41.7% Dot Balls in the Powerplay: The Gap Bangladesh's Scoreboard Keeps Hiding

At 12:40 am Sydney time last Wednesday, I logged Bangladesh's last three T20Is ball by ball, first six overs only. One hundred and eight balls, 123 runs, 45 dots. As a percentage, 41.7. What stopped me was not the dot-ball rate but the next line: the same batting unit scored at 8.9 an over between overs seven and twelve, two runs faster than its powerplay. A team that bats carefully at the top suddenly accelerates later. That inverted picture sits outside the model, and that is where this piece starts. I do not trust a number I cannot trace to a touch, so I moved from the ball-by-ball log to the shot map. One clarification up front: every figure here comes from my own tracking sheet, not a broadcast graphic.

Bangladesh's T20I story began in November 2026 in Khulna, against Zimbabwe. Beating India at the 2026 World Cup, losing the 2026 Nidahas final off the last ball after making 166/8, reaching the Asia Cup final that same year, and a Super Eight finish at the 2026 T20 World Cup — this format has always been tense with Bangladesh. The next World Cup runs in February and March 2026 across India and Sri Lanka. As a preparatory indicator, the six powerplay overs are the most honest sample available: only two fielders are outside the circle, the ball is hardest, and decisions come fastest.

I borrow my method from football. At Euro 2026, Italy's PPDA was 8.7 and Jorginho covered 12.9 kilometres per match. That pressing index became a question for me: how much pressure does a batter actually absorb in a powerplay? Which is how my Powerplay Pressure Index (PPI) came about: dots percentage times 0.6, plus false-shot percentage times 0.4. The higher the number, the worse it is for batting. In 2026 I logged 1,248 shots in Excel and built my first xG model in a Sydney bedroom. That year taught me that France scored four from 2.1 xG while Argentina scored three from 1.4, and Croatia turned 10.8 xG into 14 goals, six of them from set pieces. Measurement and outcome never align perfectly. In cricket, that gap is called shot selection.

Core analysis

First layer: the anatomy of the dots. Of the 45, I counted 21 as deliberate defensive blocks, 13 as misses or edges, and 11 as non-turning dots — the ball was hit but no run was taken. That last group carries the heaviest cost, because it is not a talent limit. It is a decision failure.

Second layer: boundary distribution. Of the 12 boundaries, eight were fours and four were sixes. Nine came in the first two overs, using the new ball. From the fourth to the sixth over, only three boundaries arrived. The middle of the powerplay is the hole. Fielders adjust, runs dry up.

Third layer: venue. In my tracking, first-innings averages at six overs read: Sylhet 44/1 (7.33 an over), Mirpur 41/2 (6.83), Chattogram 48/1 (8.00). On Sylhet and Mirpur surfaces, slow and low, the ball takes time to reach the bat. Batting at 7.33 there is a condition, not a weakness. Measured against a world benchmark of 8.8, Bangladesh sit 1.97 runs behind. But that comparison is drawn against other venues, which is why a raw number fails its own audit.

Fourth layer: role stability. Across the last 14 T20Is in my log, Bangladesh have used eight different opening pairs. Change the names and the language of shot selection changes too. A pair that plays 8-10 matches together accumulates data that eight pairs never will.

Now the opponent check. Same three matches, same venues, 18 powerplay overs: Bangladesh 123 runs (6.83), 45 dots (41.7 percent), 12 boundaries (11.1 percent). Opponents 138 runs (7.67), 39 dots (36.1 percent), 15 boundaries (13.9 percent). On PPI, Bangladesh 33.4 against 29.3. The real gap is 0.83 runs an over, not 1.5. Venue adjustment is where the model's pride gets cut down.

The contrarian turn

This is where the story bends. Had I already written the verdict on that 1.5-run powerplay leak, I would have written the wrong verdict. Between overs seven and sixteen, Bangladesh made 267 runs at 8.9, against 246 at 8.2 from opponents. A large part of the powerplay shortfall is recovered in the middle. The genuine damage sits in overs 17 to 20: Bangladesh 96 runs at 8.0, opponents 118 at 9.83. The death-over gap is 1.83 runs an over, more than twice the powerplay gap.

So the problem the team is discussing may not be its main problem. Correlation and causation blur easily here. A low powerplay rate looks like slow batting, but it can equally be wickets preserved so that men like Towhid Hridoy and Jaker Ali take over from the fourteenth over. My data shows that 0.7-run gain in the middle. When a trade-off pays off net, it is not a flaw; it is a hidden asset.

Second doubt: sample. Three matches are loud, thirty are honest. Venue, dew, the opposition bowling unit, the target — without holding those constant, 41.7 percent cannot be called a permanent trait. I set my threshold in advance: across a 12-match block, a powerplay boundary share above 14 percent is signal, below it is noise.

Third doubt comes from history. In the 2026 T20 World Cup final, India were 34 for 3 inside five overs, their powerplay a nightmare. They still won, making 176/7, because overs 17 to 20 were their strongest. The model said one thing; the empty stadium said another. In T20 cricket, the empty stadium is the last four overs.

What to watch next

Three things for the next series. First, the powerplay boundary share — whether it moves into the 14 percent band rather than 12. Second, whether the number three role settles; if the instability of eight opening pairs ends, the language of shot selection will unify. Third, and most important, the economy from overs 17 to 20, because a 1.83-run gap outweighs the powerplay one, and on Asia's slow surfaces that is what decides matches.

One question stays open on my desk: does Bangladesh raise its powerplay risk, or hold the trade-off and fix the death bowling? The first costs wickets, the second saves matches, and the data supports both. Small samples are loud; large samples are honest. The answer arrives after twelve matches, and until then I will keep logging every six overs.