We Have Been Looking for Khulna's Spin Story in the Wrong Place

Last season I was sitting in the stands at Khulna's Sheikh Abu Naser Stadium watching a National Cricket League match. A left-arm spinner took seven wickets, and all seven went into the scorecard as spin. After the match I pulled up the ball-by-ball log: of those seven, only two deliveries had actually turned appreciably. The other five came in almost straight — the batter played for the turn, the ball held its line, the stumps fell. The pitch had not changed, the ball had not changed; what changed was the model inside the batter's head. That night it became clear that much of what we celebrate as home spin dominance is not the pitch's quality — it is the batter's expectation going wrong.

For three seasons I have hand-coded National Cricket League data — 2026-20, 2026-22 and 2026-23. Ninety-six matches, roughly fifty-four thousand balls. Nearly sixty percent of those matches have no footage at all, and many scorecards are incomplete. Khulna, Rajshahi, Bogra and the Dhaka leagues — where nobody enters the scorecard, that is where my real work sits. The unrecorded matches matter more to me than the recorded ones, because Bangladeshi cricket's real signal hides in those dark places. This position is a method: building the dataset by hand is the reporting.

In 2026 I entered the industry at a Dhaka digital sports startup on eighteen thousand taka a month. I hand-coded all 14,200 events of 44 football matches, and found that Abahani Limited Dhaka had scored 23 goals from 15.8 xG across their first twelve games. My editor spiked it — tactics talk is for the boys. Abahani then scored nine goals in their next eight matches and dropped eleven points. Three weeks later the piece ran under a staff byline. The numbers were not lying; they were waiting for a better question.

Now to the actual data. For every dismissal I logged two things separately — how much the ball actually turned, and which ball the batter played for turn. I began assuming the cause of home spin success was the pitch. The numbers say otherwise. Of the wickets counted as spin wickets, nearly forty-six percent fell to deliveries that turned less than two degrees — that is, almost straight. In the fourth innings that share is highest, around twenty-two percent. The longer the match runs, the more the batter plays for turn, and the less the ball turns.

The session-by-session picture is clearer still. In the first morning session, while the pitch still has dampness, spinners do not actually get turn — yet wickets fall most heavily. The cause is not turn but the uncertainty of pace and bounce. Once the pitch genuinely starts turning after midday, batters set themselves, and the spin-wicket rate drops. The more the pitch turns, the fewer wickets the spinner takes — an inverse relationship that held across all three seasons. Home spin dominance is a sampling artifact; the real variable is which model the batter is using, and when. The spike got spiked, but the pattern stayed in the data.

This is where age and workload come in. A large share of Bangladesh's young spinners, I have watched being pushed into long spells of four-day domestic cricket at eighteen or nineteen. The body is not finished, but it is forced into senior rhythms. The result: in the last session of the third day, both their turn and their bounce drop — yet wickets still fall, because a tired batter and a tired bowler make mistakes together. The scorecard shows heroism here; the data shows decay. If I were a selector, I would look at the season-by-season decay curve, not the tournament average.

The venue comparison matters. At Dhaka's Mirpur, where the pitch generally turns more, the straight-ball wicket rate is lower — about one-third. In Khulna it is nearly half. The difference is not the pitch, it is habit. In Mirpur batters regularly play on turning tracks, so their expectation model matches the pitch. In Khulna they play a few matches a year, the model does not match, and the error returns as a wicket. In youth cricket the gap is wider still, because there the expectation model has not yet formed.

There is a parallel practice in the domestic structure. Big teams build around very young spinners, push them into senior bowling loads while half-finished, and never notice the decay. Small teams then spend a whole season producing an unfinished product for the big sides. On paper that is investment; in the data it is debt.

The natural conclusion is — change the pitch, add turn, the spinner wins. I am arguing that conflates correlation with causation. We assume the pitch turning makes the spinner take wickets because we see the two together. In my log, where the pitch turned least, the spin-wicket rate was highest. So the variable is not the pitch, it is expectation. And it matters to name what the dataset cannot see. Sixty percent of matches have no footage, so turn was measured from scorecard descriptions and my own notes — a range. I would rather give an honest range than a clean decimal, because a model should be tested, not defended. I do not chase edges; I build a monastery around them.

A word on heatmaps. Look at a spinner's wicket heatmap and you would think he always lands the ball in one corner, and that this is the cause of his success. In reality the heatmap separates him from the tactical system — who stands at slip, which end has wind, which session has dew — and all of that disappears. The heatmap is reading tea leaves again; it hides the player's actual role. The event my dataset tracks most is the event that did not happen — the session lost to rain, the bowler who was never picked, the innings that ended before it could be scored. In Khulna I learned that silence is also a dataset.

Next season my eye will be on one metric — the straight-ball wicket rate, where the batter played for turn but the ball did not turn. If that number rises, you are not reading the pitch; you are reading the batter's head. The scorecard will tell you a story; the question is which story you are willing to buy.