The Empty Column: Auditing the Asian Cricket Record That Was Never Digitised

In October 2026 I talked my way onto a Delhi digital desk's Ranji rota with a promise: I would hand-code Delhi's entire home season at the Feroz Shah Kotla. Five matches, 1,842 deliveries — batter, bowler, line, length, shot type, false shot. The sheet produced a fat false-shot rate for Navdeep Saini: 21.8 per cent against right-handers. Nobody had asked for it. Delhi won that Ranji Trophy, beating Vidarbha in Indore by an innings, and the sheet remained the only record in the office of how the seamers had actually bowled.

This is not one season's shortfall. Last month I ran a formal audit of Asian cricket's public record — eight dimensions, a clean template, one fixed question per field. Seven of the eight came back null. The single populated field was not a score or a name. It was a geographic tag: cricket_asia. Format blank. Player blank. Team blank. Auction figure blank. Governance question blank. A document that was supposed to say something about cricket had the name of a continent and nothing else.

The audit did not fail. It reported exactly what it was asked. I kept the ledger by hand until the numbers started talking back, and what they said was: the record is not here. This piece is a map of that vacancy — where the data exists, where it does not, and what the absence costs.

The terms of the audit, and why 'insufficient information' is a valid answer

No cricket analysis stands without three things: format, competition type, venue. Before those are fixed, no number means anything. 280 in a Test and 280 in a T20 are the same numeral — one is a score building toward a win and the other is a score building toward a defeat. The same bowler's economy in dewy Dubai and cold Dharamsala are two different professions. When the format is not recorded, the analyst has no route left except to say the material is insufficient.

My habit is to run the dumbest check first. If the scorecard answers it, do not build the model. If the required rate falls out in one line, do not run the simulation. If the sample is three, do not hang a rate on it. That day I asked which format it was. Nothing came back. I asked whether it was a match, a preview, an auction report. Nothing. I asked whether any player was named. Nothing. Three questions, three blanks — and everything written after that would have been inference.

The most useful finding of the audit was that 'insufficient information' is a valid professional answer, not a failure. The error happens when the template's demand for completeness overrides the evidence. A report where every field says 'not available' is telling the reader that nothing was examined — and that is itself valuable information. More dangerous is the report where every field is filled and nothing was checked.

This produces a distinction readers and broadcasters routinely miss. 'No risk found' and 'risk not examined' are different sentences, and reading them as the same is the most expensive confusion on this beat. If there is a fixing suspicion and no facts, the correct line is not 'no evidence found' but 'this was not investigated'. That distinction is ethical and professional at once.

The asymmetry of the record: where every ball is logged, and where every match is lost

Asian cricket's data is not level ground. It is two planets with one bridge, and the people standing on this side of the bridge often do not know what is on the other.

The Indian Premier League has been recorded ball by ball since 2026 — bowling angle, line and length, shot type, fielder position, DRS outcome, all in public databases. Pulling any team's powerplay run rate across ten seasons is one click. International cricket is the same: ICC events, bilateral series, all ball by ball. Analysts treat this as heaven, and therefore assume it is the default.

Then domestic cricket arrives and the map changes. The Ranji Trophy produces well over a hundred matches a season. Most are never in front of a camera. Knockouts are broadcast; group stages are not. The scorecard goes up on the board's website, but it contains what a newspaper contained a century ago: who made how many, who took how many. It does not say where the ball pitched. It does not say how often the batter was beaten, how often an edge was found, how often the ball left the seam and the batter drove in the air.

The Kotla sheet is a picture of this gap. Of 1,842 deliveries, what was recorded exists in no other file. Saini's 21.8 per cent false-shot rate against right-handers did not come from a board document; it came from a 23-year-old reporter's spreadsheet that nobody commissioned. If I had been on leave that season, the number would not exist. And the rule still holds — the largest part of domestic cricket is silent, and if anything fills that silence, it is private obsession rather than institutional process.

Asian cricket's analysis leans on IPL numbers because there is no public, ball-by-ball dataset for the domestic scorecard. What looks like a priority error is actually an availability constraint. Selectors pick from Ranji averages because Ranji has no modern data to pick from.

Thirty-six runs, and every column I trusted going quiet at once

19 December 2026. Adelaide. India bowled out for 36 in the second innings — the lowest total in their Test history, in the same year the world's biggest domestic tournament was played in empty stadiums.

I re-watched every ball for four weeks before writing a word. Short balls came, batters reacted late, a few edges flew, some deliveries landed on exactly the length they had landed on in the first innings and were defended then. Pat Cummins took four, Josh Hazlewood five. What stopped me was not the wicket count.

