Half the Story Behind Defending 106: Three Layers of Bangladesh's Powerplay Data

16 June 2026, Kingstown. A rain-dampened pitch, slow off the surface. Bangladesh were bowled out for 106, then bundled Nepal out for 85 to win by 21 runs and reach the T20 World Cup Super Eight for the first time. I watched it on an old laptop with my handwritten ball-by-ball log sheet open beside it. By full time, every conversation around me had turned into a tribute to the bowlers. My first entry ran the other way: 106 in 20 overs is 5.3 runs an over. I am not denying the bowling — the defence was real. But when a scoreboard shows a 21-run win, the story it tells usually buries the reasons for the collapse. To me a scoreline is not a verdict; it is a charge sheet.

In 2026, after a knee injury ended my semi-pro career, I returned to Mymensingh and took a volunteer data role with Sheikh Russel KC. Against Abahani Limited Dhaka in the Bangladesh Premier League, I logged every shot by hand and built a basic xG model. The model gave 2.7 against 0.8; the match finished 1-1. The post travelled to 1,200 people and to scouts in Dhaka. In Mymensingh, the first xG model was a lantern in a league of shadows — it taught me that you can count in poor light, provided you write down how you counted.

Cricket changed the method, not the principle. Every number in this piece needs a tier. Tier A: manual ball-by-ball logging from broadcast, dated. Tier B: broadcast-derived calculations where delivery labelling is partly inferred, so treat the margin as plus or minus 0.3 runs. Tier C: memory or interview-based, usable only as a lead, never as evidence. Domestic cricket here has no tracking cameras, incomplete scorecards and almost no institutional memory. So I keep an append-only ledger for data: every delivery is one entry, nothing behind it can be deleted, only appended. Some call that a blockchain-style ledger; I call it an audit trail. A model without context is just a calculator wearing a scout.

Powerplay: where the match is given away early. In my Tier-B log, Bangladesh's first-six-overs run rate across the 2026 T20 World Cup sat below seven, while the sides that went deep averaged around eight or a little above. Take a 7-8 run gap per powerplay; across four matches that is a 30-run deficit nobody recovers at the death. The cause is not intent but structure. The opening pair never settled into a combination, and the batting culture is still cast in the mould of saving wickets and hitting late. On slow pitches that is defensible, but a defensible strategy that consistently fails to reach 180 stops being a strategy and becomes a habit.

Middle overs: where spin was a shield when it could have been a sword. Between overs 7 and 15, the Mehidy Hasan Miraz and Rishad Hossain pairing was the spine of Bangladesh's bowling. Rishad's leg-spin mix of googly and slider squeezed opposition scoring rates — that is clear in my log. The problem sits on the other face of the same overs: where we bought time with the ball, we lost it with the bat. Our middle-over strike rate hovered around 6.5 while the best teams touched eight. Towhid Hridoy was among the few who attacked spin positively, but the middle order as a whole rotated slowly — turning down ones and twos to wait for boundaries simply makes the opposing captain's bowling changes easier.

Death overs: one man's skill and a squad's gap. Mustafizur Rahman kept a death-over economy under eight in that tournament; when the ball was in his hand at 16-20, there was genuine fear in the air. With the bat, the picture inverts. Our scoring rate in the last four overs ranked near the bottom of the tournament, and wickets fell quickly — because after a partnership broke, a new batter was still being given time to settle, when no such time exists. In T20, the death overs mean your three best batters facing the opposition's best four overs; too often we sent a fresh face at number four into that window.

This is where I want to stop, because the easiest conclusion here is the wrong one. The empty stadiums of 2026 taught me that silence can be a data source — and that environment distorts data. The 2026 Caribbean pitches favoured bowlers; chasing small targets, opposition sides made errors, and our bowlers collected the interest on those errors. So the verdict that our bowling was outstanding risks mistaking correlation for causation. Likewise, franchise auctions have priced powerplay hitters through the roof in recent years. The transfer market, football or cricket, is a rumour engine; I only turn its gears with data. Where domestic records are incomplete, that price is the price of feeling, not of a model. And an uncomfortable question follows: is the powerplay weakness about intent, or about the talent pipeline? We push boys into senior rhythms early, before their bodies are built — workload, travel, pressure. Those who look physically mature as teenagers carry the heaviest load, and the bill arrives later as injury and stalled strike rates. That number never appears on a scoreboard.

My pre-registered forecast for the next cycle: if the powerplay dot-ball percentage stays stuck around 45, no Bangladesh bowling attack will save consecutive matches deep in a tournament. Confidence interval plus or minus 15 percent, central probability about 70 percent. The signal to watch is not the result but the ratio of boundaries to dot balls in the first six overs. Change that ratio and the story changes; leave it and we will again celebrate a win that already had the next defeat written inside it.