The Empty Field: Where Cricket's Data Falls Silently

The three a.m. feed never announces itself. It simply drops, quietly, and by morning the highlights package plays and nobody notices. In the winter of 2026-03 I was sitting in a Mumbai production house timestamping every ball of the Ranji season — twenty-six years old, an MS in Kinesiology, and a spreadsheet nobody had asked for. One night the satellite feed went down for eleven minutes. What happens in eleven minutes of cricket became clear to me the next morning when I opened the log: the scorecard was complete, the over count was complete, the bowlers' names were all there — and where the fielders had stood during those eleven minutes was recorded nowhere.

Last week a scorecard of that same kind came back to my desk. A structured analysis document — eight dimensions, from format to governance — and not one of the eight fields was filled. Title: not available. Source: not available. Information points: zero. What had been called a "deep professional analysis" was in fact an honest confession: we do not know, and we will not guess.

This report is about that confession. The biggest crisis in cricket data is not false data. The crisis is that when a field comes back empty, the industry hands the job of filling it to a story — and the story is always ready.

Cricket's information chain is a four-storey building, and every floor is built on the assumption that the floor below did its job. At the bottom is the source — scorecard, feed, ball-by-ball log. Above it is the extraction floor, where names, numbers and events are pulled out of the raw log. The third floor is analysis, where phase maps get drawn and decisions get reviewed. The fourth floor is publication and market — commentary, columns, fantasy, and the live feeds that betting companies pay for.

No floor in this building ever announces its own mistake. The second floor comes back empty-handed, the third floor passes that empty hand along with the words "insufficient information," and on the fourth floor somebody fills the gap with a sentence like "crumbled under pressure" and the copy is filed. The reader never learns that the sentence he took for analysis was plaster over an empty field.

My own habit is a product of this building. During the 2026 World Cup I took the night shift on the Johannesburg feed myself, because nobody else wanted it, and I built a spreadsheet of India's bowling phases that nobody had asked for. In those days the archive was on tape, the tapes had labels stuck on them, and the date on the label was written by hand. When a label tore, the tape stayed on the shelf but was of no use to anyone. The file-based archive gave us the convenience that nothing gets lost. It also gave us the inconvenience that nothing gets verified — because the file still has the same name even after the data inside it has changed.

23 March 2026, the Wanderers. Australia 359/2 — two wickets in fifty overs. India 234 all out in 39.2 overs, beaten by 125 runs. Those three numbers will still be found in every cupboard. What will not be found is this: where India's fielders stood in the first fifteen overs, and how many feet they had retreated between the fortieth and the fiftieth. That was the weakest column in my spreadsheet, because that information was simply not in the feed. I left it blank.

That blank column taught me what no academic course did: an empty field is an honest document, provided it is allowed to look empty. The trouble begins when someone at a desk decides the empty field cannot be shown to the reader, because readers do not like "incomplete."

In 2026 I got my first national byline out of Johannesburg — 24 September 2026, the World T20 final. India 157/5, Pakistan 152 all out in 19.4 overs, beaten by five runs. Pakistan needed thirteen off the last over, Mahendra Singh Dhoni gave the ball to Joginder Sharma, kept his fielders up, and Misbah-ul-Haq went for the scoop and found Sreesanth. I redrew that over eleven times. The eleventh map was the one where I also recorded the fielder who did not move. That map had one empty field in it — "why he did not move" — and that empty field told me more than the other ten maps combined.

There is a way to read an empty result, and it is the same way I read a match: run the phase map backwards. When a match is lost, I walk back from the last over and find which block turned the result. When an analysis comes back empty, I walk down from the last floor and find which floor came up empty-handed. I ran the phase map backwards, and the pattern confessed.

The first disclosure is that "insufficient information" buries three completely different things under one label, and the burying is the real fault.

Case one: the information was never in the source. This is the cleanest case. If the original material is a photograph and a caption, then extracting a bowling economy from it is impossible, and that impossibility is not a defect, it is a fact.

