Middle-Overs Arithmetic: Three Gaps in Bangladesh's T20 Batting Ledger

Middle-Overs Arithmetic: Three Gaps in Bangladesh's T20 Batting Ledger

Hook

There is one row in my ledger I keep returning to. It is not a six. It is not a spectacular catch. It is a dot ball.

One evening at home, the fourth ball of the eleventh over. The board reads 71 for 2. A new batter is at the crease, a leg-spinner is bowling, the field is pushed deep — one at long-on, one at cover, one at deep midwicket. The batter defends. A zero gets entered into the book.

A single zero says nothing. But when I code Bangladesh's T20 innings ball by ball for eighteen straight months, those zeroes build a wall. In the powerplay (overs 1-6) our run rate is 8.2. In the death overs (16-20) it is 9.1. Between overs 7 and 15 it falls to 6.8.

Put the three numbers side by side and the story stops hiding. Bangladesh start well, finish well, and lose roughly ten overs in between — which is exactly half of a T20 innings.

I built the Rajshahi ledger one match at a time, and the first lesson was patience. Without patience, that figure of 6.8 would never have caught my eye; what would have caught the eye were the two or three innings where somebody made 70 off 40, and that became the next day's headline.

Context: What the Ledger's Rules Are

The rules in my book are simple and they stay the same. Every ball is a separate row: innings number, over and ball number, batter's name, bowler's name and arm, bowling type, line and length, shot type, shot intent (boundary, rotation, defence), outcome, and field placement. Then I attach three values: run value, wicket-probability cost, and the estimated difficulty of the ball.

I do this coding entirely by hand, because the real value of analysis is created at the moment of collection, not later on a dashboard. When I coded 42 matches and 3,780 shots of the Rajshahi Premier League in 2026, I learned that if the whole chain sits in one pair of hands, the shape of your error becomes recognisable. Rakib Hossain's 14 goals from 8.7 xG was part of that lesson; the second row of it was the assumption that messy pitching would settle over time. In cricket that assumption is far harder, because ball speed, seam movement and camera frame rate all shift together.

In a cricket ledger I use three segments, and each means something different. The powerplay measures the advantage of field restrictions. The death overs measure risk-taking decisions. The middle ten overs — seven to fifteen — measure the patience and intelligence that never gets written in big letters on a scoreboard.

I keep international middle-over strike rotation rate (the percentage of balls turned into singles or twos) in a separate column. That single indicator tells you more than a batter's boundary count, because a side that cannot rotate strike above forty per cent in the middle overs burns every ounce of its death-overs talent on paper, since the innings structure breaks before the last five overs even arrive.

There is an extra layer I have recorded separately since 2026. When the stadiums emptied during the pandemic, the noise-free model finally let me hear the game — which roar and which pressure belong to the cricket itself, and which belong to the stands. That empty-stadium data gets its own colour in the ledger. At Mirpur in the evening, the over in which dew settles and the over in which a commentator's voice drowns out the sound from beyond the boundary are two different effects. Ignore the split and the middle-overs picture stays incomplete.

For Bangladesh you must add one more thing: the nature of home surfaces. Mirpur's pitch is good, slow and two-paced. With the new ball the seam stays low, and when humidity is high the ball takes longer to reach the bat. In those conditions, waiting out the middle overs and then attacking the shorter boundary is easy in theory; but if a spinner holds his line through those ten overs, finding a stroke becomes genuinely hard.

That is where the second piece of context arrives: the source of the data. Bangladesh's domestic T20 structure began with the BPL in 2026, and that league is the main pool of local data. But our ball-by-ball store is thinner than in the bigger international leagues — some matches are missing speed-gun readings, some grounds have only partial camera coverage. Where the input data is itself incomplete, reaching a decision from a heat map means reading tea leaves with a local accent.

Core Analysis: Three Gaps in the Book

The first gap is not in strike rotation but in the pattern of the pause. In my ledger the dot-ball rate in the middle overs is around 42 per cent, against 36 per cent at the death. The number sounds small, but removing six dot balls from sixty balls across ten overs drags the middle-over run rate from 6.8 down towards 5.9. That is a shortfall of roughly six to eight runs, and no four-over death-overs effort reliably makes it back up.

