The Unwritten Scorecard of Khulna: How Asian Domestic Cricket Erases Its Own Data

It is nine in the morning on 13 January 2026, in the western gallery of Khulna's Sheikh Abu Naser Stadium. A left-arm spinner is running through his seventh consecutive over from the pavilion end — same footfall, same release angle, a line almost millimetric in its consistency. I logged all 42 deliveries of that spell by hand. The official scorecard is recording only overs and runs.

By the end of the day I had 204 entries in my notebook. The official scorecard had 83 overs, 17 wickets, two half-centuries and a bowling average rounded to a clean decimal. The gap is written nowhere. That left-armer's first-class average sits below 19 across six seasons, yet no file on him is ever opened, because the matches he bowls in have no ball-by-ball log. Local selection committees decide from drawn statistics, and where even statistics are absent, they decide by the smell of presence.

The numbers were not lying; they were waiting for a better question. The question is not wickets. It is survival.

Context: the measurement asymmetry of Asian domestic cricket

Three kinds of fixtures exist on Asia's cricket calendar. The first kind carries Hawkeye, ball-tracking, camera coverage of every stadium sector, a follow-through data feed — Mirpur, Dubai, Colombo, Chennai internationals. The second kind carries partial coverage: some Emerging Teams series, televised franchise fixtures, a slice of the Dhaka Premier League. The third kind carries one handwritten scorebook, which is filed into a cupboard at season's end and never opened again.

Large parts of Bangladesh's National Cricket League, many Ranji Trophy venues, outground fixtures of Pakistan's Quaid-e-Azam Trophy, the lower rungs of Sri Lanka's club tournaments — all of these are the third kind. Strike rate, dot-ball percentage, over-by-over pressure indices, field-setting maps: none of it exists. So the batter who made 700 runs at 40 and the batter who made 650 at 32 cannot be compared, because the comparison was never performed.

That asymmetry is not harmless. The entire selection architecture rests on one simple assumption: what can be measured is true. In Asia's reality, what gets measured is not a sample of cricket skill but a sample of opportunity. A player who reaches a major venue, a broadcast match, a stronger side, begins to be measured. A player who reaches the field in a lower-table side becomes invisible the moment his career starts.

When I joined a Dhaka digital sports startup in 2026 as its first data hire on 18,000 taka a month, I hand-coded all 44 matches of the Bangladesh Premier League football season — 14,200 events. That is where I learned how much information evaporates when a camera is removed. In cricket the loss is several times larger.

Core analysis: four places where data makes decisions, and one where decisions make data

One. The sample is the selection

Imagine three left-arm spinners in one domestic season. The first plays at home, televised, with a ball-by-ball log being generated. The second travels, with no coverage. The third misses 40 per cent of the season injured — his data is missing too, but the reason for the absence is not logged either.

At season's end a list is produced containing all three, but the weights are not equal. The first has a fixed number beside his name. The second has a selector's memory. The third has the shadow of a phrase: there are questions about his fitness. What is happening is not measurement. It is a ceremony that resembles measurement.

In the sample I coded by hand — roughly three thousand deliveries, six venues, three seasons — left-arm spinners bowled 38 to 40 per cent of overs but took more than 50 per cent of the wickets. The fraction is not startling. What is startling is that I had to count balls by hand to produce it, because nobody records it. In Asian domestic cricket, analysis does not always mean discovery; often it means reconstruction.

Two. Workload and a peak curve borrowed from the wrong place

Discussion of fast bowlers' peak curves in Asia is almost always conducted with a borrowed model, built from workload data in England, Australia and South Africa — where the season pauses, where long breaks exist, where indoor sessions occupy the winter. Bangladesh's or Sri Lanka's domestic structure has no such rhythm.

The result is simple arithmetic: when the same workload index is applied to different soil, different humidity, different bowling loads, the index does not protect the bowler. It indicts him. Every model is a prayer until the data says otherwise — and a model placed in the wrong context never answers at all.

