The Dot-Ball Spreadsheet: From a 106-Run Win to a 2026 T20 Blueprint

Hook: What the Scorecard Doesn't Say

On 16 June 2026, at Arnos Vale in Kingstown, Bangladesh were bowled out for 106 in 19.3 overs — one of their lowest totals ever at a T20 World Cup. Nobody holding the scorecard could have predicted what came next. Nepal collapsed to 85 in 19.2 overs. Bangladesh won by 21 runs. Player of the match: Tanzim Hasan Sakib, 4 overs, 7 runs, 4 wickets.

I watched that match twice. Once with my eyes, once with a hand-written ball-by-ball tally. The two viewings disagreed. The eyes said: Tanzim's line and length were impeccable, hence 4 for 7. The tally said something more specific — the runs came from a tiny handful of shots, and across both innings a very large number of deliveries passed without a single run. The wickets were the final scene. The dot balls were the script.

This piece is about that script. The reason is simple. In Bangladeshi T20 discussion we almost always measure strike rate. Strike rate is an output. The thing that produces it — the dot ball — we barely measure at all. And the more tournament pressure rises, the bigger this blind spot becomes, because tournament matches are won and lost in low-scoring, slow-pitch, dot-heavy conditions.

Context: A Blank Spreadsheet and Three Tiers of Numbers

I opened a blank spreadsheet and let the BPL teach me. That was 2026, and I was forty. By day I audited rice-mill accounts in Rangpur; by night I hand-coded an expected-goals model, because no public model existed for that league. The season ended with 132 matches, 3,410 shots, and my own distance-and-angle weights. In one title run, the gap between a club's actual goals and the model's expected goals was 9.4. Within a week, three betting syndicates emailed me.

After that piece I stopped writing match reports and started writing methodology notes. Every claim now carries its sample size, its weighting choices, and a stated error margin. My sentences got shorter, my footnotes got longer, and every number now sits next to a label: measured, modelled, or guessed.

In T20 that labelling discipline matters even more, because domestic Bangladeshi cricket has a severe data deficit. The BPL has no ball-tracking, no field-placement mapping, no pressure index. What exists is broadcast-driven scorecards and some over-rate data. So dot balls, shot direction, bowler type — I code those by hand from video. After each match a table fills up: runs per ball, the batter's shot, the bowler's line, fielder positions as far as the camera reveals them, and the over number.

One thing must be stated plainly. This tally is not an official dataset. It is not inside information from a broadcaster or a board. It is a crude, incomplete, personal spreadsheet. I know where the cells are empty. I know where my eye failed to read a bowler's wrist. I know where the sample is too small to conclude anything. I write from it anyway, because an imperfect count still beats a memorised slogan. The model was crude, but the missing cells confessed more than the goals.

Core Analysis

One: Confusing Input with Output

Strike rate is runs divided by balls. Fewer dot balls mechanically raise it. The reverse is not true — a higher strike rate does not guarantee fewer dots. One batter makes 54 off 40 with 18 dot balls; another makes 54 off 40 with nine. Both have a strike rate of 135. The impact on the innings is worlds apart.

Dot-ball percentage is the gravity of an innings; strike rate is its shadow. Change the gravity and the shadow moves. Chase the shadow and you never find the gravity.

Here I built a model I myself call crude. I named it the dot-adjusted scoring rate: a batter's runs, divided only by the balls on which he actually scored. Beside it I keep a second number — boundary dependence, the share of runs coming from fours and sixes. A batter with 70 per cent boundary dependence and a 45 per cent dot rate is a volatile asset. One with 50 per cent boundary dependence and a 30 per cent dot rate is stable.

Why read the two together? Because T20 batting produces two kinds of runs: runs that buy balls, and runs that buy time. Boundaries buy balls but pay in risk. Singles and twos buy time but pay in patience. In a knockout on a slow pitch, the chance to buy balls shrinks. What survives is the ability to buy time.

Two: Splitting the Innings by Phase

The powerplay, the middle overs and the death — a dot ball means something different in each, and this distinction is the most neglected thing in Bangladeshi T20 conversation.

