Cricket's Biggest Gap Isn't on the Pitch, It's in the Archive

There is a page in my Delhi notebook with eleven rows on it. Each row holds a single word: "none." Eight analytical pillars — format, player, team, league economics, governance, risk, public narrative, industry transmission. A question under each, an answer beside each. The answers are near-identical. Seven years ago, seeing that page, I would have assumed my note-taking had failed. Now I assume something else: the analysis I sat down to write is not short of analysis. It is short of raw material.

This page is not from the night after a defeat. It is from the evenings when I sit with a scorecard and realise the match I am about to write about has no ball-by-ball record, no field-placement frames, no angle from the third tier of the stand. The colleague in the next seat is writing that a player is "back in form." The press box taught me that consensus is often just a missing variable.

Context: an economy where half the transactions carry no receipt

South Asian cricket has a structural feature almost nobody discusses. Far more matches are played than are recorded. Across India, Bangladesh, Pakistan and Sri Lanka, a large share of the annual first-class and List A calendar never reaches a camera. A large share of what does reach a camera arrives through a human scorer typing in real time. And a share of what the scorer types never lands in any server at all — it lives in a book, a club almirah, a district association file.

Bangladesh played its first Test in November 2026, in Dhaka, against India. To compile a proper list of domestic bowlers with 200 first-class wickets since then, you have to reconcile three or four different sources. They do not reconcile. That is not an embarrassing detail. It is a structural one. A system that does not keep complete accounts of its own cricket forces every decision taken outside it — scouting, selection, pricing — to lean on partial inference.

I split the record into four tiers, and the split underpins everything I write.

Tier one is the broadcast match: ball tracking, release speeds, line and length, wagon wheels, field placements frame by frame. Analysis is possible here because the raw material exists.

Tier two is the streamed match: ball-by-ball scoring exists, but no field map. You know a thing happened; you do not know how.

Tier three is the scorecard only: runs and wickets, with no record of which ball, which field, which match state.

Tier four is camp games, warm-ups, age-group trials, club practice matches. No public record at all — and this is precisely where the most predictive information about a cricketer is generated.

The ICC rankings see only the first two tiers. A ranking does not measure cricket; it measures the archive. A player invisible in three tiers will not appear in the ranking — not because he is bad, but because he is unwritten.

There is a direct commercial consequence. Broadcast rights, fantasy pricing, sponsorship valuation — all of it prices the recorded portion. What is unrecorded is, to the market, non-existent. A bowler with forty wickets in a domestic season and a bowler with four wickets on television are not two members of the same profession in the market's eyes. They are two different information classes.

Core analysis: eight pillars, one absence

A long analytical document reached me last month. Eight pillars, sub-queries inside each, confidence ratings attached to every query. I sat with it and found the same answer returning in every cell: not assessable. Format unidentified. Player unidentified. Team unidentified. League unidentified. Governing body unidentified. Risk level undeterminable. Narrative phase undeterminable. Direction of industry transmission unverifiable.

My first instinct was that the document had failed. My second was that it had behaved honestly. Eight pillars with sub-queries and confidence ratings make a claim: an analysis is valid only when every assertion carries an evidence trail. A document that writes "none" instead of inventing a verdict is not an incomplete document. It is a document that resisted temptation.

I recognise that temptation in my own work. In 2026, aged seventeen, I worked as a data logger at the Under-17 World Cup in India and watched England beat Spain 5-2 in the final, Phil Foden wearing number 10 and taking the Golden Ball. I had hand-mapped every half-space entry in a 96-page notebook. The analysis that came out of it rested on a small data set — but every point in that set carried a timestamp.

In Delhi, I learned that a notebook can outlast a broadcast. Broadcasts end. Archives persist. An analyst who does not build his own notebook is dependent on someone else's broadcast notes — which is to say, on someone else's selection of facts.

