The Report Came Back Blank: What Zero Signal Actually Means in Cricket's Review Process

Two of the four camera feeds in that Indore room went black mid-review. January 2026, the Ranji Trophy final, Gujarat against Mumbai — my forty-first first-class match on the domestic panel, and I am sitting in the TV umpire's chair. A low catch had been claimed at slip. The on-field umpire had given a soft signal of out, which meant his inclination leaned that way, and the final call came to me. I needed one frame — the frame where the ball touches the edge of the bat and the gap to the grass is settled. The frame did not arrive. The two feeds were not sitting on the same clock.

I had two routes. The first: build a story out of whatever was visible. The second: record that the evidence was insufficient and send the decision back to the on-field umpire.

I took the second. Nobody called me brave that evening. The protocol says exactly this — when the sync fails, an estimate does not become evidence.

Last week another blank report landed on my desk. Not from a ground this time, from a data pipeline. A Stage-1 deconstruction of an article written about cricket. The title field empty. Article type: Unclassified. Information points: zero. In the entities field sat the instruction the system should have been asking the analyst — “identify from the information points above.” Source quality: not assessed. Time sensitivity: not assessed. Domain label: cricket_world.

The room came back to me. Same moment, cameras swapped for a spreadsheet.

What a blank report actually looks like

A match referee's report has mandatory fields: venue, date, format, over rate, pitch rating, any sanction, a short remark. If a referee files one with the venue blank, the board does not conclude that the ground failed to exist. The board concludes the report is incomplete and sends it back.

What returned here was not an incomplete report. It was a report in which every field was present and every field was empty. The difference matters. An incomplete report at least tells you which match is under discussion. A blank one does not manage that much.

The analysis document was honest about its own condition — it named the failure as a pipeline fault and refused to issue any verdict on the article's content. That is precisely the third umpire's posture. When the feed does not arrive, he does not declare that the catch was not taken; he records that he has no basis on which to decide.

Cricket's information economy does not reward that posture. Closing a blank report as “no story here” is easy. Filling it with a plausible narrative is easier. In twenty-three years I have watched pundits insert an estimate where a number was missing, and watched that estimate become a headline by the following morning.

One thing in this report was real: the document's own behaviour. Zero information points. Zero entities. No format anchor. No timestamp. No source tier. That tells us nothing about cricket. It tells us a great deal about the machine that claims to read cricket. Auditing that machine is the job.

The format anchor is the first rule to break

In cricket, format is the first question and batting average is the second. The order cannot be reversed. A 135 strike rate is respectable for a Test opener and thin for a T20 finisher. An economy rate of eight is expensive in T20 and close to irrelevant in a Test. Powerplay means the first ten overs in an ODI and the first hour with the new ball in a Test. Set-up and death-overs arithmetic is a T20 invention, and importing it into Test analysis is dishonest.

So when the pipeline reports that format cannot be established, it is not logging a technical gap. It is cancelling every headline downstream. Strike rate, economy, situational splits, recent trend, league benchmark — none of these carry meaning until the competition type is fixed.

The danger is that numbers poured into the wrong mould still look clean. In broadcast, the graphics package is settled before the match — the powerplay wheel, the death-overs worm, the Manhattan. Put a fixture into the wrong slot and the analyst picks up the wrong template, and the figures still read with authority. A bad frame produces good-looking numbers. That is the most expensive error available, because nobody catches it.

I keep returning to that Indore afternoon when the sync failed. The problem was not that I had no picture. The problem was that I had two pictures that disagreed with each other. An empty format anchor is the same condition with the pictures removed.

The real evidence is an instruction printed in the entity field

The most damning line in the report is not a statistic. It is a sentence — “identify from the information points above” — sitting where team names, player names, coaches, venues and events should have been. The system's own prompt came back instead of a name.

I sat in the TV umpire's chair, so I recognise the tell. When a feed drops in the control room and an operator accidentally loads the split-screen that was already open, the screen stops showing cricket and starts showing the system. An unfilled schema echoes its prompt. That is not an editorial judgment; it is the emptiness of the template, and it is not hard to identify.