In the powerplay only two fielders are outside the circle, so boundaries are comparatively available. A dot there is a failure, because the opportunity existed. At the death, batters take risks, so dots fall while wickets rise; the interpretation changes again.

The middle overs — seven to fifteen — are a different arithmetic entirely. Spinners bowl, two fielders protect the boundary, the ball is slow or turning. The only route to runs is rotation: one, two, and the occasional boundary. A dot ball here costs far more than a powerplay dot, because a dot in the middle means creating a hole in the centre of an innings, and that hole's weight lands on the batter in the last five overs.

My tally keeps returning one pattern: Bangladesh's T20 weakness is not the powerplay, it is the middle overs. We survive the powerplay reasonably well, because the game is easier there and batters can play their natural shots. But when a spinner comes on in the seventh over, when the ball starts to turn, when two fielders drop deep, we arrive at a place where there is no plan — only hope.

At Mirpur I have separately counted middle-over dot rates. The sample is small, so I will not offer a definitive figure, but the direction is clear: the same team, the same batters, post fewer dots in Sylhet or Chattogram where the ball comes on, and more at Mirpur where it holds. That is no mystery. But it rarely reaches the decision-making table.

Three: The False Strike Rate

In my files I keep an informal category I call the false strike rate. Two conditions: strike rate above 130, dot-ball percentage above 40. These batters look good on the card and cost their team in the middle.

Run the arithmetic. 54 off 40 with 18 dots, five fours, two sixes, and 21 runs from the other 15 balls. Eighteen dots means eighteen deliveries on which the batter at the other end never had strike, the fielding side regrouped, and the bowler could change his plan. Those 18 balls spent the team's resources without returning runs.

Now invert it: 54 off 40, nine dots, three fours, two sixes, 28 runs from the other 26 balls. Identical columns. But the second batter rotated strike every over, forced the bowler to keep revising, and gave his partner room to survive.

Same strike rate, different dot-ball architecture — the auction bids up the first and the match is won by the second. I do not say this lightly. What sits on a BPL auction table is runs, strike rate and a few highlight clips. Phase-specific dot data sits nowhere, because nobody has organised it. And the empty cell is where the decision gets made.

Four: Different Pitches, One Squad

Pitch variation in Bangladeshi domestic T20 is wider than stadium variation in European football leagues. Mirpur is slow, the ball holds, spinners thrive. Sylhet offers pace onto the bat, shorter boundaries, higher scores. Chattogram sits between the two, but evening dew rewrites the whole calculation — gripping the ball in the second innings becomes hard.

Here franchises make a basic error: they build one squad for every venue. A spin-based middle-order batter is an asset at Mirpur and a liability in Sylhet. A power hitter is gold in Sylhet and a stuck engine at Mirpur. Yet auction prices are set by a single rating, and that rating is usually an average across venues — correct for none of them.

I have computed venue-specific dot rates, and I have had to stay cautious about sample size. A few matches per venue per season, maybe eighty to ninety balls for a given batter in a given phase. In eighty balls, two lucky edges or two dropped catches can flip the whole picture. So I make no hard numeric claim here. I say only that the direction is clear enough that ignoring it needs a justification nobody has offered.

Five: The Blind Cell at the Auction

After every BPL auction I do one exercise. I take the batters bought for the highest prices and pull their dot-ball percentage from the previous season. Then I look at the relationship between price and dot rate.

The relationship is weak, and whatever exists can run the wrong way. The reason is not complicated. Auction value is set by three things: recent scorecards, television clips, and an internal story that says 'we know this player'. None of the three counts dot balls.

I do not want to assign blame. A franchise's cricket operations team has little time and almost nobody to code ball-by-ball video. Decisions get made on the data that is easy to reach. That is human nature, and that nature is producing the BPL market's largest inefficiency.

The cell nobody fills is where the decision gets made. An empty cell is not neutral — it means nobody asked the question.

