Forty Percent Dot Balls and One 22-Run Over: A Ledger Audit of a 187-Run Chase

The scorecard looks lovely. 187 for 5 in 19.2 overs, 9.67 runs per over, a win with two balls to spare. On a television graphic this is a flawless chase — the required-rate line flat at the start, then falling off a cliff, everything green.

When I laid all 116 legal deliveries of that innings into my ledger, the number that unsettled me most was not the runs. In the middle nine overs, from the seventh to the fifteenth, 22 of the 54 balls produced nothing. That is 40.7 percent. At the end of six overs the side was 52 for 1 and needed 135 from 84 balls, which is 9.64 an over. At the end of fifteen it was 126 for 3 and needed 61 from 30, which is 12.2 an over.

The real struggle of that chase sat inside those nine overs. The win arrived afterwards, close to half past nine at night, and I wrote the two events into separate accounts.

Data Verdict

Match: Rajshahi XI versus Khulna Tigers, domestic T20, regular round. Result: Rajshahi XI won by 5 wickets with 4 balls remaining. Legal balls: 116. Total runs: 187. Dot balls: 38, or 32.8 percent. Overs 7 to 15: 54 balls, 74 runs, 22 dots, 8.22 per over. Expected Runs model (xR): 174. Variance: plus 13. Repeatability score: 38 out of 100.

---

Context: A Routine Round and an Uncomfortable Coincidence

I built the Rajshahi xG ledger one match at a time, and the first lesson was patience. In 2026, hand-coding 3,780 shots across a 42-match league, the patience was football's, but the principle carried straight into cricket: a scoreline and a process are two different datasets. The scoreline says what happened. The process says what can happen again.

This fixture was a routine round of domestic T20 between Rajshahi XI and Khulna Tigers. Khulna won the toss and chose to field, thinking about evening dew. By the seventh over the ball was getting wet, the spinners were losing grip, and the line that had worked in the powerplay quietly stopped working through the middle.

Khulna batted first and made 186 for 7 in their twenty. Imtiaz Ahmed scored 64 from 43 balls, and the last five overs produced 52. The target of 187 was eleven runs above the recent average at this ground.

My coding method runs on three layers. First, ball-by-ball data: 116 rows for 116 legal deliveries, each row recording runs, whether it was a dot, whether it was a boundary, the bowler type and the batter's hand. Second, two separate coders code the same match and a reconciliation pass resolves conflicts — here four of the 116 rows disagreed, all of them over wides and leg byes. Third, the model layer, where I derive xR from four inputs: over phase, wickets in hand, the bowler-type versus batter-hand matchup, and the historical scoring rate for that exact situation in the domestic baseline.

Russia 2026 taught me that a data desk is a war room with better coffee. Tracking 1,842 shots across 64 matches that summer, the lesson was that the desk's job isn't to manufacture drama, it's to lower the noise. In domestic cricket that matters more, because the camera angles are fewer, the data resolution is coarser, and every row therefore carries more weight.

The Core: 116 Deliveries, Three Separate Economies

Reading this innings as one block is a mistake. Those deliveries ran on three different economies.

Powerplay (overs 1 to 6): 36 balls, 52 runs, 11 dots. Rajshahi attacked from the outset, and it was deliberate. Nineteen runs came in the first two overs, including four fours. Sadman Haque was bowled in the second over as the ball moved, leaving the score at 24 for 1. Rafsan Karim and Nabil Hasan then built 68 from 45 balls together, both striking above 130. Seven of the eleven powerplay dots came from the seamers, and six of those landed on the exact length that sits outside a left-hander's body line.

Middle overs (7 to 15): 54 balls, 74 runs, 22 dots. This is where the story turns. The scoring rate here was 8.22 an over, which is not bad, but 22 dot balls means 40.7 percent of deliveries produced nothing. In other words, 74 runs were built on 32 scoring balls — 2.31 runs per scoring ball, a rate that is not normal T20 shape. This is not a slope. It is a spike.

The cause of the spike was boundary dependency. Across the full innings, 106 runs came from boundaries, which is 56.7 percent: thirteen fours and nine sixes. When more than half your runs arrive from fours and sixes, every dot ball doubles in cost, because no bridge was ever built between the dot and the boundary. Through that middle nine-over block Rajshahi did not hit a single six — until Tarek Jubayer cleared long-on in the fourteenth.