It means the pipeline did not read the article — or read it and extracted nothing. Two different worlds open from here. In one, the article contained nothing to extract. In the other, the article contained plenty and the machine dropped it. Both look identical from the outside, which is why the null result is reliable only as a diagnosis that something broke, not as an interpretation of the piece.

Entity extraction is the gate. Without it, role cannot be assigned — opener, anchor, finisher, new-ball quick, spinner, all-rounder, keeper. And without role, there is no way to decide which metric even matters. Age curves, injury load, home-and-away splits: none of it can be applied to a person whose name was never recorded.

Source tier and timestamp are the report's two witnesses

A board release, a wire service, mainstream cricket media, an aggregator, a social post. Five different witnesses with five different weights. Placing an official release and a screenshot on the same scale is not auditing, it is laziness. With the source tier blank, every conclusion floats, and the reader is denied even the ability to see which claim rests on which witness.

The missing timestamp is more expensive still. In cricket, two-day-old news is usually wrong news. A fitness doubt filed four hours before the toss dies at the toss. A squad announcement goes stale overnight. Without a date, the analysis cannot even report whether the fixture in question is yet to be played, has just been played, or belongs to last season.

One clarification is necessary, because this report carries the same flaw: the absence of integrity-related language in a document is not evidence that the sport is clean. Absence and unavailability are different things, and treating them as one converts an audit into a press release.

Five warnings, kept like a sanction ledger

First and heaviest: silent fabrication. High risk. The reason is blunt — an eight-hundred-word match report can be written out of this null, will be entirely invented, and will clear every downstream check, because the pipeline contains no component that reconciles written words against the source.

Second, equally heavy: this is a broken pipeline, not an empty article. That distinction decides the remedy. The first condition needs a new story; the second needs a repair. Confusing them postpones the repair indefinitely.

Third, medium: the missing format anchor. It must not be filled by default. The correct arrangement is to make format a mandatory field with “unknown” as an explicit, permitted value, so that blankness and ignorance are both recorded and neither is mistaken for the other.

Fourth, the absence of a source tier. Without it, no conclusion can be weighted. Fifth, lightest but most stubborn: the non-standard domain label. cricket_world and Cricket are not interchangeable once routing is automated. That same carelessness will one day land a major story on the wrong desk.

The real risk is not a wrong verdict, it is a credible lie

The convenient read is that zero information points means the article had nothing in it, so the story is dead. One sentence and you move on.

That read is wrong. An empty list of information points is a statement about the machine that was supposed to read the article, not about the article. Extraction failure and content absence are indistinguishable in this payload, and when two different things look the same, an honest official picks the one that can be demonstrated — and what can be demonstrated here is only that the document returned its own prompt.

Second, a lesson cricket taught me. A captain burns a review in the eighth over on a hopeful lbw because doing something feels better than doing nothing. He spends a finite resource and receives nothing. Rescuing a blank report with something general is exactly that. It looks like action. The return is zero.

Third, an institutional point. Procedural silence from a board is routinely read as chaos — a schedule shifted overnight, a venue changed, a sanction issued without explanation. My experience says it is not chaos; it is a fragment of a process nobody published. This report is the same shape. Nothing in it looks wrong, because nothing in it is visible.

Which brings me to the firewall I have to draw, because the strongest temptation in this trade is to become the spokesperson for the institutions you audit. I am naming the document that told me nothing. I am also printing the questions it left unanswered: what was the format, who were the entities, where did the source come from, what was the date. The day those four questions stop appearing unanswered is the day this column stops.

Who signs the next blank report

I resigned once from a chair in Indore, and I have reviewed every comfortable verdict since. The comfortable verdict is the one that can be written without a single question raised. The most comfortable way to write this report was to invent a match.

The repair is not complicated. Make format a mandatory field with an explicit licence to write “unknown.” Attach a source tier to every information point. Date every document. Require at least one extracted entity. Put a validation gate ahead of Stage 2 that refuses to pass an empty payload forward.

Cricket already ran this experiment when the soft signal was scrapped. Removing the presumed inclination left the umpire no hiding place — either you had the evidence, or you did not decide. The pipeline should be held to the same standard.

The match report is never the whole match; the minutes are where the pressure leaves fingerprints. This time the minutes came back blank. The next time a report arrives like this, the question will not be what it said. The question will be who signs it — and whether a single field was filled before that signature.