Possession in Dots: What My Private Ledger Says About Bangladesh's T20 Powerplay

For two T20 seasons my ledger has kept returning to one line: Bangladesh's powerplay. When the side reaches fifty inside six overs, its dot-ball rate drops to 38-42 percent. When it stalls at 34-38, that rate jumps past 50 percent. Same batting line-up, broadly the same conditions, and yet the outcome swings that far. The scoreboard already tells you the runs. The real question sits elsewhere: what share of deliveries did the batters genuinely push toward the boundary, and what share did they merely block to survive? The gap between those two numbers is what actually defines a T20 innings.

My first xG ledger began as a private argument with the scoreboard. In the 2026-18 Premier League season Burnley finished seventh on 54 points while their expected points stood at 45.1, and they conceded 39 goals from 49.7 xGA. That season taught me a number can be true while the story around it is false. Returning to cricket, I wanted to run the same test on Bangladesh's powerplay.

Watching matches in the ground and on screen year after year, I built one habit: beside every innings I keep a small column where I write "ball played" and "ball hit." The ratio between the two later became my most valuable metric.

A translation layer for possession and penetration

In cricket the separation between possession and penetration is harder than in football, because every delivery can produce a run. In T20 a translation layer is still possible. In football, pass volume measured possession and xG measured penetration. In cricket I use dot-ball rate and the play-to-block ratio for possession; boundary rate per ball, intent-shot share, and powerplay strike rate for penetration. The equivalent of field tilt here is boundary tilt: what share of total runs came from boundaries.

This translation matters, because blurring possession and penetration ruins the analysis. A side can dot 60 percent of deliveries and still lead the scoreboard if it attacks across the other 40. The reverse also holds: 70 percent of deliveries blocked can hold a score together, only to collapse in the final overs.

I split the powerplay into three layers: the survival layer, the rotation layer, and the attack layer. Bangladesh is genuinely excellent in the first layer, technically clean, protecting wickets, rotating strike. The problem lives in the second and third.

Stated identity against actual output

When I started the BDCricTeam page in 2026, I had no data at hand, only eyes and a notebook. That notebook taught me that what a team says and what a team does are never the same thing. Seventeen years later I still hunt for exactly that gap in the table.

Bangladesh's stated T20 batting identity is "technically sound, builds an innings by playing the ball." That identity holds for the first two or three overs. From the fourth, trouble begins: the ball grows older, the field spreads, and the team's intent stops being clear. Does it hold, or does it hit?

My ledger shows that in the final three overs of the powerplay Bangladesh's boundary rate per ball drops to roughly 4.5 percent, against 7 to 8 percent for the better sides. That is the real gap. Forty runs in six overs is not bad if a minimum boundary plan sits behind it. If those runs arrive only through singles, doubles, and the opposition's extras, the foundation is fragile.

Spain completed 1,029 passes, and the goal disappeared into the possession. Many Bangladesh powerplays in T20 follow precisely that fate: possession present, penetration absent.

Testing through three channels

The first channel is the relationship between powerplay dot balls and the strike rate of the last five overs. In matches where the powerplay dot-ball rate passed 45 percent, the final-five-over strike rate generally stalled between 125 and 130. The reason is simple: the side was then forced to chase a deficit, took risk, and lost wickets.

The second channel is the first ball after the powerplay. My notes contain a pattern I call "the curse of the seventh ball." In the over after the powerplay Bangladesh frequently loses a set batter, because spin arrives and the side slows down to "handle" it. Between overs 7 and 12 the run rate falls to 6.5 to 7. Those six overs are the weakest zone of a Bangladesh T20 innings.

The third channel is boundary tilt. When less than 50 percent of an innings' runs come from boundaries, it usually signals that the side could not break the opposition's bowling plan. In Bangladesh's won T20 matches boundary tilt averages 55 to 60 percent; in lost matches it drops to 42 to 45 percent.

The Spain lesson and the two-column ledger

The Spain versus Russia match at the 2026 World Cup in Russia changed my framework. Before the match my model gave Spain a 78 percent win probability. After 120 minutes Spain had 1,029 passes, 75 percent possession, 1.16 xG, and only one open-play goal. Russia had 0.41 xG yet won on penalties. After that match I began placing a penetration metric beside every metric, and added PPDA and field tilt to the framework.

In cricket that penetration metric is the boundary and the intent shot. Playing out 70 percent of a powerplay is possession, but if only 15 percent of those deliveries were genuinely hit, it is hollow possession. My previews are now a two-column ledger: territory on one side, danger on the other.

