The Empty Spreadsheet: Indian Cricket's Biggest Number Is the One Nobody Publishes

June 2026, London. I was seventeen and could not sleep after that final at The Oval. A 180-run margin, Fakhar Zaman's 114, and one sentence echoing everywhere — it was a bad day. That sentence stayed under my skin for thirty days. I wrote a small script to parse every ball of the tournament, and the output read: India's top three struck at 78 in the powerplay; Pakistan struck at 95. India's dot-ball percentage through the middle overs was 48. A bad day is a story; a template is a structure. The difference was that the data existed to answer the question at all, because the ICC publishes every delivery. This month I opened another file. It had 84 columns — annual ball-load for Ranji fast bowlers, rest days between spells, injury history, injury type, rehab duration, post-return workload. Five columns were populated. Seventy-nine were empty. This was not a server failure. This was the system. Cricket's data economy is split into two layers. The upper layer is dazzling: ball-by-ball records for every IPL delivery, Hawk-Eye line-and-length data, field-placement maps, broadcast graphics, and the full auction ledger that fills newspapers. The lower layer sits almost in darkness — the National Cricket Academy's injury surveillance register, the bowling load divided between Ranji and India A, selection committee inputs and their weightings, and whatever data infrastructure the WPL did or did not build. After twenty years of watching cricket and ten years of writing about it, I am left with one rule: data that is not public is usually not missing. It simply has never been asked for. Money, of course, gets published. Viacom18 bought the Women's Premier League media rights for 2026-2027 for ₹951 crore in January 2026. India's share of ICC revenue in the 2026-2027 cycle sits near 38.5 percent — a figure that becomes a headline after every board meeting. But the data that ought to determine which fast bowler survives three more years of first-class cricket is recorded nowhere. The money is published; the record that would justify it is not. The visible layer delivers plenty. I pulled every death over from the pandemic years — a delivery-level log of every 16th to 20th over from 2026 onward. The spreadsheet started confessing. Jasprit Bumrah's yorker percentage climbed from 35 in 2026 to 52 in 2026, and his economy in overs 17 to 20 fell from 8.2 to 6.7. In a season played in the United Arab Emirates from 19 September to 10 November 2026, he took 27 wickets. That finding came out of a public delivery log, not a mole. The auction tells the same story. Ahead of the February 2026 mega auction I built a valuation model on Syed Mushtaq Ali Trophy numbers. One name kept surfacing — Tilak Varma, nineteen, a strike rate of 147, an average of 40. On 12 February 2026 in Bengaluru, Mumbai Indians bought him for ₹1.7 crore. He made 397 runs in his first season. Three hundred transfers, one pattern nobody in the room wanted to publish: the market was underpricing even the information that was freely available. Now the part where the spreadsheet goes silent. If I want to know how many balls a fast bowler delivered in a calendar year across Ranji, the Irani Cup, India A and the IPL, I cannot build it. Ranji scorecards exist. Spell-by-spell fatigue records do not. Injury histories do not. Post-rehab re-load data does not. Sample size is the trap here. Death overs are four overs — 24 balls. Across a season a specialist maybe bowls sixty to seventy such deliveries. Three excellent overs and one broadcast highlight reel together manufacture a reputation that franchises then pay for at auction. My rule is blunt: publish the sample size in the first line and shrink the estimate toward the league baseline. Whatever survives that shrinkage is the actual skill. The same arithmetic governs the premium on uncapped seamers. The price at auction is not set by performance on the field but by an inference drawn from four televised matches. And because workload data is unavailable, no franchise can even check how many balls a bowler it is bidding ten crore for has delivered in the last twelve months. Selection inputs live in the same darkness. No selection meeting publishes its weightings — how much for form, how much for role balance, how much for domestic numbers. The consequence is that every selection is analysed as speculation, and speculation drifts into personal like and dislike. This is not a transparency complaint. It is a question about the competitive capacity of professional cricket. The gap is starker in the women's game. The WPL secured a ₹951 crore broadcast deal for 2026 to 2027, but no declared data standard arrived with it — no published player-depth index, no public release calendar for ball-by-ball files. A broadcast promise without a data framework remains half a promise, because as audiences grow, so do the questions. Now the side of this I have to apply to myself. Opacity is not always conspiracy. Building a data pipeline is expensive; maintaining it is more expensive. State associations have neither the budget for scoring software nor permanent staff to run it. Much of this is indifference, not concealment. The other side belongs to my own trade. I had 79 empty columns. I could have filled them with respectable-looking names and estimates and nobody would have caught it, because a report with complete headings and empty cells passes automated quality checks. That is the greatest ethical risk in this work — the pressure to fill the template. So where there is no answer, one has to write: cannot assess. And one thing must stay clear. Cannot assess never means no risk. A bowler whose workload has never been audited is not safe — he is merely unmonitored. In fast bowling, a stress fracture builds quietly first and becomes a headline later. The correlation trap is laid right here. Toss-winning teams, Impact Player usage and net run rate all move together and invite a tidy causal story. Behind all three sits a third factor — the nature of the pitch and the depth of the squad. So I label any mechanism a hypothesis and attach a test to it. Nobody arrives to change my mind. They dismantle it, column by column, and then I concede. That is the whole of my method. The question must be registered before the result. The signal to watch is whether the BCCI publishes Ranji bowlers' workload data in the next cycle, and whether a declared data standard for the WPL appears. Until that file is populated, Indian cricket's biggest number will remain the one entered in no spreadsheet at all.