It Returned Zero: The Asian Cricket Coverage No Machine Can Read

There were eleven fields on the screen. Ten of them held three characters — N/A. The eleventh glowed with a single tag: cricket_asia. That was all. No title, no source, no date, no author stance, no stated purpose, and above everything else — information points: zero. A page sent in to be understood as some piece of Asian cricket reporting came back saying only that there was nothing there.

This is a new kind of news for me. In forty-eight years I have seen blank scorecards, rain-washed days, press conferences where the captain arrived, said two sentences, and left. But a blank page that announces its own blankness — that is a first. And this report is about that announcement.

December 2026. The press box at Feroz Shah Kotla. The third Test between India and Sri Lanka, five days, Delhi air so bad that players were wearing masks on the field. Suranga Lakmal left the field to vomit. The first voice note I sent from the Kotla was mostly me breathing. If that sixty-to-ninety-second fragment had been handed to a machine, it would probably have written: "Noise, unclassified." But the whole match was in that breathing — the coughs, the empty stands, a horn somewhere far off, and that strange silence that falls when ten thousand people hold their breath at once.

So when an audit report says "the Stage-1 result is null," my first question is not about technology. It is about the room. Which room became so empty that not one fact came out of it? And if the room was empty, what exactly are we printing and calling cricket coverage?

***

The first pass

The biggest change in Asian cricket over the last decade did not happen on the pitch. It happened in filing. The volume of matches has grown beyond what any human desk can read. India, Pakistan, Sri Lanka, Bangladesh, Afghanistan, Nepal, Oman, the UAE — inside the Asian Cricket Council's orbit there are so many series, so many leagues, so many qualifiers moving at once that keeping a reporter on every scorecard is an economic impossibility. The pipeline was born out of that impossibility.

The first stage of a pipeline is simple: pick up the page, pull out the facts, attach a category. Who is playing, when, where, who said what, who claims what, who denies what. That is Stage-1. Stage-2 argues over those facts. Stage-3 writes the piece. The whole building stands on the first brick.

The report we are discussing is about that first brick — and the brick is missing. Title: N/A. Source: N/A. Type: unclassified. One-line summary: empty. Author stance: N/A. Purpose: N/A. Information points: zero. The one thing that survived was the domain label: cricket_asia.

A tag is not coverage. That is the spine of this whole report. You can call a page "Asian cricket," but that tells you nothing about which corner of Asia it came from, which board's pressure shaped it, which election year it was printed in, and which player's price it was used to settle.

I learned this work in rooms, not in spreadsheets. I learned the transfer window from a room full of men raising numbered paddles. January 2026, Bengaluru, the IPL mega auction. Five hundred and seventy-eight players shortlisted, one hundred and sixty-nine sold, eight franchises sitting with purses of eighty crore each. Ben Stokes drew a bid of twelve crore fifty lakh. I sat in that room with Rajasthan Royals' analysts while the Russia World Cup played on a hotel television.

What a machine can never capture in that room is the face of an unsold player. A name is called, two bids come, a third does not, and the paddle goes down. Every paddle in that room lifted somebody — somebody's family, somebody's village, somebody's next season. In Stage-1 language this event is one line: "Player X unsold." As an information point, that is correct. As truth, it is incomplete.

***

A form with nothing to fill it

There are two sentences in the audit I read three times. The first: "Entities involved: identify from the information points above." The second: "Source quality: judge from the source fields of the information points."

Both sentences cut themselves. There are no information points above, so identify from what? There are no sources, so judge whose quality?

This is not poetry. It is bad design. The form was built, the fields were built, and it was assumed that there would be something above them. It reminds me of the whiteboard in the Kotla dressing room before a match, where the coach writes the plan. If the board has no names on it and underneath is written "insert names as per role," that is not a strategy. That is the appearance of process.

The most dangerous thing in cricket coverage is not a lie. It is the appearance of process. When a desk prints forty reports a day and twenty of them were written by someone who was never in the stadium, the reader cannot tell the difference. The sentences are the same. The structure is the same. The only thing missing is that there was no eye behind the sentence.

