The Empty Scorecard: What Cricket Learns When the Data Returns Nothing

At ten minutes to three in the morning, the Sector V office was empty and my laptop was running a script I had written myself — eight columns, clean borders, neat headers. When the output arrived, the table was perfectly formed. Every cell said N/A. No error, no red flag, no warning. A completely valid, completely hollow report.

I let my tea go cold that night and kept staring at the screen. I have been reading scorecards for fourteen years, and scorecards have always had blank spaces in them. "Did not bat." "Did not bowl." "No wickets." Those blanks do not lie to you; they tell you something. The maidan taught me that every number arrives with dust on its boots. But this table? Its boots were clean. It was not about a match. It was not about anything.

That is the centre of this story — the biggest and least discussed risk in the sports data industry is silent failure, the kind that does not give you a wrong answer but an answer so tidy and so empty that you mistake it for the truth.

The Room That Changed Cricket's Arithmetic

The room in which cricket gets written changed fundamentally over the last decade. Statsguru had already made the scorecard searchable, but what arrived after 2026 was different — ball-by-ball data, pitch maps, shot zones instead of wagon wheels, expected runs, expected wickets, and inside every franchise a data department usually run by a young man who has never watched a first-class match but whose laptop holds three seasons of it.

I am not an opponent of this shift. I have stood next to the man with a laptop at a Vijay Hazare game whose battery died in the eighth over — he wrote the overs by hand in a notebook and cross-checked them against the scorer's book. That is how real data gets made. My condition has always been the same: I need to know the name of the person who produced the number. Statistics are just footprints; the poem is the animal that made them.

In today's newsroom this happens in two stages. Stage one pulls raw facts out of a source: which team, which player, which format, what score, what date. Stage two lays an analytical frame over those facts — the right benchmark for the format, the home-and-away differential, the economics of an auction, the risk in a rule. The whole frame has one precondition: stage one must return something.

What sat in front of me that night was the entire stage-two apparatus with nothing from stage one. Title: empty. Source: empty. Facts: empty. Article type: "Unclassified" — the fallback word a system writes when it has understood nothing. Time sensitivity: never assessed.

Two Kinds of Blank

One distinction matters here, and it is the spine of the whole profession. A blank in a scorecard is a statement. When the sheet says the number ten did not bat, that is information — the innings was declared, or the target was chased inside ten overs, or the rain came. There is an event behind that zero. I counted the night shifts in scorelines, then learned to hear the tiredness between them. A bowler who did not bowl a single over and a bowler who bowled ten and took no wicket look similar in a column, but one is news and the other is noise.

The blank in a data pipeline is not like that. There, N/A does not mean the event did not happen; it means the system never reached the event. That is exactly where the analyst's room and the dressing room part ways. On the ground, a zero is an event. On a server, a zero is often not an absence of event but a failed search for one.

The most expensive sentence in sports journalism today is not "the numbers are wrong"; it is "the numbers were unavailable, so we estimated."

The Mistake You Cannot See

Three things happened together in that night's output and all three pointed the same way. First, the fact list was empty. Second, the article type was unclassified — the classifier itself never ran. Third, time sensitivity was never assessed. Two independent sub-tasks shutting down quietly without raising an error says one thing: the body text never arrived. The problem was not in the analysis. It was in the fetching.

This sounds technical, but it has a direct sporting consequence. If the pipeline captured a page title and no body — because of a paywall, a JavaScript shell, a bot block — the system marks itself successful. Nothing stops it. And the honest analyst downstream fills every cell with N/A. That honesty is itself a trap, because to a reader a fully formatted, eight-dimensional, well-ordered report looks no different from one that was actually read.

An error that announces itself buys you time; an error that produces a beautiful table survives for months.

Cricket Already Has This Precedent

None of this is new. Cricket has always worked with numbers that were true but incomplete, and those incomplete numbers have changed large decisions.

Take Virat Kohli's England tour of 2026. Ten innings, 134 runs, an average of 13.40. The next England tour, in 2026 — 593 runs at 59.30. Had someone in the winter of 2026 pulled a twelve-month window and declared that this batsman was finished against pace, they would not have been wrong. They would have been incomplete. And in sports journalism the incomplete does more damage than the wrong, because the incomplete arrives dressed as evidence.

Or take 2 April 2026 at the Wankhede. MS Dhoni promotes himself above Yuvraj Singh in the final, makes 91 not out off 79 balls, and India wins the World Cup. No model recommended that decision. The model learned it afterwards, once the outcome already existed. This is data analytics' most honest limitation: it explains the past beautifully and reads the future far less well.

Or take Duckworth-Lewis-Stern. Introduced in 2026, revised with Stern's amendment in 2026, it directly decides who won. Its legitimacy does not come from its accuracy — no model is accurate — but from the fact that its logic is public, challengeable, improvable. The day a data system starts hiding its reasoning, it stops being part of the sport.

