The Empty Cells: What Cricket's Data Ledger Never Records

Last month a file arrived on my desk in Brisbane. Fourteen fields. Every field name printed correctly — title, source, type, information points, time sensitivity. Inside, all empty. The system that delivered the file had done its own job properly; what failed was the step before it, the one that was supposed to lift information out of the text. Before I binned the file, one thought landed: a cricket scorecard often looks exactly like this. The field names are there, the interiors are blank, and we fill those blanks ourselves.

August 2026. Mirpur. Bangladesh beat Australia 4-1 across five T20 internationals. My notebook holds five pages for five matches. Anyone reading only the runs and wickets columns today gets an easy answer: Australia collapsed against spin. The sentence is not false. It is also not the answer. Most of what actually settled that series is written in my notebook, not on the card.

Cricket's data apparatus is a sequence of decisions about what counts as information. Ball-tracking records release point, speed, roughly the seam position. The pitch map records where the ball landed. The wagon wheel records where it went. No instrument records how high the wicketkeeper's gloves were at the moment of release. None records how far the non-striker had walked out of the crease. None records the depth of the slip cordon, the angle between first and third slip, or which direction a fielder's first step went.

There is no point lengthening that list, because the real point sits outside it. The instrument that collects the information also decides what information is. In cricket that decision is almost always taken looking at the ball, not at what the people around it are doing before and after release. Fourteen years of watching matches has given me this repeatedly: the distance between the scorer in the pavilion and the keeper behind the stumps is not only geographic.

One distinction is worth keeping: 'no information' and 'nothing happened' are not the same statement. Cricket journalism blurs them constantly. A name missing from a squad announcement becomes 'he is injured' in print — that is an assumption, not a fact.

Back to Mirpur. Australia's batters were dismissed mainly by two kinds of delivery — the ball slightly shorter than their natural length, and the ball landing before their trigger movement had finished. Both are two faces of one event: timing. The spinner was putting the ball on the pitch earlier than the clock inside the batter's head. The scorecard calls that lbw or caught behind; the mechanism calls it tempo.

What Mustafizur Rahman's cutter was doing in that series was not variation in pace — it was the angle of his wrist at the release point. Two different balls leaving the same action, and a batter guessing for one of them. That angle never appears on a pitch map, because a pitch map tells the story after the ball lands; the wrist tells the story before.

With Shakib Al Hasan the thing is subtler. His arm speed stays almost identical over after over, but through the middle of a spell he was releasing slightly wider. That makes both the sweep and the cover drive difficult — one needs the ball coming into the body, the other needs it outside. On the scorecard the result is 'dot ball'. But a dot ball is not an event; a dot ball is an outcome. Dot balls rise when batters delay their decisions, and decisions get delayed when the bowler refuses to break his own routine.

Field settings read the same way. A defensive field is not a wall; it is a contract with time — the captain is saying, I will concede runs now and buy the over in exchange. The scorecard does not record that purchase, only the runs at the end of it.

My objection here is not to data culture but to over-trust in data. 14 July 2026, Lord's. Kane Williamson's New Zealand 241, Eoin Morgan's England 241, level again after the Super Over. Ben Stokes was named player of the match, and the World Cup was then settled by a boundary count: England 26, New Zealand 17. The information was clear. The number was not wrong. What the number would mean was decided by a rule, by a contract. The final was settled not by data but by a contractual arrangement made with data — one nobody had read that evening, because everyone assumed the cricket would settle it on the field.

This is where cricket has exactly one honest instrument, and it is DRS. What is 'umpire's call' really? It is the ball-tracking manufacturer's own admitted margin of error — the machine is uncertain whether the ball is hitting the stumps, so the decision goes back to the umpire in the middle. Cricket's only transparent data instrument is the one that publishes its own error bar. In fourteen years of reading analytical reports, that honesty has been rare.

Where honesty is absent, price gets set by another arithmetic. 23 December 2026, Kochi. Sam Curran sold for 18.5 crore rupees at the IPL auction, the most expensive buy in the auction's history at that point. The price came from a single-tournament sample: the 2026 T20 World Cup, where he was player of the tournament. The auction room had abundant information — pace, strike rate, economy. One cell stayed empty: how suited he was to that specific franchise's bowling plan, that specific ground's dimensions, that specific position. Nobody filled it, because nobody had the information to fill it. An auction price is not a measurement; it is a confession — an accounting of what is known and what is not.

Walk into a scouting room and the problem inverts: not too little information but too much. Deciding on a young player genuinely needs four things — who, in what role, in which format, and one indicator: average, strike rate, economy, or a twelve-month trend. Everything beyond those four is ornament. Our reports carry more of the ornament and frequently omit the first four. A comparison made without naming the format is not a comparison, it is a translation — and a bad one.

Another cell stays almost permanently empty: fatigue. The 2026 ODI World Cup packed 45 matches into 46 days; nowhere was it logged how much a fast bowler's third-spell pace had dropped from his first. On the fifth morning of a Test, a spinner's revolutions fall away, and the scorecard records that as 'no wickets'. Fatigue is not an excuse; fatigue is a measurement — one we do not take, only infer from results.

The largest version of that emptiness is not money but venue. Home and away is cricket's biggest structural performance gap. Australia's batters in Mirpur were playing in conditions they encounter more than four or five times a year approximately never. Bangladesh's batters on Australian bounce play close to zero matches a year in it. Both are structural constraints, both equally real. The difference is only in the naming: we call the first 'pressure of conditions' and the second 'technique deficit'. Same event, two names. The name is decided by whoever is collecting the information.

The talent-pathway question has to be read from both ends too. What a young Bangladeshi spinner learns — low bounce, slower pace, long spells — is the logic of his environment; what a young Australian quick learns is the logic of his. Labelling one as the other's deficiency is easy, but the information says something else: neither has collected data from the other's environment, because neither was given the opportunity. Calling information that was never collected a lack of talent is the oldest habit in our profession. When Bangladesh won the 2026 ICC Trophy in Kuala Lumpur and with it ODI status, that too was a rule-built door — not a playing result but a tournament structure deciding who entered which room.

Now the expected reading, and then the inversion. Expected: as cricket's information grows, analysis grows more accurate. Mine says otherwise — as information has grown, our confidence has grown and our accuracy has not. Because once the dashboard fills, people stop watching. And the cells still blank get filled inside our heads, with assumption, story, habit.

That is why the empty file reads to me as a warning rather than a failure. A system that can say 'I do not know' can be trusted the next time it says 'I know'. A system that fills every cell has to have every number verified separately.

My notebook has a line in it: The data did not explain the collapse; it timestamped it. Data does not explain a collapse, it stamps the time on it. The explanation comes from the people standing on the field.

The more I tracked the keeper's glove, the less the ball mattered. Where the ball ended up matters less than what was happening behind the stumps in the half-second before it left the hand.

So what do you watch in the next series? In Bangladesh's next home series, watch two things separately: how far the non-striker is walking out of the crease, and how far back the keeper is standing. Dot balls rising while the keeper's position does not change — that is the signal where we should stop before starting the story. The question, in the end, is not about numbers: what will you fill that empty cell with — information, or habit?