Empty Spreadsheet, Full Broadcast: The Invisible Gap in Asia's Cricket Data

Last week I opened a file and sat with it for a long while. Twelve rows, the same answer beside each one: insufficient information. No title, no source, no single information point. A report on Asia-region cricket had gone into the analysis pipeline, and somewhere inside it turned out never to have arrived. The file looked immaculate—cells neatly arranged, labels in the right places, nothing inside. In data language this is a null payload. The spreadsheet did not interrupt the broadcast; it simply outlasted it. And that empty file pushed me toward a question I have been circling for months: Asia's cricket is broadcast more than almost any sport on earth, yet how much of it is actually verifiable?

Why the file came back empty is a story in itself. First the article never reached the server—an ingestion failure. Then the classification label arrived as cricket_asia, where the required field was simply cricket. A sub-label had taken the top-level chair, and every downstream template was about to route to the wrong address. The most dangerous part came last: the system raised no error at all. It produced a well-formed, confident-looking shell with not-applicable written in every cell. Blocked entry, wrong routing, silent resolution. I have watched matches for nine years, from the concourse to the scoreboard, and I have seen those same three failures play out in selection meetings, auction tables and workload calculations.

Asia's cricket broadcast economy is enormous. Asia Cup, IPL, BPL, PSL, ILT20—the calendar keeps the screen full. But broadcast hours and structured data depth are not the same thing. Ball speed, line, bounce, a batter's footwork, a fielder's starting position: if these are not stored at ball-by-ball resolution, then even after the match ends we do not know what actually happened. In my experience most Asian domestic competitions keep a flawless scorecard and almost nothing behind it. Which fielder stood where in which over, how much short ball a particular bowler bowled in the powerplay, the yorker ratio at the death—these answers usually cannot be found in any public database. Where answers are missing, discussion runs on memory and inference.

That is exactly where the second problem sits: label drift. When a sub-category occupies the slot meant for the parent category, the machine cannot decide where the file belongs. People do the same. When a selector looks at Taskin Ahmed's death-over economy, he files it under cricket—that is, bowling. But the number actually belongs under cricket_asia—it was produced by that pitch, that humidity, that matchup. The same economy tells a different story at Sher-e-Bangla than at Mount Maunganui.

The real crisis in Asian cricket is the gap between the volume of broadcast and the volume of verifiable data—and that gap shapes decisions before a ball is bowled.

Take 1 January 2026 at Mount Maunganui. Bangladesh beat New Zealand by eight wickets for their first Test win on New Zealand soil. What went least discussed afterwards was the rhythm of the bowling changes. Who came on in which over, in what length of spell, off how much rest—a full picture of that requires ball-by-ball field mapping. It does not exist. So we know the result and not the process. And without the process you cannot repeat the decision, only the story.

The same gap sits on the auction table. BPL and IPL prices are set largely by recent performance, star value and a handful of coarse numbers—average, strike rate, economy. But cricket's most expensive skills often hide inside coarse numbers. How many balls a finisher faces between overs 16 and 20, how his strike rate shifts, his scoring rate against left-arm spin—these narrow splits never enter the auction model, because the data is not there. Where data is absent, price is set by narrative. The biggest lesson of my working life is this: an auction price and a performance price are not the same number; the distance between them is the data deficit.

That deficit has an unexpected consequence. Where information is thin, rumour markets grow fat. Fantasy platforms generate hundreds of thousands of teams per match, and those teams are usually built on three scorecards and a podcast opinion. At the scouting level it gets sharper. If a scout wants to know whether a 19-year-old quick can hold his pace in the 55th over, he needs speed data from the last two sessions. Without it he falls back on the eye—and the eye is not unbiased. None of this is betting advice; the point is how missing information lets narrative drive a market.

Workload is the clearest case. In an Asian calendar one player can feature in three formats, two countries and four different pitches in a month. Yet workload is usually counted in matches, not deliveries. For a fast bowler the real load is spell length, rest between spells and recovery between matches—three variables combined. If those are not captured at domestic level, the national team's physio and its selector stand in the same dark room holding two different maps.

In 2026 I made my English-language commentary debut in Bangladesh women's ODI series against India. My biggest discovery there was where the hole is—women's domestic ball-by-ball data is so thin that comparative analysis is nearly impossible. Yet that same series showed clearly which spinners were losing control in which phase. With the data, we would have seen it before the match.

Here is the uncomfortable part. A null result is not a failure; it is an admission that the measurement failed—and the cricket industry almost never publishes admissions. An organisation that does analytics will rarely print the line: we do not have the answer to this question. What it prints is a clean, confident story. I know from my own work how strong that temptation is.

There is a deeper trap too, the most common mistake analysts make. A number is not proof. During the 2026 World Cup our log showed that 9 of England's 12 tournament goals came from set-piece situations. The easy reading is that set pieces were their superpower. The less easy reading is equally true: the team that wins more corners scores more set-piece goals, and the team that wins more corners is the team attacking more. Correlation is not causation. Cricket repeats the error endlessly. This bowler succeeds in the powerplay—is it his swing, the new-ball conditions, or the opposition's fragile top order? Separate none of those and the number cannot anchor a decision, only an argument.

I grew up in Bangladesh and now work in Manchester, and I feel the data-culture gap daily. In England even a second XI county scorecard is online, ball by ball. In Bangladesh domestic competitions it is often hard to find a full first XI scorecard. This is not a shortage of talent—the emergence of Taskin Ahmed or Litton Das proves talent exists. It is a shortage of infrastructure. Talent and data do not flow through the same pipe; one existing does not conjure the other. The heaviest cost lands at selection, when two contenders must be separated and the comparison rests on unequal depth.

In September 2026 in Manchester I hand-built a pressing spreadsheet—20 Premier League clubs, PPDA logged match by match. On 9 September at the Etihad, Manchester City beat Liverpool 5-0 after Sadio Mané's 37th-minute red card. My log showed City's PPDA at 12.4 before the dismissal and 6.8 after it. The scoreline was less a story about City's dominance than a story about the card. A 16-year-old's PPDA spreadsheet outperformed the broadcast, because the broadcast said City were magnificent and the spreadsheet said City had the advantage. Explaining pressing metrics to non-analysts taught me that people hold on to the decision behind a number, not the number. Cricket can make the same distinction—if it stores the data at ball-by-ball resolution.

So the empty file should not be thrown away. It is a signal. When the next Asia Cup report lands, the first question should be: how many information points sit behind this claim, and where did they come from? The organisation that can answer will keep its spreadsheet. The one that cannot will keep its broadcast—in memory, not in evidence.