The Dot-Ball Ledger: Bangladesh's T20 Crisis Isn't at the Death, It's in Overs 7–15

22 June 2026. North Sound, Antigua. India 196/5, Bangladesh chasing 197. By the close of play the scorecard said this: Bangladesh 146/8, India winning by 50 runs. What the scorecard did not say is written in my ball-by-ball ledger. Overs 7 to 15: thirty-one dot balls across eight overs, one six, nineteen runs.

The stadium was full. The ledger was almost empty.

In Bangladesh's T20 discourse we almost always talk about the last five overs — who hit how hard, who got out, who lacked intent. But the overs that manufacture that death-over risk are never counted anywhere. A scorecard is a lossy compression. A hundred and twenty balls go in; six numbers come out.

This piece is the arithmetic of what those six numbers threw away.

Method first, claim after

My ledger holds 68 Bangladesh T20 innings, ball by ball, from January 2026 to December 2026. The sample is small by Full Member standards and the confidence intervals are wide. I proceed with that admitted. Let the ledger breathe before the narrative does.

Definitions, fixed before I looked at outcomes:

• Dot-ball percentage = legal balls with no run scored ÷ total legal balls.
• Middle overs = 7 to 15.
• Boundary rate = fours plus sixes per 100 balls.
• Acceleration tax = the per-over wicket-loss rate across overs 16 to 20 when the required rate at the start of the 16th is above 9.5.

I am using two custom roles — middle-over rotator and new-ball dot creator. Both were defined before outcomes. I will not build more. If I keep inventing finer roles and the arbitrage never closes, the role was my artefact, not the market.

The number that should make you uncomfortable

Across those 68 innings, Bangladesh's middle-over dot-ball rate is 41.2 per cent. Over the same window the top eight T20 sides averaged 32.4 per cent. A gap of 8.8 points. In T20, eight points is an enormous amount.

More uncomfortable still: Bangladesh's powerplay dot rate is 44.1 per cent and its middle-over dot rate is 41.2 per cent. The distance between the two phases is 2.9 points. For leading sides that distance runs between 12 and 14 points.

Think about why. In the powerplay only two fielders are outside the circle — more room for boundaries, but also more dots, because the ball is new, it moves, and the batter is rushed. In the middle overs five fielders are out. Boundary options shrink, so an international batting line-up does one specific thing: it rotates strike. One-two, one-two. The dot rate falls.

Bangladesh is not doing that. It carries the powerplay rhythm into the middle overs. That is the real fracture. Not the death overs — overs 7 to 15.

Litton Das, Najmul Hossain Shanto, Towhid Hridoy and Jaker Ali all sit between 39 and 46 per cent middle-over dot rates. Each has a career strike rate between 120 and 135, which does not sound bad at first read. But a strike rate is an average. It does not tell you the sequence in which the balls were faced. A batter who plays 14 dots in 30 balls and scores off the other 16 finishes 30(25); a batter who plays four dots and scores off the other 16 also finishes 30(25). Identical on the card. Not identical in the match.

The causality usually runs the other way

The familiar explanation: Bangladesh has no power hitters, so it starts behind.

My ledger shows a different variable. At the start of the 16th over Bangladesh's average score is 92/3. The average required rate at that moment is 9.8. When the required rate sat below 9.5, Bangladesh lost 0.31 wickets per over across the next five. When it sat above 11, the rate was 0.68. More than double.

The wickets are not falling from a shortage of shots. They are falling from an obligation to play them. A batter who walks out in the 16th over hunting boundaries loses freedom in shot selection. The dot-ball ledger creates the shortfall in the earlier overs; the death overs merely settle the bill.

Pitch conditions fold into this. On a 165 to 175 surface both sides absorb the same pressure and the difference is hidden. On a slow, two-paced Mirpur or Chattogram deck, where 148 to 155 wins, the shortfall converts directly into the result. What the public market calls a lack of intent is a deficit accumulated ball by ball.