Which is why I do not read "information points: zero" as bad news. It is an honest result. A machine that writes "nothing found" when nothing is found is a thousand times better than one that invents an inference to fill the gap. The audit did exactly that — it refused to assume.

But honesty has a price, and the price is paid in the next step. If I sat down to write a "deep analysis" on this empty report, I would have to pull some story out of my forty years of memory about Asian cricket, treat it as if it were this page, and then praise or attack it. You would read it as analysis. That would be the worst crime available — because crimes of that kind are never caught. With no facts, there is no contradiction.

***

What the machine cannot hear

If I had to name what goes missing first in automated Asian cricket coverage, I would name three things. All three come from my own experience, not from a model.

First — the second XI. When a system decides what to cover, it always picks what gets searched most. Which means the national team, which means the star, which means the IPL. The November round of the Ranji Trophy, where a twenty-three-year-old takes five wickets for the first time and is not named in the paper the next morning — none of it is recorded. This is not a small omission. The entire structure of Indian cricket stands on that second XI. The day that line disappears from coverage, we will not even know what we lost, because there will be nobody left to report the loss.

I see this through the lens of the auction room. One hundred and sixty-nine names were sold at that auction, but to reach those one hundred and sixty-nine, more than four thousand players had already played domestic cricket. Not one of those four thousand exists in any analytical system, because no camera watched their matches. A pipeline can only read what has been written, and there, nothing was written.

Second — sound. This is my strongest testimony. Deadline taught me to write fast; the voice note taught me to write true. And a large part of truth is the sound of the ground. The hollow echo of the Kotla at eight in the morning. The Sher-e-Bangla in Dhaka when the crowd shouts "Bangladesh" as one and the stadium's roof seems to shake for a moment. The Pallekele in Kandy when hard rain comes and every umbrella opens at once. The deep silence at Karachi's National Stadium that lasts three seconds after a wicket falls, and then the noise.

Not one of these fits into a Stage-1 field. What fits is: "Attendance: not available." But that is the real record of a match. No scorecard will tell me that on the fifth day, in the afternoon, when Dinesh Chandimal survived an lbw by a hair's breadth in the twentieth over, somebody in the press box stopped breathing. That breath was the story.

Third — the political geography of the auction room. The three things a franchise's analysts were doing that night — the arithmetic of the purse, the record of a player's injury, the ledger of relationships with the board — none of them get logged as an "information point." Yet they decide who plays. I watched a bidder hold his hand back simply because he calculated that this player's price would rise by next season. That logic never appears in a scorecard. And if you do not know it, you are not writing about an auction — you are printing a list of bids.

***

Six cricketing cultures in one tag

Now to the single living cell: cricket_asia.

The label may be correct. It may be a default — some systems are built so that when they understand nothing, they fall back to their widest category. And it may be a fuzzy match on a single word pulled from a URL fragment. All three are possible, and with not one information point, I cannot choose among them.

But I will say this much: "Asian cricket" is not a category. It is a geographic convenience. Any reporter who has set foot on these grounds knows that this is not one game. At least six games are running at once.

In India the question is administration — the board, the selectors, the calendar, and the real meeting that happens in a corridor and produces no minutes. In Pakistan the question is security and domestic habit — home matches are not played at home, so a fast bowler first meets a foreign pitch after he has already become an international. In Sri Lanka the question is migration — bowlers leave for T20 leagues abroad and the domestic season hollows out. In Bangladesh the question is patience — what a team learns in fifteen years is forgotten by the next series. In Afghanistan the question is refuge and exile. In Nepal and Oman the question is how a small board asks for space in a big board's calendar.

Put those six worlds into one tag and the analysis will always be done on the largest world's terms. And the largest world is India. So every Asian story ends up as a story about the Indian team's preparation. That is not the machine's mistake. The machine is doing what it was taught. The mistake is ours, for teaching it this way.

