The Over That Was Never Bowled Does Not Score Zero
One afternoon in 2026. My seat was behind the sight screen at the Sylhet...
One afternoon in 2026. My seat was behind the sight screen at the Sylhet International Cricket Stadium — bought with my own money, not granted by the credential office. A hand-drawn field map in my lap, a stopwatch on my phone. Mid-innings I was timing every bowler's run-up separately, because that was the only instrument I had. After the match, a groundstaff member handed me a photocopy of the scorecard.
On that sheet, two zeros sat side by side.

Against one name, a zero — he did not bowl a single ball that day. One row above, against another name, a zero as well — he had bowled four overs, taken no wicket, conceded twenty-four. On paper the two zeros were identical. In reality they were not remotely the same event. One man was never in the game. The other was in the game and failed in it.
That night, writing a four-hundred-word note for a student page, I nearly filed a false sentence — "he went wicketless." The line was ready, waiting for print. Just before I typed it, I remembered: he never bowled. I learned that at the very start of learning the game, from a seat the credential office could never grant me.
Cricket is the oldest sports database on earth. In nineteenth-century England the scorecard was the first structured record — every ball, every run, every dismissal, in its own cell. And within a few decades that scorecard taught itself something remarkable: a vocabulary of absence.
"Did not bat" was written down. "Not out 0" and "did not bat" became two different cells. "Absent hurt," "retired hurt," "no result," "match abandoned" — these words appeared for one reason. The game understood that if you keep an empty cell and a zero in the same place, one day somebody will misread it.
The misreading has already begun.
A single match at a major tournament today generates thousands of data points. Ball-by-ball feeds, tracking, speed guns, spin revolutions, field maps, live win-probability. Every broadcaster, every fantasy platform, every newsroom buys a different version of the same feed. From it come match previews, player profiles, "form analysis," "injury updates."
Cricket is no longer only a game. It is a real-time information economy. And the weakest point in that economy is not in the numbers. It is in the sentences.
In 2026 I spent thirty days inside the bio-secure bubble of the Bangabandhu T20 Cup in Dhaka — five teams, zero spectators, the same hotel, the same dining tables. Twenty-two pieces filed in twenty-six days. A bubble without a crowd teaches you the sound of a game talking to itself — the click of bat on ball, a coach's cough, shoes squeaking down an empty concourse.
With the crowd gone, the easiest weapon in a writer's kit was taken away. Instead I recorded ambient audio. Listening back to those files, I understood that sound is data too. But the bubble taught me something harder: how merciless the demand for a verdict becomes inside a tournament cycle.
A match in the morning, analysis by night. The reader wakes up wanting explanation. "I don't know yet" is an unpublishable sentence during a tournament.
That is where the problem sits. Cricket's information flow now runs through stages. Raw event — ball, run, wicket, venue, time. Then extraction — which facts get pulled from which events. Then analysis. Then publication.
Any of those stages can fail silently. And silent failure is the most dangerous kind.
Imagine that at the extraction stage, for whatever reason, the list of information points comes back empty. No cricketer, no team, no format, no venue, no series, no date. Now ask the analysis stage — who is batting, in which format, on what pitch? The honest answer is one word: unknown.
But the analysis engine has memorised every template of the sport. Test session logic, ODI powerplays, death-over economy, DLS par, DRS, home-away splits, form curves — the whole framework loaded and ready.
Faced with empty input, that framework has an easy path. Put a zero in every cell. The empty cell became a zero. The absence became the absence of absence.
And here the old scorecard lesson returns. The scorecard knew that "did not bat" and "out for nought" are different things. The data pipeline does not yet know this.
As a child I read a match report that called a bowler "superbly economical." His figures were: zero overs, zero runs, zero wickets. Nobody had ever seen him bowl. Every number was true. The sentence was false.
Errors of this kind are not new in cricket, only larger now. Football has long carried a statistic I consider the most deceptive in the sport — possession percentage. Sixty per cent possession can mean ten sideways passes and zero shots on target. The cell is full. The meaning is empty.
In cricket's data world this is now happening at scale. Twenty columns can be filled, thirty metrics can glow green, and what is actually known may be one thing. Or nothing.
A few questions catch the trap.
Is the cell zero, or is it empty? Was the thing measured and the result zero, or was it never measured at all? Collapse those two into one in a database and everything downstream floats free.
Which format, which venue, which sample? If a batter's "average" is built across three formats, it is three averages from three different games, not one. Set a spinner's home economy beside a fast bowler's away economy and the table looks tidy. It decides nothing.
What is the denominator? A strike rate can be produced from a ten-ball sample. The number is legitimate. The decision is not. Small samples generate match hype; in international cricket that hype usually does not hold.
And the last question is the most avoided — did the writer receive information, or receive a blank box where information should have been, and fill it in himself?
There is a structural example of this that is almost comic. A system is instructed: "Extract the names of cricketers and teams from the information points above." Except the list of information points above is empty. The thing you are told to extract names from does not exist.
Cricket has a familiar version of this. The decision is referred to the third umpire, but the on-field umpire never gave a decision. A review is requested, but there was no appeal for a wicket. The machinery is running. The question is standing in the wrong place.
From here comes the most insidious risk of all — silent propagation. If one layer of analysis writes "insufficient information" in every cell, the next layer can read it and write "no risks identified." The two sentences look close. Their meanings are worlds apart.
In cricket's language: rain abandons the match, the book records "no result." Someone reads that and concludes the match took place and nothing happened in it.
This silent propagation has a market price too. Broadcast rights values had been inflating for years; many streaming platforms are now discovering they repeated old television's mistake in a new package — buying more content for more money while having no way to buy meaning. The same structure is forming in the data market. Everyone is buying more feed. Nobody is asking who the source is, when it was created, or who verified it.
This is one place where part of the new technology can genuinely help. Some leagues and boards are now testing verifiable ledgers for ball-by-ball records, digital collectibles and fan-engagement products. The token itself confers no cricket knowledge. The ledger's real gift lies elsewhere — an audit trail. Who wrote what, when, and who later changed it, all on record.
The scorecard was always a ledger. After the match, nobody could quietly alter the runs in the book, because the writing stayed. The data pipeline's problem is exactly here — the book is gone, only the output remains.
For the same reason my suspicion of transfer markets and auction rumours is old. The transfer market is just a rumour with a heartbeat and a deadline. The heartbeat creates excitement; the deadline forces a decision. The same machine runs inside data analysis — time is running out, so an answer must be produced. Whether the answer is true is a matter for later.
The conventional reading goes like this: the system broke, so fix the system. The analytical tool received empty input, screamed about it, and that is a weakness.
My reading is the reverse. **When a system breaks loudly, that is a sign of health. The danger is the system that breaks