The distribution of those dots sharpens the picture. In my coding, four of every six middle-over dots come in a spinner's first two overs, and a large share of them come from a left-arm orthodox bowler angling the ball across outside off, using the extra bounce of the two-paced surface to close off the gaps.

Cover, then long-off, then long-on, then deep midwicket — with twelve of the thirty fielders standing back, the single becomes a long walk.

The second gap is about impact, and about the time given to it. In my numbers, an impact batter's per-ball boundary probability in the middle overs is about 1.3 times an opener's, and yet that impact batter faces an average of only 22 balls a match. Some of those arrive after the seventeenth over, by which point the opportunity has already narrowed. In other words, our set batters are not carrying the middle ten overs; a mismatch has opened between role and time, and I call it role-time divergence.

The third gap is not technical but sequential. In the ledger, the twelve balls after a wicket produce a run rate of 5.4. As a single event that is not unusual; but across the season, in the sequence of 34 wicket events, the pattern keeps returning. It means we treat a wicket as a crisis and spend three or four overs simply absorbing it — when in T20 cricket, keeping rotation alive from the new batter's very first balls is what divides the scoreboard pressure.

Look at strike rotation rate against stroke choice and a subtler point appears. Right-handers take time to take a single against left-arm spin, working along the off side; balls that leave the line are often left alone. Meanwhile, a strong arm at mid-on creates second-run chances, and in my ledger that has been captured only ten times in eighteen months. The count of twos speaks at once to fitness, fielding arms, and running decisions; Bangladesh's gap here is small, but it is worth four to six runs a match.

The fourth layer is the weather. Evening dew deadens the catch and the spin, and in my ledger spinners' economy rises by about 0.6 in the second innings over the last ten overs. That change is not a change in batting skill, it is a change in conditions; once you accept it, both the toss decision and the question of who bowls in the middle overs have to be recalculated.

The same ledger says that of the runs Bangladesh score in the middle overs, 63 per cent come from running and 37 per cent from boundaries. For England, Australia or India the split is far closer to even, roughly 52 to 55 per cent running and the rest boundaries. Our middle-over structure, then, is not only dependent on fewer balls; it is over-dependent on strike rotation, precisely where the capacity to take singles is most limited.

Cross-check strike rotation rate against dot-ball percentage and another fact appears. In matches where Bangladesh rotated more than 45 per cent of middle-over balls, the result went their way roughly 68 per cent of the time over the past five years. Where rotation fell below 35 per cent, a favourable result came in only 21 per cent of cases. The number shows the single most reliable predictor of win or loss is middle-over strike rotation — not individual flair, not a one-off performance.

One important distinction follows. In the powerplay the chance to hit boundaries is high, but it depends on the quality of the ball. At the death the chance is high, but it depends on the decision to take risk. The middle overs are neither — there, success comes from the consistency of small decisions. The middle ten overs are therefore a kind of arithmetic test, in which a batter's speed of decision is most exposed.

On technique, our batters' use of the sweep and reverse sweep against spin has fallen. In my ledger the sweep accounts for 6.4 per cent of middle-over shots, against 11 to 13 per cent for England or New Zealand. Even without steep bounce, a sweep or pull can use the ball above short leg; our batters take that route less often. Instead they look for the boundary only at long-on, which is the easiest route for fielders to close.

One more thing shows up in the ledger that rarely appears in match reports — the outcome of the first ball of an over. The rate of taking a single off the first ball is 38 per cent; off the second ball it falls to 29 per cent. That small difference reveals something psychological: starting an over, the batter plays the first ball almost by reflex, and then defends a little more on the second.

Another important finding: split the middle overs into two-over bands and the gap between overs 7-8 and overs 13-15 becomes the widest. The worst band is exactly overs 11-12, when the spinners often bowl as a pair and the batter is already halfway through. Risk-taking falls in that window, because the head is calculating wicket preservation for the last five overs. That is our biggest strategic mistake.