In 2026, at the Russia World Cup, France beat Croatia 4-2 and Kylian Mbappé scored in the final. I was writing for a Delhi-based new media outlet, working on France's shift from 4-2-3-1 into a 4-4-2 mid-block. An editor told me women do not understand tactics. I sent back twelve annotated clips with timestamps and pass maps. That day is my standing proof that evidence is the only durable rebuttal, and that rebutting without evidence is just conceding more slowly.

The limits of the zone map

A large part of my work is zone mapping — numbering the pitch, labelling the half-spaces, separating the boundary lanes. It is the natural language of my notebook. It also carries a specific danger, and the danger is greatest in my own hands. When the same grid is pushed onto a different format, different conditions or a different sample size, zone data stops carrying signal.

I write my working rules down, because if I do not write them down I start making exceptions for myself.

Rule one: every zone claim must state its ball count. Overs are easy to count; balls are not. A 24-ball zone map tells you where a bowler pitched it that day. It does not tell you where he bowls.

Rule two: do not use zone maps on a sample where the same outcome is more likely to repeat by chance than by mechanism. In practice, two or three innings from one series cannot establish a bowler's line-and-length pattern.

Rule three: change the format and the zone changes meaning. In Test cricket, a fourth-stump line is a pressure tool. In T20, the same line is a run-conceding tool. Same zone, inverted meaning. An analyst who discusses zones without naming the format is not discussing zones. He is discussing his own habits.

Rule four, the least comfortable: change the conditions and the decision inside the zone changes too. On a pitch without seam movement, a bowler hitting fourth stump is not erring — he is doing the only useful thing available. On a green top the following week, that same spot becomes an error. A zone map is not a picture of the pitch. It is a picture of the relationship between pitch and bowler.

Empty stadiums as a control group

In 2026, aged twenty, during the global sporting shutdown, I worked on the Bundesliga's ghost games. On 16 May 2026, Borussia Dortmund beat Schalke 4-0 at an empty Signal Iduna Park. I calculated that the home win rate fell from 43.3 percent before the restart to 33.3 percent after it. That work became a university paper on crowd absence and referee bias.

Empty stadiums gave me the control group I never dared to request. On a normal day, noise and performance move together, so cause and effect cannot be separated. In an empty ground, the noise goes to zero and the cricket remains. Then you can see which changes belonged to the crowd.

The crowd is a variable; the noise is a confound; the silence was data.

Cricket discards its equivalents. Dead rubbers, A-tours, warm-ups, low-attendance domestic finals — we file these as lesser cricket. But where the hype is thin, the quality of decision-making is most visible. Who was brave in a high-pressure match is a question about hype. Who made the correct decision in an empty ground is a question about structure. Scouts should spend more time there, not less.

Dhaka to Kolkata: two corridors, one gap

India's Ranji Trophy has run since 2026 — nearly a century of first-class competition. Bangladesh's National Cricket League and Dhaka Premier League have run for decades. The question is not the age of the tournament. The question is the continuity of the record. Much of the Ranji calendar is now streamed with ball-by-ball data. The DPL is now largely televised. Beyond that lie the matches nobody films — one-day warm-ups, district trials, age-group tours — where a handwritten scorebook is the only source.

I have watched a scorer write every ball of a spell onto a separate page because it had to be typed up for the next morning's paper. Where that file sits today, nobody knows. That is why I call scoring infrastructure rather than data entry. An organisation that will not invest in scoring forgets its own past, and an organisation that forgets its own past repeats the same error under a new name.

My first cricket writing, covering the Wills Cup in Dhaka for Prothom Alo in 2026, taught me the basic discipline: what happens outside the scorecard is the story. Since then my notes have carried two columns — the score on one side, what the field actually showed on the other. When the columns disagree, the score is right and the understanding is wrong.

A crore-rupee decision built on three innings

A transfer rumour is a model with no priors and too many narrators.

In a franchise auction that model becomes visible every year. How much information sits behind a bid? First, matches in the same format. Second, among those, matches that were televised. Third, among those, matches where the player was in the XI and bowled a full quota. Fourth, among those, matches where the returns were memorable. What survives four filters is often ten or twelve games. A season gets built on it.

