Khulna's Ledger, Dhaka's Selection: The Session Rows Domestic Cricket Never Publishes

Sheikh Abu Naser Stadium, November 2026. A Test against West Indies, Bangladesh's first innings coming apart. The ninth wicket has fallen and out walks Abul Hasan, a debutant, batting at number ten. He makes 113, sharing a record ninth-wicket stand with Mahmudullah. On the scorecard it is a bright, clean number, and the broadcast reports return to it again and again. In my ledger the first row of that innings reads differently. He did not lay bat on ball for the first ten deliveries, true, but his feet never settled and his body kept drifting off line. I opened the Khulna ledger, and the first column taught me patience. The 113 matters, but which session it came in, under what pressure, at the end of what selection process — the domestic scorecard does not write that down. I started writing in 2026, covering the Wills Cup in Dhaka. The habit formed then: alongside the scorecard, a separate ledger in which every match splits into four kinds of rows. The first row is session-based: overs 1-30, 31-60, 61-90 and the final-day third session, each with its own run rate and wicket count. The second is stroke-based: how many balls before a batter's first false shot, what share of deliveries he left or blocked. The third is pitch-based: morning moisture, grass cover, second-day bounce, third-day turn. The fourth is load-based: overs bowled, days between matches. Beyond those four rows sits my hardest discipline — regime splitting. Four-day domestic cricket, franchise T20 and international cricket are three separate countries with three separate currencies. One country's banknote does not spend in another. Building a pressing model for France's 2026 World Cup run taught me this plainly: the PPDA map was not a picture, it was a confession of where France pressed and where they deliberately stopped pressing. Cricket demands the same separation. Mix the format ledgers and the misreading enters before the analysis begins. The problem is that none of these rows get published. Newspapers carry averages, strike rates and best bowling figures. The columns that actually decide matches never reach the selection meeting. The Khulna wicket lives as its own geographic entity inside Bangladesh's domestic structure. Winter mornings hold moisture and the grass cover stays on, so the first hour belongs to seam. After lunch on day one the surface changes colour, the ball starts keeping low late on day two, and by days three and four spinners can hold a side to between 2.4 and 2.9 runs an over. I have logged that figure session by session myself; it is not an estimate. Which means a Khulna first-innings score can never be read alone. One 280 and another 280 are worlds apart, because one arrived after three wickets fell in the morning session and the other arrived on a flat third-day surface. The scorecard holds no such distinction. The session ledger does. The real currency of first-class batting is not runs but balls. For the top six I watch two columns hardest: balls per dismissal, and balls faced before the first false shot. A batter averaging 45 at 62 balls per dismissal and one averaging 40 at 40 balls per dismissal sit close on average and are entirely different assets in a Test. The first can buy you a session at number three. The second can score quickly at six and still fail on a hard day. Bangladesh's top order needs survival, not speed — and the column that measures survival is the one domestic cricket never publishes. Selection committees generally read the average column and the strike-rate column. The error is not in the column, it is in the choice of column. Across 52 years of watching this game, my read is that Bangladesh's selection cycle has lost the most ground by discarding players whose averages were modest but whose balls-per-dismissal numbers were high. A batter who can face 70 balls on a domestic turner is an investment on international surfaces — it pays over three years, not three matches. When the ledger shows only the average row, that investment is booked as an expense. At Mirpur, on a flat BPL deck, the same batter's strike rate climbs past 140. At Khulna, in the NCL, he stalls at 45. Two numbers, two realities, yet plenty of analysis places them side by side as if one explains the other. It does not. A flat T20 deck tests fast hands; a four-day turner tests patience, footwork and bounce adjustment. One batter can score well in one exam and poorly in the other. That is not failure; that is regime difference. My fourth row belongs to the bowlers. In a four-day NCL match a seamer can bowl 34 to 38 overs, and the next match may start three days later. Then comes the franchise league, then the tour, then the commercial friendly. The language of load management appears precisely when the calendar holds a commercial commitment; the 38 overs of a domestic four-day game never earns that language. In September 2026 Bangladesh won their first T20I series against New Zealand, 3-2 — the first three matches to Bangladesh, the last two to New Zealand. In my own death-over economy charts the difference was clear: the spinners varied their bounce early in the series and flattened out late, and the line became predictable. Load and tactics are not two separate questions here. Which brings me to my strongest warning: the relationship between runs and runs is not a straight line. In 2026 I kept a ledger for a Dhaka website on domestic football. Abahani Limited Dhaka scored 28 goals in 14 matches from 21.4 xG. My regression warning went to print; three of their next five matches ended in draws. Plenty of people said the luck had turned. It had not — shot selection had changed. The table gave a correct warning and the explanation attached to it was wrong, and a wrong explanation does more damage than a correct table. Cricket carries the same trap. The link between domestic averages and international averages is weak, because bowling pace, fielding density and the scouting database all change in between. So I attach a confidence tier to every judgement: Tier A means three seasons of consistent rows, Tier B one season, Tier C six innings. I never make a recommendation on Tier C, because that is really a rule of thumb leaning on six small samples. One more thing belongs here. NCL scorecards carry no fielding data, no ball-by-ball, no catching positions. That is not tactical silence; it is structural absence. An empty ground does not lower the standard of a match; an empty column blinds a decision. I keep the empty-stadium account in one ledger and the missing-column account in another. In the next round I will watch two columns: balls per dismissal for Khulna's top six, and third-session economy for the frontline spinners. If the same name rises twice in three rounds, I will call it a signal — not before. A clean row of data will outlast a thousand hot takes, and on selection questions it is the only account that can actually be settled.