I have to record one more thing here, because it is the biggest theft in my profession. Over the last two decades, data analysts have walked into dressing rooms — and their conclusions are often cut off from the actual rhythm of a match. I have sat in meetings where a bowler's average is quoted and not one man in that room mentions that this boy cannot bowl a third spell in August humidity. The average is right on paper. On the field, it is a lie. That is the difference between a number and a witness, and that difference is the subject of this whole report.

Football went the same way. The inverted winger homogenised the game — everyone cuts inside and shoots, and the old winger who hugged the touchline has vanished. In cricket, that touchline is the reporter who arrives at eight in the morning to see which side of the pitch has more grass. He has no metric. So he was removed.

***

Perhaps the room really was empty

Now I want to stand against my own argument, because if I do not, this piece becomes nothing but a lament about machines, and lamenting machines is as useless as lamenting the weather once was.

My contrarian case is this: the null result may not be the page's fault. It may be the fault of what we assume a page is. We proceed on the belief that if a page loaded, there is reporting inside it. That belief is false. In the Asian cricket market, hundreds of pages are produced every day with a headline, a photograph, two paragraphs of agency copy, and the rest written for SEO. Hand a page like that to an honest extractor and it gives the correct answer: there is no information here.

Which means the machine did not fail. It diagnosed. It said what an editor would have said twenty-five years ago when he threw the copy in the bin — "there's nothing in this." We removed the bin, so now a system does that job, and it frightens us because it writes N/A and puts it in front of our faces.

The second contrarian point is sharper. Ranji Trophy coverage did not die when the machines arrived. It died when we started leaving it — when newspapers stopped sending reporters to domestic matches, when television pushed it into a corner of paid streaming, when we writers decided for ourselves that "readers don't read Ranji." The machine only occupied the space we vacated. Every paddle that went up at that auction went up at the cost of a domestic match's coverage. Nobody itemised that bill, because that bill is invisible.

The third point is the one that stings most, and it is the biggest lesson of this audit: when a system fills an empty space, it covers it with clean fields. And clean fields are the most dangerous thing there is, because looking at them you believe the work is done. An incomplete report that says "I don't know" is better than any report where every field is filled and not one fact is inside.

I learned that once, at great cost. At fifty-five I walked away from a print desk of twenty-two years. It was not bravery, it was panic — at that Kotla match I recorded forty-seven voice notes instead of filing eight hundred words, and I felt that what I was writing was not present in the room. What was present was the coughing, the empty stands, and the horn. Since that day I have put one question to every piece: read it aloud — can you hear the stadium? If not, it goes back.

***

I put that question to this pipeline now, and the answer is clear: no, you cannot hear it. A system that cannot tell me which country was on that page, which tournament, which year, can tell me nothing about cricket, however correct its tag.

But I will not end here, because mourning zero is easy and looking past zero is hard. So what next?

First — ask for the raw text, not the result. Until the actual text, title, source and date of the page are available, any "deep analysis" is inference, not analysis. And passing inference off as analysis is the oldest disease in this trade — it used to be done by hand, and now it is done by systems.

Second — where extraction keeps returning zero, look at the source before the technology. JavaScript-rendered pages, copy behind paywalls, non-Latin script and mixed-language content, and reports published as PDFs or images — those are the four places this pipeline comes back empty-handed again and again. A large share of cricket coverage in India, Pakistan, Bangladesh and Sri Lanka falls into exactly those four categories. Which means a system built for English websites will call half of Asian cricket zero, and we will read that as a finding.

Third — and this matters most — write the second XI back. If we want the next pipeline to be able to read something, somebody has to write it first. In the November Ranji round, in Dhaka's domestic tournament, in a club match in Kandy, in Afghanistan's domestic series — somebody has to sit down and build the scorecard that a machine can later read. That is the least popular and most necessary part of coverage, and it is exactly the part the auction's glare taught us to ignore.

One last thing. What struck me most in this whole affair is that of eleven fields, the eleventh — cricket_asia — was the only one containing a word. Everywhere else it was N/A. Which means the system knew this much: that this was Asian cricket. It just did not know what was in it.

There is no better portrait of this profession. We all know we are covering Asian cricket. The question is what we are writing inside it — and whether that, dropped into an empty field, would stay empty too.