Death overs (16 to 19.2): 26 balls, 61 runs, 5 dots. At fifteen overs it was 126 for 3, needing 61 from 30. The sixteenth over gave 9, the seventeenth gave 22, the eighteenth 11, the nineteenth 11, and the final two balls brought 8. The win came from squeezing the sixth bowler.

That seventeenth over was bowled by Sujon Mia, Khulna's sixth option, a left-arm spinner. It cost 22, including two sixes and two fours. That is 11.8 percent of the match's total runs arriving from six deliveries out of 116.

I look at that number separately because the football ledger taught me the instinct of insurance: what does a single goal actually weigh? The arithmetic here is simple. Had that over gone for 13, the death-over average at this ground, Rajshahi would have needed 39 from 39 balls after it — one run a ball. With three genuine sloggers in that line-up, a one-run-a-ball equation is winnable.

There is one more entry that scorecards never show. In the eighth over Rafsan Karim was dropped at slip off Imtiaz Ahmed, on 14. My catch-cost model puts the average price of that kind of middle-overs drop in domestic T20 at 26 runs. Rafsan finished unbeaten on 72, so the real cost was closer to 58 than 26. This isn't a fielding failure in isolation; it's a continuity failure, and it is the single largest event embedded inside the match.

The Contrarian Read: A Win That Proves Almost Nothing

The easy reading is that Rajshahi absorbed middle-overs pressure and finished the job, therefore the side is mature. I don't buy it, for three reasons.

First, sample size. 116 balls is one innings, but my ledger demands a minimum of six innings before I make a claim. A spike economy in one match may or may not return in the next five. Across my 2026 domestic ledger, average middle-over dot balls came to 34.1 percent. This match sat at 40.7, six points above, entirely possible in a single game, and that is exactly what makes it an unstable statistic.

Second, Khulna's death plan actually worked. Their two specialist death bowlers delivered overs eighteen and nineteen for 22 runs in total, which is 11 an over. On a ground where the target was 187, 11 an over at the death is good bowling. The leak was in the over number and the bowler type — a matchup decision at the seventeenth.

Third is the oldest part of my professional suspicion. When the stadiums emptied in 2026, the noise-free model finally let me hear the game. That period showed me how much crowd noise and media hype can dress a fragile process up as a winning story. The same applies here. The scoreboard will record a five-wicket win, while the process photograph says the batting side was once 68 off 63 balls and was staring at defeat well before the sixteenth over.

This is where my problem with visualisations lives. A wagon wheel and a heat map now feel to me like reading tea leaves — pretty, colourful, and largely silent. A wagon wheel shows where the ball went. It does not show the field restriction, which bowler was bowling to which hand, how many fielders were inside the circle, or what plan was pinning the batter down. Rajshahi's wagon wheel would suggest they scored all round the ground. The ball-by-ball column shows they really played two zones, long-on and cover, while seven fielding positions were effectively unemployed.

So which is today's most important entry? It is the Khulna captain's decision before that 22-run over. Four overs left, two set batters at the crease, dew settling in — bringing on a sixth bowler there is a tactical choice open to question. The alternative was splitting the two seamers held back for the eighteenth and nineteenth, or using the left-arm spinner against right-handed Nabil Hasan on a surface turning away rather than into the left-hander. The match rests on a strategic misstep, not on the superiority of a system.

Esports taught me that reaction time is just football with a shorter clock. In esports I learned that reaction time and decision time are not the same thing. Choosing the over was right; choosing the bowler for it was not.

The Final Entry: What I Will Watch Next Round

Next round, I will not look at the scoreboard while Rajshahi XI bat. I will watch two indicators.

One, powerplay dot-ball percentage. Here it was 11 dots in 36 balls, or 30.6 percent. If that climbs above 35 percent in the next match, I will assume the opposition has fixed its length inside the first two overs and that Rajshahi's attacking intent is turning into a trap.

Two, the Boundary Dependency Index through the middle overs. Here it was 56.7 percent. If BDI does not drop below 50 percent across the next three matches, I will treat the innings economy as spike-based rather than slope-based. Spike economies win you one match, the way a final can be won. They do not win you a series.

One prayer is written into my ledger: repeat, reconcile, and never trust a single match. Today's 187-run chase is the strongest argument for that rule, because the easiest story is the least true one.