The problem was that every column I trusted refused to apply. Run-rate analysis at 36? A nonsensical figure. Partnership depth? The largest stand was nine. An expected-runs model? It had been trained on older data, and older data had no sample of an innings like this. Thirty-six runs did not tell me how bad the batting was; it told me how fragile my entire vocabulary was in front of an extreme event.

That is where the lesson I now carry into every number came from. Attach 'against' to every figure or do not publish it. What is the venue, what is par at that over, what happened on this pitch in the first innings. A number without a comparison baseline is not news; it is decoration.

The xG detour: meant to last a weekend, it lasted three years

In June 2026 the desk was short-staffed and handed me the Russia World Cup data blog. I watched all 64 matches with a notepad, logged shots and locations, published a daily xG table. My model had France at 10.4 tournament xG against 14 goals actually scored. The lesson was clean: finishing is noise, chance creation is signal.

The xG detour was meant to last a weekend; it lasted three years. That September, at the Asia Cup in Dubai, I applied the same logic to cricket and built an expected-runs floor for India's top order. On 25 September 2026 Afghanistan tied with India, and that night I understood my model had no term for pressure. Mohammad Shahzad made a century, Afghanistan chased 252, the game went to the last ball. My model could compute the innings. It could not compute what was in the hands of an Associate side that evening.

Borrowing football's vocabulary is the easiest and falsest move available in cricket. There is no cricket equivalent of xG. Name the cricket-specific measurement — seam position, matchup splits, phase run rate, false-shot rate — or admit none exists. A borrowed word delivers the impression of precision, not precision.

Somewhere between the empty seats, the model finally found its voice

From September to November 2026 the IPL was played at three UAE grounds with nobody in them. Everyone wrote about the canned crowd noise. I logged what actually changed.

Coding all 60 matches produced a pattern: death-overs economy rose roughly 0.6 runs an over on 2026. Kagiso Rabada took 30 wickets, Jasprit Bumrah 27. This was the year bowlers could hear themselves think — no noise, no pressure, no reaction. And when I opened my noise file before writing the number, the same question appeared: skill or environment? Death bowling gets easier in an empty ground because a six is not followed by a roar, and the bowler does not bowl the next ball under the weight of it. Somewhere between the empty seats, the model finally found its voice, and the voice said I had spent years ignoring a variable I filed under 'atmosphere'.

I keep a noise file — crowd, dew, travel, toss — a running sheet of context variables I check before crediting any number to skill. When a stat surprises me, my first instinct is not to hunt the story that explains it. My first instinct is to hunt the variable I forgot.

Money and proof: one auction, two ledgers

The IPL's 2026–27 broadcast rights sold for ₹48,390 crore. In the same cycle the Board of Control for Cricket in India's share of the ICC revenue distribution sits near 38.5 per cent — the number that has fuelled years of platform debate. And at the November 2026 mega auction in Jeddah, Rishabh Pant went to Lucknow Super Giants for ₹27 crore, Shreyas Iyer to Punjab Kings for ₹26.75 crore, Venkatesh Iyer to Kolkata Knight Riders for ₹23.75 crore.

Put those three figures side by side and something odd appears. Every auction valuation comes out of a public model, and that model sits on data generated by broadcast and tracking. But what is bought at an auction is not only a batting average. Availability is bought — will he last a season. Dressing-room load is bought — will he fracture the room or hold it. Sponsor pull is bought. None of those variables appears in a public dataset, which is why a bid that looks irrational against the public model is often rational against a private one.

The auction is a pricing problem, not a soap opera, and the premium is what gets paid for the thing no public table can show. That connects directly to the domestic shortfall. With no credible modern record of domestic performance, a franchise has one route left to value a young player — what was seen, heard, remembered. Reputation-based valuation is always expensive.

One more distinction the coverage has collapsed too fast. The IPL's real casualty was not the Test match; it was the Ranji Trophy's evidentiary base. Tests still sell in India — Ahmedabad, Chennai, Mohali fill. What collapsed was the documentation of the country's oldest domestic tournament: un-digitised scorecards, coverage confined to knockouts, no public dataset on which a selection argument can be built. The remedy is not nostalgia. The remedy is coverage infrastructure.

Coverage without a backbone: the women's game

The Women's Premier League has run since 2026, and in 2026 India hosted the Women's ODI World Cup. The rise in coverage volume is real and welcome. Volume and foundation are two different things, and the second is trailing.

WPL seasons are short, matches per team are few, and every player's sample shrinks accordingly. Separating a brilliant campaign from an average one on that sample is not a statistical exercise, it is guesswork dressed as one. Without split-level data — against spin, against pace, in the powerplay, in the last five overs — coverage inevitably becomes narrative. It tells stories where numbers should have been asked for.

