. A three-page model is now opened before a bid is placed at an auction table; a spreadsheet sits beside the scout's report in a selection meeting. That change was necessary and overdue. But every new system carries a new blind spot, and the blind spot has moved out of the scorebook and into the pipeline.

In February 2026 at Mount Maunganui, India's under-19s beat Australia by eight wickets for a fourth title; Shubman Gill's 372 runs came at an average of 124. Four months later a Bengaluru desk sent me to Russia as a systems correspondent instead of a match reporter. When the notes I filed from Kazan and Samara were laid beside the NCA's Bengaluru intake, one thing became clear. Two pipelines ran through one winter, and only one of them had a memory. That 4,000-word comparison was read by nine hundred people and two selectors. Ever since, every systems report I read gets one question: where is the evidence kept, and who keeps it?

In March 2026, when the academy froze, I took a forty percent pay cut so that two junior video analysts on ₹28,000 a month kept their jobs, and told nobody for a year. From a Bengaluru balcony I ran ninety-minute video sessions for thirty-two under-16 boys, nineteen weeks straight. Two of them made Ranji debuts in 2026. I wrote about neither the cut nor the sessions; I wrote about the analysts. A version of that ledger paragraph still goes into every report I file, because a cancelled season lives on in a seventeen-year-old's contract for years.

Back to the PDF, because what happened inside it is not cricket news. It is cricket-journalism news.

The audit was blunt. Eleven expected fields — title, source, article type, information points, core viewpoints, author stance, article purpose, entities, time sensitivity, source quality — and zero usable. One value was populated: cricket_world. That is a category tag, not evidence.

Here is the first and most important distinction. A partial null describes a thin article; a total null describes a broken pipeline. Had the title and source been present with everything else empty, we would say the piece was thin. When eleven of eleven are blank, the likeliest cause is not that the article contained nothing — it is that the article was never read. The fetch failed, a paywall intervened, a JavaScript-rendered page returned an empty body, or the text-decomposition stage was truncated mid-run.

There is a second possibility, and it comes from the archaeologist's habit: perhaps the 'article' was never prose at all. A photo caption, an embedded scorecard, a video clip, a live-blog stub — none of these survive a text-decomposition pass. My cupboard held such things too: objects on paper that could not be read, only named. Naming is what turns them into records.

Second point, and every selection meeting needs it. 'Insufficient information' never means 'examined and cleared'. The governance section's risk table was empty throughout. Now imagine how a busy editor or an exhausted selector reads that page — 'no rule controversy, no betting suspicion, low risk.' The truth was not that risk was low. The truth was that risk had never been assessed.

Third point, and this was the most honest thing in the whole document: it did not hide its emptiness, it declared it. I respect the method, because my own work runs on the same rule — you do not find a player; you find the trail someone kept. Where the trail breaks, writing 'the trail breaks here' is the honest act; filling the file with ink to manufacture false continuity is not. I hold a private condition: no archive claim reaches print until two independent sources agree.

But declaring the null is not enough. The most useful line in that document was not a cricket judgement. It was a technical condition that had not yet been enforced: Information Points must be at least one entry long — any record with not a single hard fact should be quarantined before the next stage runs. The eleven-field audit comes later; that gate comes first. Any organisation spending money on data will get several times its return from that one line.

There is a trap the document itself flagged, and I repeat it because it is the trap cricket buys most often. The governance checklist listed, purely as reference cases, the 2026 Cronje affair, the 2026 Lord's spot-fixing case and the 2026 IPL case. All three belong to the framework of precedents an analyst is permitted to invoke. They are not allegations. Printing them as 'findings' in a piece where nothing exists would be a serious professional error. And it is the commonest error of all: stuffing whatever material is available into the space that happens to be empty.

Now the part that bothers me most.

The standard complaint against analytics in cricket is that analysts walk into dressing rooms with conclusions detached from the rhythm of the match. The complaint is fair. But the real danger here arrived from the opposite direction — nobody said more than the data allowed, and nobody said less than nothing either; the shape of a conclusion was manufactured and left hollow. A wrong report is dangerous but correctable: someone stands up and says the numbers are wrong. Nobody stands up against an empty report, because there is nothing in it to point at.

And the institutional truth is this: cricket now buys the visualisation layer but not the man beneath it. A franchise will budget for a dashboard; paying the person who reads its fetch log feels harder to justify. Much like the seven deliberations a state association went through in 2026 before handing over the keys to a storage room.

What to watch is therefore plain. First, the re-run: same source, fetch logging enabled. If the information points come back empty a second time, the fault is not in the pipeline but in source accessibility — paywall or robots-block — and every future run against that source is a wasted count. Second, the null-rate across the batch: if several records from one run return totally empty and share a domain, that is a source-level problem, not a one-off. Third, and most important, a written warning at the editorial desk — this record was not analysed; it is not low-risk, it is unexamined.

That PDF reminded me of something old. A scorebook never lies, because its blanks are visible — you can see which over went unaccounted. The new system dresses the blank as 'N/A' and it starts to look full. The day we begin reading empty as full, we will have plenty of data and not one piece of evidence.