Eight Dimensions, Zero Facts: The Cricket Data Report That Admitted It Knew Nothing

Eight dimensions. Several dozen cells. Not a single number.

The report that landed on my Liverpool desk last week opened with 'Not applicable' where the headline should have been, 'Not applicable' where the source belonged, and 'Unclassified' for the article type. Then came the eight sections — format and match analysis, player technique and data, team standing and rankings, league and commercial structure, rules and governance, risk, public narrative and expectation, industry transmission. Every table returned the same sentence: 'Insufficient information, cannot assess.' Beside every conclusion sat a confidence rating: 'High.'

I read it twice. The first time as an editor, the second as a man who has spent thirty-seven years between a scorebook and a spreadsheet. On the second read, one thing became clear: the document that could not supply a single number was the most honest document of the year. The information-points field was empty. Yet one thing survived — the domain label: cricket_asia. The hinge of the whole story is that one word.

Cricket analysis no longer happens only at a journalist's desk. In the Asian market — the IPL, the Asia Cup, the PSL, ILT20 — every series comes with a dashboard, every client wants a weekly report, every broadcaster wants 'insight'. The supply chain runs in two stages: one system reads the article and pulls out facts, a second writes the analysis from those facts. When the first stage comes back empty-handed, the second should stop. It did not.

Keep the economics of this industry in view. Asian cricket content carries more commercial weight than content from any other region, because the audience is densest here and advertising rates are highest. In a market that publishes thousands of reports a day, sending a note that says 'we have nothing' feels like losing a client. So the empty cell gets filled — not with facts, but with prose.

In 2026 I opened the batting for Udity Club in the Dhaka league and kept wicket too. I still remember those scorebooks: if a box was blank, we knew the over had not been bowled or the batsman had retired. Nobody filled the blank with a paragraph. Today's pipeline does the reverse: it wraps the blank cell in elegant sentences so the client cannot see that nothing is inside.

The first error hides here, and it is linguistic rather than technical. Every cell says 'cannot assess'. But on a dashboard, 'cannot assess' and 'low risk' look almost identical. Zero signal and zero risk have never been the same thing, yet our eyes are trained to read them as one. All six rows of the risk matrix should have read 'unassessable'; the reader will take them as a green light.

The format question sits at the centre of this, because without it no cricket comparison holds. Test, ODI, T20 and The Hundred are four separate universes. A strike rate of 130 is ordinary in T20 and explosive in a Test; new-ball economy and death-over economy cannot be judged on one ruler. The ICC itself keeps a separate rankings table for each format, because the institution knows that applying one format's benchmark to another produces a wrong sum. Put Shakib Al Hasan's Test average and his T20 strike rate in the same cell and the resulting number describes no cricketer at all; it describes our laziness.

In this report the format cell is empty. Any model reading the document has no way of knowing whether it is applying Test standards or death-over standards. The failure has a name: cross-format contamination. Every conclusion built without a known format is a guess, not an analysis; and a guess that introduces itself as analysis is no longer an error but a falsehood.

The second problem is the label. The contract specified the domain as 'Cricket'; what arrived was 'cricket_asia'. That is not a misspelling but a different taxonomy. 'Asia' spans everything from an India-Pakistan grudge match to a domestic franchise league. Anyone who reads that label and assumes the IPL is probably wrong. The telling detail is that the label survived while the body text died. That mismatch says the classifier never read the article — it worked from URL or metadata keywords. The machine telling us 'Asian cricket' has not read a single sentence.

The third problem is the most dangerous: the beauty of a structure can cover for the absence of evidence. Eight sections, tables, confidence ratings — all in order. Readers will quote this document, show it in meetings, act on it. A visible error gets caught; an elegantly arranged emptiness gets believed. In November 2026, after SK Telecom T1 lost 0-3 to Samsung Galaxy in the League of Legends final in Beijing, I wrote 'The Requiem' — because I had a scene and a number. A template has neither.

There is another layer that cricket analysis routinely buries: the accounting of luck. Toss, dew, rain interruptions and DLS — leave those out and no result can be measured; without DRS controversies, umpiring effects cannot be separated from performance. This document contains not one word on governance or integrity. That does not mean there is no risk. It means there is no information. Conflate the two and we sell ignorance as innocence.

The final observation is procedural. This empty record did not arrive alone. If one record in a batch has zero information points, the rest need checking too. If the 'Unclassified' tag keeps recurring, assume the problem is not a one-off fetch failure but a training gap in the classifier. One null record is an accident; a cluster of null records is a system's habit.

Now the sentence that throws a little cold water on all of this. The conventional fix is clear: install a hard gate, block stage two when information points are empty, repair the taxonomy, audit the batch. All correct, all necessary. But the problem is not only technical. It is a market problem.

We have built a market in which an empty document must wear the costume of analysis before it can be sold. Clients pay for structure, subscribe to dashboards, expect to hear 'eight-dimension analysis' in the meeting. Against that demand, saying 'I do not know' feels like professional failure. Yet that empty report was the system's last honest sentence — the one place where the machine admitted it had nothing.

The real danger is not that report. The real danger is the report with a plausible number in every cell — the kind of model that romanticises youthful potential and leaves dressing-room chemistry outside the ledger. There is no 'not applicable' anywhere in it, so nobody asks a question. I analyse like an accountant and dream like a kid in the stands; the difference is that a good accountant knows which cell he does not know. And the rookie problem is the silence after the highlight reel — nobody keeps accounts of that silence.

Before the next Asian season begins, three things should happen: stage two stops on zero information points, every risk cell maintains a permanent distinction between 'low' and 'cannot assess', and the two stages reconcile their taxonomy. But the larger task is cultural — a final cell in every report that states what the analysis does not know. The question is simple; the answer is uncomfortable. Of all the 'analysis' we have read this decade, how much was an empty record in a suit?

Every data point is a ghost story waiting for a narrator.