The women's game's legacy test will be collection, not sentiment. In ten years the judgement will rest on whether anyone holds every ball of a 2026 match in Dubai, not on how many articles were written that year. The fifty-year shortfall Asian men's cricket built for itself is available to be repeated here. That opportunity is expensive.

The commentary box is the last unwon argument

The truck knows. The microphone does not say.

Cricket's tactical argument was settled by data desks twenty years ago. Nobody needs to be told a powerplay wicket is valuable. The live fight is broadcast. On my screen this bowler's false-shot percentage has gone from 9 to 31 across five overs; on commentary he has 'lost his rhythm'. One sentence measures, the other guesses.

From that habit comes the thing that irritates me most: making the hundred the story and leaving match state in the margin. A hundred is a historical fact. It does not answer whether the team was 180 for 5, whether 42 overs were gone, whether 90 were needed. Situation first, milestone as residue. The order inverts whenever commentary has to fill air.

Where controversy went: from the pitch to the review room

DRS did not end disputes. It moved them. The argument used to be on the field, about whether the ball hit bat. It now happens in the review room and on the grey line of the rulebook called umpire's call.

There is an analytical consequence we skip. When the decision moves to technology, the question stops being 'what happened' and becomes 'which margin is admissible under the rule'. In my tracking the number that should matter is the wicket; now three or four semi-decisions taken inside the tolerance of the technology enter the match analysis, and they are deliberately ambiguous. That ambiguity is not an accident. It is a design: it shifts accountability from the player and the umpire to the system, and the system is harder to question.

Why the eight fields were empty

Eight dimensions — format, player, team, league, governance, risk, narrative, industry. Seven returned not one populated field. That is not a coincidence of numbers; it is an absence of information in the input.

An empty format field does not mean the format was undeterminable. It means the material under audit never stated a format. An empty player field does not mean no player existed. It means nobody named him — and unnamed sourcing ('a player said', 'a senior said') is growing in cricket writing.

The empty team field is the most expensive, because the single most consequential variable in team assessment is the home-away differential. No strength claim survives without it. Neutral-venue status is routinely dropped from media summaries and changes the entire meaning.

The league field being empty means one simple thing: no amount, no currency, no cycle. Indian cricket figures arrive in crore and get compared across incompatible cycles — the 2026 auction purse against the 2026 purse, the 2026 rights deal against the 2026 deal. That comparison is an analytical offence.

Governance is the most dangerous blank, because that is where unexamined risk and examined absence get confused. No-objection certificates for overseas leagues, central-contract discipline, the ICC's red line on government interference — none of these had a date or a decision attached. Without a date and a decision, governance analysis does not exist.

The contrarian part: this is not a technology problem

Here is where the comfortable conclusion has to go.

The comfortable conclusion is that domestic cricket has no data because the technology is expensive. That is false. A high-definition camera, a tracking system, one coding operator — that is small change against a Ranji budget, provided anyone counts it as spending. The real problem is that nobody pays a price for its absence. The board loses nothing. The selector loses nothing. The broadcaster gains nothing. When a system contains no penalty for a job not done, the job does not get done. This is an incentive failure, not an engineering one.

The second comfortable conclusion is that an 'empty report' therefore contains no risk. I said it earlier and will say it plainly: this is a category error. If a matter was not investigated, the report cannot say 'no risk'. It says 'unexamined risk'. Unexamined risk is not smaller than risk. It is larger.

The third I have to write against my own reflex. My brand is the counter-intuitive finding, and brands calcify into tics. There is no inversion here. The boring finding is the true one: Asian cricket's data base is thin, that says nothing about the quality of the data that does exist, and it is not evidence of weakness in any team, player or league. Absence is not evidence. It is only absence.

The easiest leap is from correlation to cause. One could conclude that domestic cricket lacks data therefore lacks talent. The two are unrelated. Saini's 21.8 per cent false-shot rate existed that season whether or not I recorded it. The shortfall in the record belongs to our eyes, not to reality. A number is only as honest as the person who typed it, and only as complete as the process that typed it.

What to watch next

I will not predict a winner. I will say which signal would change the whole analysis.

The first is in the Ranji scorecard. The day group-stage matches carry public ball-by-ball detail, every selection argument changes ground. This is a logging question, not a broadcast one — no camera is required for a coder to write down line and length.

The second is in WPL data. If split-level numbers become public over the next two seasons — against spin, in the death, domestic versus overseas — women's coverage can move from narrative to analysis. If they do not, the current boom will be a paper boom that fades with a generation.

The third is in the broadcast. When commentary states the situation before the milestone — five overs left, six down, who has how many overs — the data desk will have won. Until then the gap between the truck and the microphone stays open.

The fourth matters most. The next time a report says 'no risk found', ask what was examined. I did not find the pattern; it waited for me to finish counting. This time the count ended and produced one empty column. Next time, at least, we will know what that column is called.