Blank Spreadsheet, Crowded Middle Overs: Bangladesh's Data Forensics at the Asia Cup

In a match at the last Asia Cup, in the 31st over of the second innings, I was staring at a spreadsheet open on my laptop. The cell next to the batter's name was empty. That cell was supposed to hold a number — what Bangladesh did in the first three balls of the over. The stadium scoreboard said 140/4; my spreadsheet said something more uncomfortable. In that match, of Bangladesh's 37 dot balls in the middle overs, 21 came against spin, and a large share of those arrived with a left-hander at the crease. The television commentary kept reaching for the word 'pressure.' I had an incomplete table and one question: where is pressure actually born — in the batter's head, or in those empty cells?

I opened a blank spreadsheet and let the Bangladesh Premier League teach me. In 2026 in Rangpur, I balanced rice-mill accounts by day and hand-coded a model by night. What professional football calls xG has no direct equivalent in cricket. So I had to set my own weights: which ball counts as 'success,' which as 'wasted,' and which exists only to inflate a number. This piece is an extra page from that old habit — the shape of Bangladesh's innings in the Asia Cup, its gaps, and what those gaps let us infer.

The Asia Cup is a curious thing. It is not as long as a World Cup, and not as comfortable as a bilateral series. A short group, a quick semi-final, and a do-or-die tone in every match. Bangladesh's history is oddly tangled with this format. On 22 March 2026, in Mirpur, Bangladesh lost the Asia Cup final to Pakistan by 2 runs — two needed off the last ball, and they did not come. In the 2026 T20 final, an 8-wicket defeat to India. On 28 September 2026, in Dubai, a 3-wicket final loss to India in which Liton Das made 121. In 2026, a group-stage exit.

Place those four results side by side and a pattern surfaces: Bangladesh reaches the big match, but cannot hold the tempo of an innings through the middle overs. That is not commentary mood; it is a hypothesis I wanted to test. And the first problem in testing it was data: for cricket in this region, advanced public data barely exists.

The 2026 edition is no exception to the pattern. Bangladesh went out in the group stage there, losing to Afghanistan and Sri Lanka. In my notebook, the middle-over run rate in those two matches was even lower, closer to 3.7. When tournament pressure rises, the old habit returns rather than a new one forming.

So I worked with hand-counted information — 14 Asia Cup matches, phase-by-phase figures written into a notebook. The sample is small, and I will not hide it. Every number of mine is one-third measured (scorecards, ball-by-ball events), one-third modelled (my own weights), and one-third guessed. Where it is a guess, I say so.

In my hand-counted data, the biggest recurring gap is not the powerplay — it is the ten overs after it. With fielding restrictions in the powerplay, boundaries are more available; between overs 7 and 15 the field spreads, the gaps for singles shrink, and the dot-ball count jumps. In my count across those 14 matches, Bangladesh's run rate was 5.8 in the first six overs and 4.1 in overs 7–15. The gap is 1.7 runs, which looks small; but by dot balls the picture changes.

The dot-ball rate was 38 per cent in the first six overs and 52 per cent in overs 7–15. That is roughly one extra dot per over. Ten extra dots in sixty balls means nearly two overs burned — and in a short tournament's knockout, two overs decide matches. In the 2026 final, Bangladesh made 222 in 50 overs; India chased 223 in 48.2. That match is a story of defeat, but it is also a story of the middle overs — beyond Liton's 121, almost no one else settled.

And these dots are not all alike. One pattern kept returning in my notebook: in the middle overs, a large share of Bangladesh's dot balls came in the left-hander versus off-spin or leg-spin matchup. The reason is simple — most Asia Cup pitches are slow, the ball keeps low, and spinners then hold the line outside leg stump and turn it. For a left-hander, pushing that ball to cover or mid-off becomes hard.