I will not flip the argument and claim bigger samples always produce better decisions — they do not, because a large sample across changing conditions is still heterogeneous. What actually happens is worse: we decide on a dataset of wildly unequal density and then call the outcome analysis. Too much information on too few matches; too little information spread across too many opinions.

There is a structural parallel with how the Saudi Pro League buys ageing stars. A market that sells the name rather than building the product does not raise the product's quality over time; it raises the packaging. In domestic cricket the same logic holds: a system that does not record its own cricket can still produce talent, but it cannot recognise it. Its scouting becomes market-dependent — buying players who have already been seen rather than building players nobody has looked at.

The invisible portion is the largest one

I once did an estimate for my own use: how much cricket-related information a franchise generates around itself in a year — warm-ups, practice games, internal data — and how much of it is public. In my rough reckoning, more than ninety percent is never published. That is not a scandal. Publication costs money and attracts no audience. But the consequence is that cricket generates a complete picture of its own game and publishes a preferred one.

That is my central professional problem. What I call signal is really selected noise. What the camera caught is signal; what it missed is zero. The game, however, is played in all its parts.

Contrarian angle: the problem is not too little data, it is biased data

Here I turn against my own profession. The data-analytics movement made a promise: more information means better decisions. What actually arrived was the reverse — more information means more pride in the information you preferred.

The mechanism is ordinary. Where cameras reached, we went deep. Where they did not, we did not try. Analysis is now more precise than at any point in the game's history and probably less representative. From ball-tracking we can detect a bowler's wrist-position drift, while not knowing how many overs he bowled in domestic cricket last month.

This forces an uncomfortable question on my own tribe. Those of us who talk most about sample size publish the most 12-ball zone maps. Those of us who mock the eye test build entire profiles from a single television highlight. If I apply one standard of proof to consensus and a looser one to my own dissent, the dissent stops being evidence and becomes a posture.

And a note on myself. My weakness is not an incomplete framework; it is the refusal to publish until the framework is complete. There is always one more variable worth adding, and every addition delays the piece. An incomplete model can be published. An unpublished model helps nobody. My working rule: publish at eighty percent completion and label the remaining twenty as open questions. A reader who knows which parts are inference is holding an honest document.

The thing everyone repeated

Last domestic season a bowler was written off across the board as out of form. The reason was simple: not enough wickets. I deliberately set the bowling table aside and looked only at the field placements in my notebook. Over after over the field was set toward third man while the bowler worked outside fourth stump. Runs came slowly. Wickets did not come.

The absence of wickets was not a form problem. It was a catching plan that accepted boundaries in exchange for breaking the run rate. Whether the plan worked is measured by economy, not by wickets. Nobody checked that column, because that column is unglamorous and tells no story. The wicket count tells a story. The number that tells a story gets repeated most, and the most repeated number becomes the consensus.

This is where my work sits. I am not trying to win the argument. I am trying to identify the variable. I do not chase patterns; I build cages strong enough to test them. The cage is the control group, the timestamp, the sample size, and one plain question: who wrote this down, and why?

What eight empty boxes told me

Three conclusions, after years of working with incomplete archives.

First, an information gap must be declared as a risk. A document that says "not assessable" has not failed; it has identified a risk inside itself — a risk to analytical reliability. Seven years ago I was ashamed to leave a box empty. Now I think the empty box carries the most information on the page.

Second, building an archive builds power, not memory. A district association that preserves its scorebooks will one day sit inside the decision-making process. One that does not will remain the subject of somebody else's account. A large share of power in South Asian cricket sits in the right to write, and the right to write comes from the right to preserve.

Third, my notebook is not for my own use. A year of notes that never becomes a published piece is worth nothing. Eight years of notes are worth nothing if they remain a private hobby.

Not a conclusion — a question for the next match

The next time you see a wagon wheel or a zone map, ask one question: how many balls built this, and who was scoring? The next time a domestic bowler's name surfaces, ask how much of his record reached a camera. Once you know the answer, you will see that cricket's biggest gap is not in the pitch. It is in the record nobody kept.