Case two: the information was in the source, but the extraction floor failed to catch it. This is the most dangerous case, because here the emptiness is lying. The source is fine, the document is fine, and still the field is empty — because the parser never received the text, or the feed was broken, or the language was one the model did not recognise. In this case "insufficient information" does not mean "there is no information." It means "we stopped looking."

Case three: the field does not apply to the subject at all. Asking for a "powerplay split" in a Test review is exactly this. The emptiness is not an error, it is a category mistake.

The cure differs in each case — change the source in the first, fix the pipeline in the second, change the framework in the third — but all three get written with the same word. When one word is the name of three different diseases, the wrong treatment is guaranteed. The document that arrived last week repeated that one word across all eight dimensions. Reading it, nobody can tell whether the source was empty, the pipeline was broken, or the question was meaningless. The document did its first job — it did not guess. It did not do its second — it did not diagnose.

The second disclosure comes out of cricket's own history. The 2026 log already had this season written in the margins — and the most important part of that log was the part that was not in it.

That Wanderers final is still remembered as "a bad evening." I do not call it a bad evening. I call it a match whose structure was already written, and whose clearest evidence is an empty field. Australia lost two wickets in fifty overs. Two. Across fifty overs. That does not mean they batted well — it means India's bowling plan had no block at all in which wicket pressure could be built. When two wickets fall in fifty overs, the scorecard says 359/2, but what it is actually saying is that the opposition was never put in a position where it had to take a risk.

Now the field that was empty. My spreadsheet's column for "field positions, overs 40-50" was blank, because the feed did not carry it. The consequence of that blank was that for twenty years we called that defeat "pressure," because a story about pressure needs no data. Had the column been filled, we might have seen how far the field had retreated in the last ten overs, and whether that retreat was the captain's decision or forced by the bowler's line. That information is still unavailable. And the place of unavailable information is always taken by a story.

The third disclosure is at player level, and here I have to be most careful. When a player's name is not recorded in an analysis, commenting on him is not technical analysis, it is inference. I do not infer. But I can say what happens when the field carrying the name is empty.

Joginder Sharma's last over is the best example, because two stories circulate about that over and both contain the player's name, not the number. The first says Dhoni took a risk and won. The second says Misbah played the wrong shot. Both stories are true and both are useless, because neither records where the fielders were. I drew that over eleven times, and each time something new appeared — but the most important thing was what was not on the map: the fielder who was not moved. Keeping fielders up to defend thirteen is one strategy. Keeping fielders up and then not moving them is a second strategy. The difference between the two shows only on the map, never in the commentary.

The fourth disclosure is at team level, and it is the least comfortable. When there is no solid information about a team at all — no ranking, no home/away profile, no batting depth, no bowling combination — then there is only one honest form of team analysis: silence. But silence does not sell. So in its place arrives a sentence like "a balanced side, but the middle order lacks stability," which applies to every team and says nothing about any. That sentence is the most dangerous form of the empty field, because it does not look empty.

The fifth disclosure is about leagues and money. The biggest structural change in cricket over the last decade is that the franchise scout has replaced the zonal selector, and the scout's main weapon is data. When the data is right, that change is an improvement. When the data field is empty and somebody fills it, that change produces a recruitment system in which a player is chosen not on his performance but on a model's estimate — and that estimate is never audited. This is where I fear my own habit. Franchise cricket's data is the finest and richest I own, and therefore the most seductive. Dropping it into Test or ODI analysis is easy, because the models look familiar. But the moment the format changes, the same model starts lying. Format leakage is the most common theft in analysis, and it is never caught because it looks good.

The sixth disclosure is about governance, and it is the least sporting and most institutional of the lot. The Ranji Trophy's status has changed quietly over the last fifteen years — its place in the calendar is the same, but the journalist watching it has become rarer and its footage has become scarcer. The day a tournament stops being logged is the day its players stop having selection value on record. A claim of "merit" in selection can only be falsified while its data exists. When the field is empty, the merit claim becomes unfalsifiable — and an unfalsifiable claim is the single greatest corruption-enabler in cricket.