I first understood this inside Bangladesh Premier League data. Domestic-tournament information is incomplete and even flawed; but that is exactly where these matchups recur. On the slow Mirpur surface, what spinners do between overs 7 and 15 — holding the ball outside the stumps and taking the bounce away — appears in almost every home match. Before the Asia Cup, that was the clearest signal on my table.

Shakib Al Hasan's name keeps returning to this discussion because he plays it differently. The shape of his innings often shows he is unafraid to take risk in the middle overs — hunting small boundaries alongside the singles. Mushfiqur Rahim also keeps the scoreboard moving in that phase, but the job of raising the tempo usually falls on the next batter. In my count the difference between their roles looks small, yet its effect on the direction of a match is large.

The 2026 final is the exception. Liton Das made 121 there and showed that holding the tempo through the middle is possible — if the batter is willing to change his line. But one man's exception does not become an indictment of the whole structure. If the rest of the innings plays 52 per cent dot balls in those same overs, one brilliant knock cannot win the match.

The death phase is unclear too. Bowlers like Mustafizur Rahman and Taskin Ahmed build pressure in the final overs, but that shows up in the opposition's data as well — so teams try to take their runs before the last over arrives. My notebook suggests that a side scoring well in the first 30 overs can take fewer risks in the last ten. In other words, death-over success is often the harvest of the earlier thirty overs, not just the skill of the last five.

The bowling side is just as blank. Bangladesh's spinners choke runs in the middle overs — that is clear in measured data — but how many balls were on a good line and length, and how many were short, no one counts. In my notebook, in at least five of those 14 matches the spinners' economy was under 4, yet the wickets were few. Controlling the ball and breaking the match are two different jobs, and we routinely fold them into one.

And here comes my favourite part — the empty cells. Public data carries runs per over, dot balls, boundaries; it does not carry field settings, the type of line and length, or which bowler was brought on against which batter. In Asia Cup matches I noted myself who bowled which over; the television graphics never show it. The cells that stay empty are often the ones hiding the real reason for a decision.

That model was crude, but the empty cells whispered more truth than the goals did. Cricket has no goals, so the truth arrives elsewhere — where runs did not come, which over wasted the ball, which matchup kept trapping the batter. I never treat data as a verdict; I treat it as a mirror, to be held against the naked-eye reading.

A caution is due here. Every number in this piece shows a relationship, not a cause. Dots are rising and the team is losing — the two happen together, but that does not mean dots are the sole cause of defeat. The pitch may have been slow, the toss may have mattered, the opposition spinner may have bowled better than on any previous day. In a small sample, any pattern quickly builds a story. My job is not to build stories; it is to see which story the data will hold.

So I keep the habit of watching twice. Since Russia 2026 I have learned to watch Germany twice: once with the eyes, once with PPDA. The same rule applies to cricket. The scorecard says 140/4, but the eye sees a batter who cannot move his feet, a bowler holding the ball outside the stumps, and a keeper standing up to the stumps. Only when the two readings agree do I dare write it down.

One more thing is worth remembering: data being absent and data being meaningless are not the same. Who collects the data decides what becomes visible. Broadcasters put on the table what they want shown — big names, big shots, big runs. Where cameras are few, as in domestic cricket, decisions are often made from memory rather than data. In the Asia Cup context, my biggest discovery is not that Bangladesh is slow in the middle overs; it is why we kept that fact outside the count for so long.

A model is a monastery: you enter to escape noise, then hear it clearer. Three thousand four hundred balls of data, 14 matches, one notebook — none of it can settle a team's fate. But this small structure pulls me toward a specific question that gets lost in the noise of commentary. And if the question is right, even a wrong answer teaches something.

For the next Asia Cup my notebook now carries one line: count overs 7 to 15, not only runs but dot balls and the number of line changes. If a side can cut one extra dot per over in that ten-over block, that single run may be the difference in a short tournament's knockout. The question remains — will we ever learn to fill that empty cell on the scoreboard, or will we keep trusting the word 'pressure' in the commentary box?