Contrarian Angle: Correlation Is Not Causation

The easiest explanation is that Bangladesh cannot play spin. My ledger does not fully support that. The same batters score well against spin in the powerplay, and yet the same spinner's economy rises in the middle overs. The difference is created by the age of the ball and by field placement. The weakness is not in the ability to read spin; it is in the timing of strike rotation.

Another popular idea — that the middle overs go slow because of less physical power than India or Pakistan — is not supported by the physiological data either. By international benchmarks Bangladesh's fitness scores and average ball speed have risen sharply over four years. Yet the scoreboard has changed less, because a mismatch remains between the bowling type and the distribution of roles.

The most important caution is this: you cannot jump straight from a low middle-over run rate to a verdict on results, because pitch conditions, the quality of the opposing attack and the presence of dew all work at once. Naming one cause means denying the others, and that is the biggest error in data analysis.

My view on heat maps is relevant here. Almost every analysis now shows a warm-and-cold map and assumes the map tells you the batter's role. In reality the tracking data shows only the small patch where the ball landed; the batter's decision, the field, the pressure of the over and the type of delivery are all missing. The heat map in modern cricket is therefore the new tea leaf — pleasant to look at, but not enough to make a claim about a batter's actual role.

The second thing that rarely enters the discussion is technical and organisational. At some Bangladeshi venues tracking-camera coverage is limited, and in some matches speed data is fragmentary. Filling that gap often means leaning on overseas tracking data, where bowling types and pitch behaviour differ. When someone uses an imported measurement to explain a local structure, the chance of error rises — and that pressure lands hardest on young batters.

There is one more factor that arrives from beyond the boundary: the coaching structure and the weight of expectation. In domestic cricket, when a young batter is told he must survive the first two balls, the space to learn middle-over rotation shrinks. That instinct is best built in club-level practice, which in our country is largely informal. So a large share of the problem you see on the scoreboard is actually made away from the table — and it cannot be locked shut by looking at a ledger alone.

Finally, the small numerical clue most neglected in any country's middle-overs analysis: the first ball after a bowler change. In my ledger, for Bangladesh the outcome of the first ball of an over is about 27 per cent worse than other balls. The reason is probably the re-evaluation a batter must make when the bowling type changes, which in most cases becomes a dot ball. That single thread has the power to change the direction of a whole innings.

Sources

The core of this analysis is my own ball-by-ball ledger, covering eighteen straight months from November 2026 to April 2026. It holds 114 innings across T20 internationals and the BPL, and 12,480 deliveries. The indicators are defined by me, and readers can verify them themselves.

Bangladesh's first T20 international was played against Zimbabwe in 2026, the starting point of our journey in this format. At the 2026 ICC Men's T20 World Cup, Bangladesh reached the Super Eight for the first time in their history, which forms the backdrop to this discussion. Domestically, the BPL began in 2026 and remains the largest and most consistent source of local T20 data.

Shakib Al Hasan is Bangladesh's leading run-scorer and leading wicket-taker in T20 internationals; both his presence and absence directly shape the middle-over structure. Mahmudullah Riyad carried the death-overs responsibility for years, while more recently batters such as Mehidy Hasan Miraz have shown signs of middle-over consistency.

Takeaway

Next season I will watch three indicators. The first is strike rotation rate between overs 10 and 13. The second is the dot-ball percentage of the number four against left-arm orthodox spin. The third is the overs in which the second spinner is used, especially outside the 7-10 and 13-15 blocks. Cricket's prayer stays the same: repeat, reconcile, and never trust a single match.

Russia 2026 taught me that a data desk is a war room with better coffee, where every number is doubted. But the lesson Rajshahi gave me is harder: without patience, no calculation has the last word. Sitting at Mirpur in the evening dew, you see that numbers never tell the story by themselves — they tell you how much you are rushing.

So the question is simple: next season, will Bangladesh stop in the middle ten overs, or will they run?