The Empty Cell, the Full Scorecard: The Data Fault Cricket Still Cannot Flag

September 2026, Chepauk. Three years into The Hindu's sports desk, and the press box was so full that I was sent into the stands — wedged between a retired railway clerk and a schoolboy, from where you could see the pitch and also see the groundstaff drag the roller.

India were chasing 348. For the last wicket, Greg Matthews trapped Maninder Singh lbw. All out for 347. Only the second tied Test in history. Dean Jones had made 210 in that brutal heat and had vomited on the outfield.

I filed 1,400 words that night. The clerk was named. The schoolboy was named. At the bottom of the last column of the scorebook sat two numbers, and they agreed with each other. The scorebook says tie; the stands still say something else — and that very disagreement is what makes the record trustworthy, because it could be cross-examined against the ground.

Last month a data sheet with forty columns landed on my desk. Clean headers, correct formatting, the full colour code. One entire block in the middle was blank. The software that assembled it certified the sheet as "complete." Nobody filed a complaint.

The most basic rule of my generation of reporters was that the scorebook is the honest witness and star worship is the unreliable one. That honesty has now been outsourced to a machine that green-lights an empty cell. The deepest flaw in cricket's data economy is not the number that is printed wrong. It is the space that is left empty and raises nobody's eyebrow.

Cricket's data chain has reversed itself in twenty-five years. Once it ran one way: match → scorer's pencil → newspaper box → reader. Today it runs like this: match → ball-by-ball feed → data vendor → broadcast graphics desk → fantasy platforms and betting markets → and finally the same reader, who believes he is looking at a score.

Every link has increased its claim to precision while decreasing the habit of verification. DRS's "umpire's call" is the cleanest example of the opposite instinct: the technology measures the ball's path, and at the exact point where its own margin of error begins, it hands the decision back to the field. The technology declares its own limit. I do not see that honesty in the rest of the data trade. Whether it is a DLS recalculation or the WTC points table, the rules are transparent. But the same feed that draws the broadcaster's wagon wheel also settles a market within seconds — and the audit trail for those two uses is not the same.

It is worth remembering where the demand came from. 20 February 2026, Mumbai. At the first IPL auction, Chennai Super Kings paid USD 1.5 million for MS Dhoni — the day's highest bid — and USD 600,000 for Muttiah Muralitharan. The fifty-one-year-old man in that room was covering a transfer window in cricket for the first time in his life, because Indian cricket had never had a transfer window. The first auction paddle sounded like a new season being born.

That week fixed something: once a player's price is settled in crores, the compulsion to measure his every ball, every strike rate, every matchup is created automatically. Statistics had been a by-product of cricket; after 2026 they became its raw material. Whether it is a Right to Match card or an NOC file, a number system stands behind every document. And the same question applies to the women's game after 2026 — whether the WPL's broadcast promises survive contact with the schedule will be known over the next two cycles.

Now the empty cell. It shows up in three forms.

First form — structurally valid, substantively zero. The sheet on my desk had eleven names, eleven roles, and a completely blank performance block. In cricket's language, that is a team sheet with players and positions but no way of knowing who opens and who finishes. If the role is undetermined, then which metric even matters is undetermined — average, strike rate, or economy. A T20 economy below seven is considered excellent; the same seven means nothing for a Test bowler. This is my second-largest worry: the pipeline that passed the empty block also never established which format the data belonged to.

Second form — the number that travels without its context. An auction price and international cricketing strength are two separate ledgers, yet we merge them into one paragraph. A Ranji average and a Test average are different currencies; adding them without stating the exchange rate is dishonest. In fifty years of notebooks, that is where most of my errors have lived — turning one brilliant innings into a career, and ignoring the required rate at the time it was played.

Third form — the water flowing downstream. This is the gravest. The ball-by-ball feed that draws the ball's path on television also fixes a stake in a market seconds later. The first use still carries editorial accountability. The second does not. The ICC's Anti-Corruption Unit exists precisely for this exposure and has more than twenty years of record behind it — the Cronje affair of 2026, the Pakistan spot-fixing case at Lord's in August 2026, the 2026 IPL spot-fixing scandal. The question in all three was identical: who held the information, and when did they get it. Today the information reaches first the person whose name appears nowhere in the scorebook.

A football comparison earns its place here. A goalkeeper who can kick sixty yards but cannot get a hand to a low shot sells for a fortune. Cricket's data world has a cousin of that species walking around: the analyst who can build a heat map in forty colours but cannot tell you which format and which innings the number belongs to. The dazzle of distribution has buried the basic skill.

The outsider's reading is that cricket's problem is too much data, or bad data. Mine is the reverse. The problem is data that looks filled in. An empty cell is a warning — it shouts, it invites the question, it forces you to go ask the scorer. A cell that looks full but carries no source is more dangerous, because nobody questions it. The 2026 tie is trustworthy because it had two witnesses — the book and the stands — testifying from different places. Today's feed has one witness wearing every hat: it produces the data, verifies it, sells it, and dresses it for the graphic.

Blaming the algorithm is easy and cheap. The algorithm is working fine. What failed is the door the material comes through, and the habit that decided a lock on that door was unnecessary. Real reform begins the day data vendors print their limits of incompleteness the way DRS prints the limit of umpire's call. Until then we fall back on the method I used at Eden Gardens in 2026 — on 13 March, India chasing 252 had slid to 120 for eight, the stands were on fire, Clive Lloyd awarded the semifinal to Sri Lanka, and before writing a word I walked the concourse for forty minutes recording what people were shouting, because a collapse without fan testimony is only half-reported. I went looking for the roar and found the silence after — and the silence was the story.

I have only one final test, and it is old: can I stand beside this number in the place I stood when it was made? If not, it is not yet a record. A record is not a number until someone remembers where they stood. I count the crowd the way others count runs: one by one. Every cell deserves the same treatment, one by one, with the question attached — who filled it, and when.

Three things are worth watching from here. Does any board publish its data vendor's audit trail; does the technical standard that halts a report when a field is blank, instead of green-lighting it, become normal within the next two seasons; and does the WPL's broadcast promise hold through its third cycle. The day those answers arrive, we will know whether cricket has started keeping its own ledger again or is still trusting a vendor's formatting.

And the next time someone places a clean, colourful, complete-looking data sheet in front of you, ask one question: which of these cells is empty, and why did nobody say so?