Nine Boxes, Zero Information: The Football Analytics Report That Wrote 2,200 Words and Said Nothing

Last week a 2,200-word analytics report landed in my inbox. Nine sections. A table in each. Four to six rows per table. A tidy answer in every row. And in every single answer, the same phrase: 'N/A — insufficient information'.

The title was blank. The source was blank. The one-sentence summary was blank. The author's stance was blank. Yet the document ran to its last page in perfect order — a risk matrix, a likelihood-impact-mitigation grid, recommendations, tracking signals, and a closing glossary of nineteen professional terms. A definition of xG. Of PPDA. Of FFP. Of the sell-on clause. Nineteen terms were defined. Not one of them was used in the analysis.

In October 2026 at Salt Lake, Jeakson Singh headed India level against Colombia in the 82nd minute; sixty seconds later Colombia won the game. That night I opened a spreadsheet — 21 names, 21 clubs, 21 contracts. In March 2026, locked into the Goa bubble with the ISL running behind closed doors and no press credential to my name, I hand-tagged forty I-League and reserve-team matches. Two hours per match. One rule governed the whole exercise: what I did not see, I did not write. This report ran on the same rule. The difference is that it printed its empty ledger across twenty-two hundred words.

The complaint that Indian football lacks data is three decades old. The truth runs the other way. There is still no data; a format has simply taken its place.

After the 2026 Under-17 World Cup every club wanted GPS vests, every academy wanted Wyscout scrolling, every media officer wanted a 'post-match data package'. A media manager at a Kolkata club once told me he receives three unsolicited 'AI scouting reports' a week. Nobody asked for them. Nobody read them either. At most I-League clubs the full-time analyst headcount is one — and that one person usually cuts the video and books the travel as well.

The problem is not that machines have started understanding football. The problem is that we handed them a form and withheld responsibility for the content. An analytics pipeline is supposed to read a match body, extract players, clubs, shapes and numbers, and distribute them across nine boxes. On the day the body itself is missing — paywall, JavaScript-rendered page, empty response — the pipeline has two options. Stop and say nothing was found. Or fill the boxes.

This report took the first option. That is its only honesty, and it is the reason it deserves reading.

The nine sections were tactical-technical, club finance and transfer market, results and public-opinion cycle, league landscape, rules and governance, management and dressing room, risk profile, media narrative, and industry transmission. Every section carried a table. Every table carried 'insufficient information'.

Then came the risk matrix, with six categories. Five read 'cannot be enumerated'. The sixth read 'Systemic: analytical-process risk — this output could be mistaken for substantive analysis.' Level: High. Likelihood: High. Impact: Medium. In twenty-two hundred words this was the only sentence containing a claim. And the claim was not about football. It was about the document.

For me that is the most useful passage in the entire artefact. An empty analysis named its own emptiness. It refused to convert 'no information' into 'zero'.

That distinction looks small. It is everything.

In 2026, when I ran Cohort 21 against senior minutes, only three of the twenty-one had crossed 500 senior minutes by the age of twenty. Three. The number was ugly and I published it. But had a script filled that spreadsheet, every player whose record could not be found would have been marked '0'. And '0' means measured, found nothing. 'No information' means never measured at all. Every analytics failure I have seen in Indian football has come from collapsing those two words into one.

The report did not make that error. Not in a single one of its nine boxes.

One field carried the instruction 'identify from the information points above'. There were no information points above. The template was ordering the analyst to extract names from an empty list. That is the moment a form outruns its content. I went back to the ledger and found a name nobody had circled — and it belonged not to a player but to a blank cell in a template.

In its recommendations the report wrote: check upstream logs, look for empty body responses, look for paywall blocks, look for JavaScript rendering. That is not a football recommendation. It is a maintenance ticket. And it is the most honest line in the document, because it concedes that the underlying article probably existed, and was lost.

One more detail. The report attached 'medium confidence' tags to a few inferences — but only where the inference concerned the pipeline, never where it concerned football. Where football was at stake it recorded no confidence at all, because there was no basis for any. And at the close it graded itself one star out of five on all four information-value measures: sporting, industry, timeliness, reference. One star across the board.

An old sub-editor on a Kolkata desk does not think that way. He fills the empty space and writes 'understood'.

Now the part where the easy conclusion fails.

The convenient reading of this report is: see, AI analysis is worthless, you need humans. I do not agree. I think it is the reverse.

Hand the same brief to a human pundit — blank title, blank source, blank information — and he will absolutely produce twenty-two hundred words. He will pick a formation. He will name a manager under pressure. He will write that the dressing room is fractured, because it sounds like reporting. The machine did the thing humans do not: it declined.

This report's failure is not fabrication. The failure is the inverse. The failure is that a null result was dressed in the full uniform of a finding. Somebody shipped a template. The template ran. Nobody had placed a gate at the door saying: if information points equal zero, do not generate a report.

And the glossary is the witness. Nineteen terms defined, not one used. That is the form built before anyone knew what it would carry. In the AIFF documents I worked through for years, that is the commonest artefact of all — the form first, the question of what goes in it later.

And when reports are built on matches from the Northeast, where names like Chhangte come from, the footage often has not been tagged at all. Tag the footage first; otherwise you are just arguing about your memory.

What would have to change starts with one line.

Every analytics report should carry a header, the way a match sheet carries the weather: 'This report had sufficient information in 0 of 9 sections.' That is all. Then the reader knows what he is holding — an analysis, or the shell of one.

In an empty stadium you can hear everything the crowd used to cover. In an empty report you can see everything a full one hides.

Because when nine boxes look full and there is nothing inside, the question is not what the machine said. The question is why the man who sent it never opened it.