Stratifying by venue and format

Born in Sri Lanka, working in Bangladesh, I have watched both countries' cricket closely, and seen how fast the same data carries different meaning in two places. At Mirpur the ball stays low and spin bites, so timing breaks down when batters try to hit through the powerplay. At Chattogram the ball comes on a little more. At Sylhet dew falls and the second innings eases. Three venues, three different games. I do not blend Mirpur data with Chattogram data.

Format differs too. A T20 powerplay is not an ODI powerplay, because ODI wickets cost more. In Test cricket possession means something else entirely, where playing out deliveries can itself be the objective. Without this stratification every conclusion drifts the wrong way.

The mirage file

I keep a "mirage file" of sides whose results outrun expectation. Burnley's seventh place was the first entry. In cricket that file has taught me that one good series or one good season does not equal structural improvement. A good Bangladesh powerplay series does not fool me either, not until the relationship between dot balls and boundaries shifts permanently.

Stratifying by opposition quality

There is a wide gap between Bangladesh's powerplay strike rate against top-five sides and against associates. Top sides bowl to a clear plan: they force the top order to play the ball and squeeze in the middle overs. Against associates that squeeze is looser, so the numbers look flattering. Averaging the two tiers together produces nothing useful.

The variance tribunal

Without separating correlation from causation, analysis becomes a comfortable story. I did not trust the table until it survived a season of variance. So I kept a holdout: built the pattern on one season's data, then tested it in the next.

The negative relationship between powerplay dot-ball rate and final-five-over strike rate survived into the following season, though weaker. A signal exists, but it alone does not decide a match.

When stadiums emptied in 2026, I found Bundesliga home win rate fell from 43.3 percent to 33.8 percent, and home goals per game from 1.74 to 1.29. From then on I folded context variables into the model rather than leaving them outside: crowd absence, travel, rest days. In cricket, crowds and venues play the same role, which is why every angle in my newsletter must pass a context filter.

The player layer

At player level my ledger surfaces a few names. Litton Das is the team's best on both the powerplay play-rate and boundary-rate metrics. His problem is not consistency but fluctuating intent. Najmul Hossain Shanto's profile is the inverse: he plays more balls and takes fewer risks. Towhid Hridoy is the fastest-scoring middle-order batter, but his innings frequently begin after the twelfth over.

Here sits a structural problem. Bangladesh's top order bats to protect wickets while the middle order bats with wickets already lost. The top order's conservatism therefore loads weight onto the middle order, and nobody accounts for it. The biggest counterintuitive discovery for me is this: Bangladesh's T20 problem is not the quality of the batting order, but the habit of thinking about the three phases of an innings in isolation.

The possession paradox in bowling

The possession-penetration framework applies to bowling too. Mustafizur Rahman creates dot balls in the powerplay with cutters and slower balls. But a dot ball is possession, not penetration. If the side concedes 45 in the powerplay but takes two wickets, it has traded possession for penetration. If it concedes 35 without a wicket, that is hollow possession, and it returns with interest in later overs.

My ledger shows that in Bangladesh's won T20 matches the number of powerplay wickets matters more than the dot-ball rate. Squeezing the opposition and breaking its batting structure are not the same thing, and the second is worth more.

In spin, the rise of Rishad Hossain has added a new dimension, taking middle-over wickets that were once a chronic gap. His economy and wicket count still need balance; chasing wickets alone sometimes leaks runs. Taskin Ahmed faces the same arithmetic: a powerplay dot ball and a death-over wicket carry different prices. Shakib Al Hasan's experience steadies the middle overs, but stability and penetration are not the same commodity.

The counterintuitive angle

Now the point where I want to stand against my own table. A dot ball is not always bad. Saying otherwise is half a truth. A dot that pins a set batter for an over and brings the next batter to the crease is a good dot. A dot that wastes a free-hit opportunity is a bad dot.

A table can only count; it cannot understand. That is the trap of data blindness. Many times I have seen someone wave a table and declare a side passive, while the match showed the side striking at the right moments, only for the strike to end as a catch rather than a boundary.

Opposition quality is another trap. Bangladesh's dot-ball rate is naturally higher against top-five sides. Reading the table without that context makes every decision wrong. So every contrarian claim of mine must beat a simple base-rate model. If the base rate for powerplay run rate is 7.5, showing 8.2 proves nothing. Proof comes from boundary tilt, the timing of wicket loss, and the pattern of ball played.

What I will watch next series

The regular season is a game of patience. The tactical and fitness currents running beneath the table show up late. In the coming series I will watch three things. First, Bangladesh's boundary rate per ball in the final three overs of the powerplay: if it clears 5 percent, the risk of a middle-over collapse falls. Second, the run rate between overs 7 and 12, the real examination. Third, the intent of Shanto and Litton: are they playing the ball, or hunting the boundary?

The scoreboard will say how many runs were made. My ledger will want to know where those runs came from.