Six: The 2026 Evidence and Tanzim's Seven Runs

Back to the Nepal match. Tanzim Hasan Sakib took 4 for 7 in four overs. The scorecard calls that an outstanding bowling performance. I would call it outstanding, but incomplete as a story.

Bangladesh's 106 was no ordinary innings — Arnos Vale was not Mirpur, but it was not easy either, and Nepal's bowlers held their lines. Both sides fell into the same trap: dots accumulated, batters could not rotate strike, and the pressure attached to every boundary became abnormal. The 21-run margin was not a batting margin. It was a margin of dot-ball management.

A bowler does not take wickets; a bowler buys time, and the wickets arrive later. I learned that sentence from a ball-by-ball tally. When a side is pinned to one or two runs an over for four or five overs, a batter is forced into a shot he would never play in normal conditions. That compulsion manufactures the wicket. The wicket is the symptom; the dot ball is the disease.

Seven: The Two-Eye Habit

After Russia 2026 I watched Germany twice: once with my eyes, once through PPDA. Before the tournament I had written that Germany's press had already decayed — their PPDA had drifted from 8.9 in qualifying to 12.6 at the tournament. Germany went out in the group stage.

But my model had ranked them third favourites. So I hedged the language, left both doors open, and lost the argument anyway — because I chose safety over evidence. That lesson produced a two-track habit: a loud public thesis, and a quiet appendix listing everything I got wrong.

This article has an appendix too. It is written into the next section.

Contrarian Angle: Do Dot Balls Win Matches, or Do the Causes of Dot Balls Win Them?

There is a trap here, and I have nearly fallen into it twice. Fewer dots correlates with winning. But the cause is not the dot ball.

Consider. Dots rise on slow Mirpur pitches. That is the pitch, not the batter's failure. A side chasing 180 takes risk, so its dots fall — not because its batting is better, but because it is sprinting toward a defeat and the arithmetic gets messy. A side that posts 120 and defends will win while its batters keep playing dots.

So are dots meaningless? No. A dot is a symptom — the sum of bowler, pitch, match state and batter decision. A sum can be measured, and that is exactly why it matters. But blaming an individual because of a sum puts the analysis in the wrong place.

The second trap is subtler. Cricket has imported an 'intent' metric — positive intent, attacking-shot percentage, the courage of a rising run rate. This is football's distance-covered and high-intensity sprints in a different shirt: easy to measure, pretty to print, often meaningless. The batter who tries to hit every ball for six tops the intent chart and fails to lift his team. The batter who accumulates singles sits at the bottom of the chart and moves the scoreboard.

I am not arguing against risk. I am arguing that cricket's most valuable work is usually silent — stealing a dot, rotating strike, taking a single while protecting a boundary. The unmeasured acts are the ones that keep an innings alive.

And one admission about myself. My model is crude. The numbers I call measured, I pulled off a screen with two eyes, so they contain errors. The numbers I call modelled rest on weights I invented, so they contain bias. Some middle-over figures I mark as guesses outright, because the sample is eighty or ninety balls, where two dropped catches can reverse a conclusion. I will never call my spreadsheet the final word. It is a handkerchief, and for now it keeps the dust off.

Takeaway: What to Watch in the 2026 Cycle

The 2026 T20 World Cup is in India and Sri Lanka. That venue list means Bangladesh must play on two different kinds of ground — India's batting-friendly surfaces and Sri Lanka's slow, turning, spin-friendly pitches. Two different run-economy models inside one tournament. Anyone arriving with a single plan will be unprepared for at least half their matches.

In the next cycle I will watch three signals. First, the middle-over dot-ball rate across the first ten BPL matches — if it does not fall, Bangladesh's real T20 crisis will never be solved. Second, whether phase-specific data reaches the auction table — because if one franchise starts buying it, the rest will follow. Third, the powerplay boundary rate — because the baseline set on slow pitches in the first six overs fixes the entire arithmetic of the last five.

And one question I am keeping for myself. The next time Bangladesh are bowled out for 106 and still win, will we look at the number — or will we leave the empty cells behind it in the dark again?