The seventh disclosure is about the India-Pakistan fixture, which I file under nostalgia marketing rather than sport. The commercial value of that fixture rests on the story built around it, and that story draws less and less support from present numbers. The record in ICC tournaments over the last two decades has been one-sided, and yet the "evenly matched contest" frame arrives before every meeting. The frame is not false — it is merely data-hostile. And when a frame is data-hostile and still sells, it means somebody in the market has already filled that empty field.

The eighth disclosure is about the framework in which this document was written. Eight dimensions, four or five sub-fields in each, a risk flag in every sub-field. That is a good framework, and that is its problem. When a framework is so complete that every question has a field waiting for it, the pressure to fill the empty field rises — because an empty field exposes the framework's own fault, and nobody likes admitting the framework's fault. A good framework has room to write "this field does not apply." A better framework has room to write why it does not apply.

Now the part where I have to stand against my own argument.

My first argument was that an empty field is honesty. That is half true. The whole truth is that an empty field is safe only as long as it looks empty. The day it starts looking filled, it becomes more dangerous than a lie, because a lie can be challenged and a filled field cannot.

I have seen this with my own eyes. In one scorecard the catch record was incomplete — some catches were logged against "the team," not against a player. The next season that same incompleteness had become a "fielding efficiency" number, and on the strength of that number a player was labelled a poor fielder. That player had probably not dropped a single catch. His only fault was that his catches were not entered in the column they belonged in. This is not a cricket story. It is a data-management story, and it happens every year. A no-ball is never just a no-ball; it is a ledger entry. And the day that entry lands in the wrong account, a career's arithmetic changes.

The second argument I want to take back is that an empty result is a finding in itself. That is true only under one condition: when the empty result follows an investigation, not a lapse. Last week's document was proud of its honesty, and there was something hollow in that pride. It wrote "insufficient information" across eight dimensions but never said why the information was missing. A doctor who says "I do not know what is wrong with you" and a doctor who says "I do not know what is wrong with you, and I did not examine you either" are both honest, but only one is trustworthy.

The third argument, and the most important: is it possible there was no story here at all? I ask this question first every time, because my greatest weakness is that I feel compelled to produce a counter-intuitive read — because my reputation was built on counter-intuitive reads. When a source is empty, the easiest thing in the world is to turn the emptiness itself into a "big revelation." "The data pipeline is failing" is a very good headline. But it is also possible that on that day one file simply opened from the wrong location.

So I state the null hypothesis first: perhaps this is not an industry crisis at all, merely a technical glitch. If that is the case, then writing three thousand words about it is itself an example of the disease whose cure I am demanding. I accept the contradiction, and it is the reason I have invented no cricket number in this report. The dates that exist are in the scorecard. The ones that do not exist, do not exist.

The fourth argument is a direct attack on my own profession. Two generations of Indian cricket journalism ran in a system where the press-box seat, the squad announcement and the interview all came through the board's door. In that system, filling an empty field was a professional skill. A journalist without a document had "a source close to the team," and that source was never verified because verification needs documents and documents need independence. I am a product of that system and I am its critic, and both things are true at once. Whenever I write a board-critical piece, I have to check that I have not softened a claim. Softening means filling, and filling means that next time something more will be filled into that same field.

The fifth argument is the most uncomfortable. When I say the market puts a price on an empty field, I am also assuming the market knows about the emptiness. Perhaps it does not. Perhaps the market is also trusting the filled field that some sub-editor filled at two in the morning. That is more frightening, because there is no conspiracy in it — only a chain in which every link assumed the next link was reliable.

The next time you read a match report, or watch a "deep analysis," check three things and let the rest go.

First: the date of the source. If the document does not say which day the information belongs to, it cannot tell you whether it is still true. Second: the names. If no player, no team, no ground is named in the analysis, then the analysis applies to nobody — it is merely well written. Third: the field that is empty. Every good report should have at least one empty field, and it should be visible. A report in which everything is filled has had something filled into it.

My verification list for the next match is ready, and it is three lines long: what time the feed started, what time it broke, and who was standing on the field when it broke. The answer to the third line will not be found in any scorecard. It will only be with the man who was watching the screen at that moment and wrote it down. The most reliable data in cricket is still the data somebody recorded without being asked.