In the auction world this is clearest. On 24 and 25 November 2026 the IPL mega auction was held in Jeddah — the first outside India — and the Right to Match card was brought back. RTM is essentially a philosophical instrument: it measures the gap between a market price and a team's attachment. When a franchise official lifts that card, he is saying that three years of memory are worth more than a one-year contract. That number does not get made in a spreadsheet.

I saw this up close on the road to Salt Lake. In June 2026, when ATK announced Roy Krishna, every paper in the city ran the same wire copy. I was alone at the airport at two in the morning. Over the next eleven weeks I watched how eleven thousand kilometres, a new language and a first monsoon add up to fifteen goals. A transfer window is a long way to Salt Lake, and every rumour is a road sign. But the real data in those eleven weeks was not the fee — it was the first day's food, the first monsoon, the first night's sleep. Agents started calling me before the fee was agreed because I had asked about the family, not the number.

Every number has a person behind it; a report that loses the person has lost its most important information too.

The Politics of Silent Failure

Now the part where this technical story meets money.

Fantasy sport, betting markets and data aggregators all live on one thing: fast, complete information. In their systems a blank is a problem to be filled, and the cheapest way to fill it is a guess. I am not offering any prediction of outcomes here and never will; my point is about the shape of information itself. When an empty column is filled with an estimate, that estimate flows downstream, other systems cite it, and three weeks later it is the basis of a decision. The correction never comes, because for a correction someone has to know the column was never filled in the first place.

This connects to a second suspicion I have held, and had confirmed, many times over fourteen years. Data analysts have entered the dressing room, and their conclusions are often cut off from the actual rhythm of the match. It is not because they do the maths wrong. It is because they apply correct maths to the wrong question. They have a twenty-year chart, but they did not see the swelling in the bowler's knee that morning. They have every batsman's shot zone, but they do not know his mother is in hospital. And the day they fill that missing knowledge with an estimate instead of writing N/A, the dressing room's trust is gone.

I remember buying my last nine hundred rupees' worth of ticket to watch the Under-17 World Cup final at Salt Lake in October 2026. That night I started a notebook in which crowd noise and match minutes were logged side by side. It was my first real dataset and it was completely incomplete — the microphone never worked, the rain ruined two pages, and the most important moment of that evening is not recorded on any page. And yet that incompleteness taught me the truest thing I know, the thing I now apply to every report: assuming that what was not recorded did not exist is the biggest lie of all.

The Reversal: The Empty Cell Is the Most Valuable Cell

The industry's reflex is to fill. The editor says the table must look complete. The board says every column must come back populated. Nobody says that an empty column makes a report stronger, not weaker.

My reversal, and I can test it against any scorer's agreement: completeness is not accuracy, and a newsroom that demands zero blanks is technically an estimate factory. A filled estimate and a verified figure look identical — same font, same table, same illusion of authority. An empty cell, by contrast, cannot be cited. That is its entire defence.

Open any scorebook. It is honestly full of blanks — the overs not bowled in rain, the innings that was declared, the runs awarded as penalties. The scorer never fills those gaps with a guess. He leaves them, and it is precisely that honesty that makes his book more trustworthy than any dashboard.

The second reversal is that we blame the wrong man. The analyst did not lie — he wrote N/A everywhere. The real failure belongs to the editor who never asked where the number came from, who made it, what date the snapshot belongs to. Without provenance, every number is a rumour with a decimal point. And most of the large errors I have seen in this profession were not failures to check whether a number was right; they were failures to ask where the number was.

There is another layer here that is easy to miss. The largest part of Indian cricket still sits outside the data that gets debated — the third day of a Ranji match, a Vijay Hazare game in a smaller town, a women's team's training camp, the reserve who carries drinks for a whole season. The data available on these is so thin that it defaults to N/A on its own. And what is not measured drops out of the argument. IPL valuation is not a proxy for the health of the sport; look from the maidan and the club ground outward and the picture changes. The quietest fan in the stadium is often keeping the longest ledger, and that ledger appears in no survey.

The App That Will Never Be Written

Back to that night.

When I closed the empty table, something had become clear. The instruments we built over the last decade to measure cricket are excellent, and they have made me a better writer. But their hardest test is not how much they know. It is whether, when they know nothing, they are willing to say so — and whether we, who depend on them, are willing to print that.

When the audit of cricket's data decade gets written, I hope nobody counts the columns that were filled. I hope somebody counts the places that were honestly left empty. I know that count will never appear in an app, because there is no money in selling blanks.

Even so, the next time a table arrives in front of you perfectly complete, perfectly clean, with no empty cell anywhere — ask one question. How did this table get so complete? Did the information arrive, or was the blank filled in with a good story?

Because a blank on the field is always an event. And a blank on paper is always a choice — to tell the truth, or to build a narrative. The ledger always remembers which one we picked.