The Auditable Ledger of Asian Cricket: How Hand-Logged Ball-by-Ball Data Prices a Bowler's Overs and a Calendar's Load

Hook

The dew at Mirpur's Sher-e-Bangla Stadium that November night was so heavy that the spinner could not grip the ball, yet the scoreboard kept describing the pitch as a batting paradise. I was logging the 17th over of the second innings ball by ball, by hand. The official feed recorded five runs and four dot balls. My notebook recorded something else entirely: four full tosses, one slog-sweep, three inside edges, and two dots that were actually missed stumpings. Across the last five overs the pitch's carry fell from 1.9 to 1.3 — and the market still priced that surface as a 175-plus deck.

I logged every shot by hand before the market learned to price it. In Asian cricket today the most valuable asset is not a bowling action or a bat swing. It is an auditable ledger. That same night in Mirpur, a trading desk in Dubai was setting an over-under line on a number I had built four hours earlier. They got the number; I made it.

Context

The men's T20 World Cup was staged in India and Sri Lanka across February and March 2026. Once it ended, Asia's calendar did not empty — it filled with a new kind of pressure. The International League T20 in Dubai across January and February, the IPL from March to May, the Lanka Premier League, the Nepal Premier League, Karnataka's Maharaja T20, the Abu Dhabi T10, and the Bangladesh Premier League in December and January. An Asian fast bowler is asked for overs in at least four separate franchise windows a year, and owns exactly one body.

The worst-priced part of that calendar is not international cricket. Ball-by-ball data from India, Pakistan, Sri Lanka and Bangladesh is now in every major vendor's system, with tracking and multiple camera angles. The gap sits in Asian domestic and associate cricket. Nepal, Oman, the UAE, Malaysia, Hong Kong — in those markets the feed often arrives late, or arrives incomplete. A bowler who has taken 40 new-ball overs at 7.2 an over gets priced off a two-minute highlights package.

The Asia Cup stage, ACC age-group and Emerging tournaments, and the build-up to the 2027 ODI World Cup: across those three layers, selectors in Bangladesh, Sri Lanka and Afghanistan are making decisions without a measurable ball-by-ball foundation. What I have tried to do since 2026 is not to support a team, but to fill that vacuum. My notebook is personal, small, and knows its own limits. It nonetheless holds information that can change a decision before the market ever sees it.

Core

In 2026 I hand-logged 1,140 shots from 96 domestic T20 matches, one grainy stream at a time. I was the only data seat on a desk in Dhaka. A senior columnist called it a girl counting shots. Two head coaches later asked for the spreadsheet. The lesson: value lives in the ledger, not the headline. I stopped writing adjectives. Every match now gets five layers — ball-by-ball outcome, shot type, contact quality, field-setting coordinates, and bowling load. Those five layers produce an auditable ledger where every claim carries a ball number.

From that ledger comes my expected-runs model, a cricket-scale cousin of football's xG. Line, length, field, phase and shot type combine into the expected output of a controlled delivery. In my logged sample, a pull against a short ball averages 1.42 runs per shot, a slog-sweep against fuller length 1.68, an inside-out drive against a wide yorker 0.60. Big feeds do not record that layer; they record '4 runs'. Yet eleven fours off inside edges and eleven middled fours are not the same event, even when the scoreboard says they are.

Dew is not a constant. It is a variable, and every variable carries an expiry date. I keep separate dew coefficients for Mirpur, Chattogram, Colombo's R. Premadasa, Dubai and Sharjah. Of 214 night T20s I logged at Mirpur between 2026 and 2026, the side batting second won 58.4 percent; in day matches, 47.1 percent. In the last five overs my dew coefficient adds 0.42 runs per over, and reduces measurable spin turn by roughly 18 percent.

I date that coefficient. This version, set on 15 March 2026, expires on 31 October 2026 unless the surface is re-measured after re-turfing. After 2026 I do not forget this: when the stadiums emptied, not only did home advantage shift — pitch behaviour shifted too. Franchise cricket carries no crowd pressure, so carry and dew relate differently than in internationals. A model still using a 2026 coefficient to read Mirpur in 2026 is not wrong. It is expired.

Workload must be counted in overs, not matches — and recovery in days. For Bangladesh's seam group I run a live ledger on three pillars: competitive overs, days between spells, and travel load. My pre-registered alarm band is 44 overs inside a 21-day window. That is my threshold, not a universal rule, and I publish it so readers can see the limit I set for myself.

The travel pillar is the most neglected. Dhaka to Dubai, Dubai to Colombo, Colombo back to Dhaka: more than 21 cabin hours on three legs, with average sleep debt in my log at 6.2 hours per night. A fast bowler's delivery speed in the second spell frequently drops 2 to 3 km/h from the first. The market never pays for those 2 km/h, because the feed records average speed, not spell-by-spell speed.

A price band converts speculation into an instrument, and every instrument has a ceiling. I treat cricketers as stocks and contracts as assets. Before a franchise auction I build a band for a seamer: powerplay dot-ball rate on one edge, death-over economy on the other, injury density in between. For a bowler logging 118 new-ball overs, 62 percent powerplay dots and 9.4 death economy, my fair band was 2.3 to 2.9 times base price. When the market clears the top of that band, I write — and I say why. A transfer rumour only carries value when a logged spell sits behind it; otherwise it is an unhedged position before the medical.

Asia's largest mispricing sits in associate cricket. Across recent ACC events I hand-logged more than 300 overs where feed coverage was incomplete. Omani left-arm seamers were bowling 45 percent of deliveries inside six metres on slow, low pitches, yet the market priced them off tournament wicket totals. The skill was hidden in the feed, not absent from the bowler. The cheapest asset in Asian cricket is the data nobody has taken responsibility for collecting.

Contrarian

Dhaka's convention holds that workload destroys fast bowlers and rest is the cure. My hand-logged injury ledger reads differently. Across the bowler-seasons I tracked from 2026 to 2026, injury incidence correlated most weakly with total season overs. It correlated most strongly with three things: returning to bowl directly after a break longer than nine days, a load increase above 35 percent in a single block, and inadequate recovery between two spells in one match. The problem is often not how much load, but the shape of it.

Correlation is not causation. A bowler who breaks down has usually bowled a lot; but plenty bowl just as much and stay fit. In my sample the correlation between overs and injury sits below 0.2 — a weak signal. Base rates matter: even at zero workload, a fixed percentage of fast bowlers miss matches every year.

So I publish a counter-consensus read only when my logged edge clears 0.30 runs per over, or when modelled win probability diverges by 6 percentage points. Threshold first, article second. Belgium taught me the discipline: hold a position only while evidence and price both justify it, then close the book. I do not chase edges. I audit the assumptions that create them.

Takeaway

In the next window I am watching two things. First, the dew coefficient in the 19th over: if second-innings turn drops more than 18 percent, the pace-upset line is being mispriced. Second, the ball-by-ball feeds in Nepal and Oman — where no ledger exists, the opportunity is largest. Scorecards will not explain Asian cricket. A ledger will, provided every line carries the date it expires.