What the market is mispricing

Now the mispricing itself. Domestic T20 auctions, franchise scouting and broadcast narrative all price a batter in two columns: strike rate and sixes. They price a bowler in two: wickets and economy.

The cricketer who saves three dots every ten balls but never hits a six has no column at all.

Take two middle-order batters. One makes 30(24) with four dots. The other makes 26(16) with seven dots and three sixes. At the auction table the second is worth far more, because his tape looks better. But in the kind of match Bangladesh actually plays — a 150 to 155 target, seven to eight an over — the first profile wins more games. His dots are fewer, so his team's required rate at the 16th over stays below 9.5, and his team loses fewer wickets.

That gap is the arbitrage. The market is paying for punishment and not paying for strike rotation.

Across the border the mispricing is plain. In the Indian Premier League the demand for the bat-through middle-order player has fallen for several seasons, because the top order scores so quickly that the middle order's only job is finishing. In the Bangladesh Premier League the picture inverts. Openers leave at 40 off 35, and the middle order has to drag the match through 14 overs. Same role, two markets, two prices. Which price is correct is decided not by the player's talent but by the team's scoring architecture.

The innings that decided Fortune Barishal's title in February 2026 came out of the strike-rotation ledger, not the six-hitting ledger. My log says the same thing.

The bowlers are being read upside down too

The same error applies to spinners.

A spinner's card reads 4-0-23-1. The eye catches one wicket. The ledger shows eight dot balls, five of them in a left-hander matchup. It is because those 23 runs were suppressed that the opposition's required rate crossed 11 at the 16th over — and lost a wicket an over later.

What does the camera show? The wicket.

The strongest signal in my ledger: spinners bowling overs 7 to 15 have a combined economy of 7.1, and that economy correlates negatively with Bangladesh's death-over wicket-loss rate at a Pearson coefficient of minus 0.42. It is precisely because these bowlers were not punished that wickets fell later — and the credit column carries a seamer's name.

Rishad Hossain, a leg-spinner, is priced on his wicket count. A controlling attack like Mehidy Hasan Miraz's is priced lower. Both are needed, and across a full team contribution the distance between them is far smaller than the distance between their columns.

Where my argument is weak

This is the point to stop and say it.

Dot-ball percentage is an outcome, not purely a cause. A side chasing big totals has fewer dots; a side fighting has more. Causation can run backwards — weak batting raises dots, and more dots make batting look weaker still. What I am claiming is not correlation but a flow: dot balls, then over-by-over pressure, then narrowed shot selection, then wickets. To demonstrate that flow you have to analyse by over sequence, not by match-end strike rate.

The second problem is sample. Sixty-eight innings is small. The 41.2 per cent middle-over dot rate carries a confidence interval of 37.6 to 44.8 per cent — the true value could sit below 35. One good home series will change the story. I keep a standing base rate for my own overrides: over the last four years I have gone against consensus seven times, been right four times and wrong three. Fifty-seven per cent. Not magic. A ledger.

The third problem is the most important: bowling quality. When the opposition fields a Kuldeep Yadav or a Rashid Khan, dots rise. I have controlled for that only by restricting the sample to Full Member fixtures, and that is not a perfect fix. So I am not saying the dot rate is the only cause. I am saying it is the least discussed variable.

A pre-registered claim for the next series

Here is the forecast, timestamped.

If, at the next Asian competition, Bangladesh plays 45 per cent or more dots in the powerplay and holds 38 per cent or more in the middle overs, my model puts the probability of successfully chasing any target above 170 at under 25 per cent, across five thousand simulations.

I will log the inverse as well: if Bangladesh can bring middle-over dots below 34 per cent — without adding a big hitter, purely by reordering the batting line-up — I expect its T20 win rate to move above 50 per cent from the 38 per cent of the previous three years.

Looking at two prices for the same player in two Asian markets, only one conclusion survives: the market buys sixes; the game is lost in overs 7 to 15.

Bangladesh's next step will not begin in the six-hitting ledger. It will begin in the ledger of uncounted, uncounted dot balls.

I count the silence between the balls. Noise misleads sometimes. Silence does not.