This is where something else hides, and it is what I most want to write about. In Asia's domestic structure, the load falls fastest on young, early-maturing quicks. Before he has finished school he is called into the national nets, then given consecutive overs in the domestic league, and then an injury management plan is written for him from a template whose age band does not match his body's age. Nobody built this path deliberately. Several people built it simultaneously, because if you find a young quick who can bowl at a major venue you will not rest him, and if you bowl him you generate data, and if data exists it becomes a selection basis, and the firmer that basis becomes the harder it is to question your system.

Three. Home-spin dominance: fact, or sampling effect?

Nobody disputes that spinners dominate Asia's domestic first-class fixtures. Mirpur, Rajshahi, Khulna and Bogra produce slow, turning surfaces; spinners deliver 55 to 65 per cent of overs, and results are frequently settled by the final session. The figure is so familiar that it is treated as geographical truth.

Reverse the question — is this dominance purely a product of the pitch, or of the selection sample? — and the picture starts to move. Left-arm spinners play more at major venues because preparation against left-arm spin is thinner; the sides that get the most coverage are the ones with strong pace attacks and flat-session batting; so spin-heavy innings disproportionately occur away from the pitches of teams stationed outside the venue.

Bowling at an average of 19 to 19.5, however much turn you possess, your record often rests on a single column: average. Strike rate, phase-by-phase economy, your role in a defensive fourth-innings effort — none of it is there. The gap is filled by narrative. Who is a big-match bowler, who cannot bowl on other pitches — these legends are manufactured outside the data but spoken in the data's language.

Four. A heatmap is a map with no streets

I hold a clear and unpopular position on heatmaps. Twenty square metres of grid squares do not tell anyone whether a left-arm spinner was flighting the ball, building pressure, or simply bowling a safe over. A heatmap measures where the ball went. It does not measure why the ball went there, who set the field, or how many runs needed saving in that session.

Where Bangladesh's domestic league has no heatmap, the advantage runs the other way. Bowlers can be seen in a system, across a timeline, inside a partnership. Sitting three days at Khulna, I saw that left-armer release only one ball a day to the outside; the other forty turned in or skidded on. Without data, that structure is visible but cannot be recorded.

Heatmaps help decisions — true, but only for decisions whose question is already correctly framed. Unfortunately, many Asian selectors do not frame it themselves. They borrow the map. In Khulna, I learned that silence is also a dataset — but silence must be read, otherwise it is read like a drum.

Contrarian angle: where correlation is not cause

The reflex response is this argument: Asia's domestic cricket is poor, therefore no players emerge from it. I do not want to arrive there, because my own data argues against me. The absence of a record does not mean there was nothing worth recording; it means that during the period you could not record, you can never prove who was genuinely good. That is a negative result, not a damaging verdict. The difference matters.

One more thing should be stated plainly. A portion of what looks like curated success in Asia's domestic structure is simply rain. Matching scorecards across several seasons, I noticed many matches never completed four innings, entire first-day sessions were lost, and several spin-friendly fixtures ended inside 40 overs of the second innings. Where a session is cut away, a seamer's average inflates artificially, because the chance to bowl on a hard pitch never arrived. Nobody records this, because rain is not bad news. It is nobody's fault.

This is where my most familiar principle returns: the spike got spiked, but the pattern stayed in the data. When that left-armer's average reads 22 across his first three seasons and 18 in his fourth, the improvement is not really a bowling improvement; it is the compound product of fitness, one correct partnership, and a changed pitch set. An analyst who reads only the average reads a correlation as a cause. In Asian cricket that is the most expensive error available, because it happens invisibly, without a single press conference photograph.

So to those who say data has transformed Asia's domestic cricket, the answer is that data has not yet arrived here. What has arrived is something more damaging than a data shortage: the habit of passing off that shortage as statistics.

What to watch next season

Three things deserve separate attention when you open the next domestic scorecard. First, when you see the season's leading average, look at that bowler's venue list; if most of it is home, the average matters less to you than its structure. Second, keep a monthly over-count for bowlers, not just wickets; before an injury, the pile of overs is always visible first. Third, make a list of the matches that have no log at all — because your next best player is probably bowling there, or bowled there, and nobody knows it. Khulna's silence stays in Khulna; the question is whether you will come back to open the notebook.