Null Result: When Cricket's Data Says 'Don't Know'

Eighteen June, 2026. Ahmedabad. A seventeen-year-old sits with a hardback notebook open on his knee and writes down every ball of the Champions Trophy final at The Oval. Pakistan make 338 for 4 — Fakhar Zaman's 114 — and India fold for 158 in 30.3 overs, Mohammad Amir taking 3 for 16. That evening he logs 483 deliveries by hand: bowler, length, line, shot, outcome. It takes eleven nights to type them into 483 rows of a spreadsheet his cousin built for him.

When the sheet was finally finished, one column stayed empty. I had wanted to fill it with 'intent' — what the batter was actually trying to do. I had the length of every ball, the direction of every shot, and no method at all for recovering the thought behind it. I left the cell blank. Of everything I have written in nine years, that empty cell is the most honest line in the file. What the data does not contain must not be filled in — it must be left visibly empty.

After that year, the data side of Indian cricket coverage expanded fast. From the IPL auction tables of 2026 to the IPL staged in the UAE in 2026 — 60 matches between 19 September and 10 November, won by Mumbai Indians against Delhi Capitals by five wickets — broadcast graphics, win-probability meters, impact-player arithmetic and auction valuations all arrived on screen. Borrowing a model became easy. The question is who pays the bill.

I have a word I use in every production meeting: the ledger. A ledger means every claim about the game is a debt, and it has to be repaid in logged evidence. 'He doesn't look in rhythm' is an opinion; '34 dot balls and 19 balls on the wrong length inside the first ten overs' is a claim. Indian cricket writing has plenty of the first sentence and too little of the second. But the shortage this piece is about is of a third sentence, the one that says: I do not have the answer to this question.

There is a live place to test that this season. On 2 November 2026, at the DY Patil Stadium in Navi Mumbai, India beat South Africa by 52 runs to win the Women's World Cup. What followed was a queue of announcements — talk of expanding the Women's Premier League, demands for match-fee parity, plans for new domestic women's scholarships. The 2026 win can do for girls what 2026 did for boys, but only if the arithmetic is audited rather than celebrated. I have opened that file. It has dates in it, and figures in it, and a few rows where nothing has been written yet.

That old spreadsheet comes back to me here. The empty cell taught me something no model did: a null result is not a failure; it is a reading. When you count 483 deliveries and find no pattern, the correct conclusion is that there is no pattern — not that the model is broken and a second model should be tried.

In the summer of 2026 I made exactly that mistake. After the 64 matches of the Russia World Cup I reverse-engineered expected goals, then tried to build an 'expected runs' equivalent for the IPL. I hand-coded 13,286 legal deliveries into shot-quality zones. The result: my model overrated death-over six-hitting by 14 per cent. The cause was obvious — the original weights came from surfaces and bowling habits that do not exist in Indian domestic cricket. I re-fitted the weights before trusting a single number. Borrowing a model is easy; paying it back is the actual work.

In 2026 I gave myself another chance. The IPL was in the UAE through the pandemic, the stands were empty, and the stump microphones were the only honest witness. I hand-coded 2,914 fielding sequences into a twelve-variable pressure index — a direct translation of football's PPDA, counting fielding interventions per ten balls. It came back and it disappointed me: the index explained six per cent of wicket variance. The null result came back clean, so I wrote it down anyway. Since that day every report I file carries its sample size and an uncertainty range beside the number, however short the report.

Auction valuation is the fastest-working example of the same disease. A franchise wants a player's 'value' delivered as a single number. That number is usually built from county surfaces, from American ball-sport positional frameworks, and from three seasons of data — and something else again by the time it has been translated onto Indian pitches. Impact-player arithmetic is thinner still, because the rule itself is so new that the sample has not formed. When someone asks me how much a specialist batter is worth as an impact player, my honest answer is: little enough that I will not stake the column on it. The calibration cost of a borrowed model is paid by whoever repeats it without naming it.

And all of this connects to the thing I have written about most — the domestic pyramid's attention. The real casualty of the IPL era is not Test cricket; India is the last country where Tests still sell out. It is the Ranji Trophy's media footprint. That is an attention problem, not a talent problem. The BCCI applied a new domestic match-fee structure from the 2026-25 season — ₹40,000 per day for the playing XI, ₹20,000 per day for the bench. That money matters and should be counted. But paying the fee is half the job; the other half is who logs their deliveries. The hardest team in world sport is India's second XI, and nobody has counted who nearly got there.

Milestone worship runs on the same arithmetic. The hundred is a broadcast format, not a performance measure. Put the match situation beside every fifty — dead rubber, mid-innings run-rate pressure, or a knock that actually changed the result — and half the milestones lose their shine. I have run that exercise twice. The second time the finding was duller than the first, and I printed that too. The ledger does not care what I hoped it would say.

The reverse side of this argument matters just as much, because that is where the trap sits for a writer like me. Our problem is not that we have bad models. It is that the market pays for certainty and not for uncertainty. A graphic reading '76% win probability' travels further than a column reading 'this question will not resolve from the data'. The moment we fill a number into an empty cell to please that market, we commit the same offence we accuse the broadcast desk of committing.

Personally, my biggest risk is the reflex for the inverted take. Contradiction is part of the job, so the brain starts hunting for the counter-intuitive conclusion before the data has arrived. I have made the fix procedural: the hypothesis and its threshold are written down before the numbers are pulled. If the result confirms the consensus, I print that — and state plainly that I went looking to disprove it and could not. It reads like weakness in a byline. It is the contract with the reader.

So the next time an auction table is on screen, or a new broadcast deal is declared, or a WPL expansion is announced, keep a space between the celebratory lines for the column that is empty. The real inheritance of 2 November 2026 is not a photograph of a trophy; it is the file in which the next three years of dates are being written. The question is not whether Indian cricket has data. It is whether we will publish the silence when the data has nothing to say.