The Quiet Powerplay Deficit: What Bangladesh's Six-Over xG Keeps Hidden

Where the scoreboard promises, the data asks for an audit

Across three domestic T20 matches I have watched this season, the same scene repeated. Under the Sher-e-Bangla floodlights, the first six overs closed at 48/1, 53/0, 45/2. In the commentary box, among former cricketers in the next row, even inside the dressing-room language, one line held: good powerplay. Then, later in the night, I opened my own shot-quality table and found the expected runs behind that fifty were only 39 to 43.

Nine to thirteen runs arrived for free — over the edge, on a dive in the deep, or two inches inside the rope. The scoreboard behaves like a promise in Bangladesh. Fifty in the powerplay means the plan worked, and that belief is so fixed that nobody asks where the runs came from. I am not here to assign blame. I only want one question asked: if that fifty becomes thirty-three across the next three matches, what will we have learned?

How the accounting was built

In 2026, at twenty-four, I joined Golpo Sports as a junior data analyst from a flat in Rajshahi. My first task was coding 1,248 shots from the 2026-17 Bangladesh Premier League. Back then I treated data as scripture — Abahani Limited scored 34 goals from 27.6 xG, Sheikh Jamal Dhanmondi 29 from 31.2. That twelve-part series taught me that replacing the word 'deserved' with 'xG differential' changes the reader's questions. In Bangladesh, I taught a league to see its own xG; in cricket that work is still unfinished.

Football machinery does not bolt directly onto cricket. The value of a powerplay shot depends on field setting, bowler length, how much of the line the batter missed, how much bounce the pitch offered. PPDA showed me Germany, but before importing that lesson I had to decide what 'press' means in cricket. My definition is simple: a dot ball created inside the fielding circle is a press; a single conceded by pushing a fielder back is a press break. That mapping is an assumption, not doctrine — and I write it down before tagging any match.

Now the data reality. Tracking-camera density is thin in Bangladesh, and Hawk-Eye is not installed at every venue. My model runs on field mapping and video tagging — two scorers, one video analyst, six angles. That limitation reduces precision, but I do not hide it. An ESTJ builds the pipeline first and the poetry second.

What actually happens inside six overs

My powerplay model separates three things: contact quality by line and length, the intended direction of the shot, and where the ball travelled relative to fielders. Combine them and the output is expected runs for the six-over block, shortened to ER6. Across 2,094 powerplay deliveries from the last two domestic seasons, the picture is uncomfortable.

Twenty-nine percent of a team's total powerplay boundaries came off the edge or a mis-hit. Shots straight over cover are rare; attempts to clear third man are frequent. The batter is attacking rather than respecting the defensive field, but the ball is not meeting the middle. The second number is more familiar: an average of 2.4 full tosses or half-volleys per innings, sixty-three percent of which are hit for boundaries — meaning half the powerplay runs come from bowler error, not batter planning.

The third number teaches the most. Dot-ball rate is 51 percent in the first over, falls to 42 percent in the third, and climbs back to 49 percent in the sixth. That dip is not coincidence. The new ball grips slightly, the batter takes two overs to settle, then the seamers shorten their length and dots return. Teams that read this curve attack in the third and fourth overs and reduce risk in the first and sixth. Our domestic cricket still treats the powerplay as one undifferentiated block.

Bowling type widens the gap

Two kinds of seamer share a match: those who get the new ball to swing upright, and those who extract bounce from the cross-seam. In my data the second group shows a better powerplay economy, yet the ER6 they concede is nearly identical. The ball reaches the bat, contact happens, the edge arrives. I do not play fantasy cricket, because tracking data does not exist there; but in domestic video rooms I see this gap clearly.

On spin, an uncomfortable fact: of the 22 percent of powerplay boundaries that come off spinners, half are attempts at the slog sweep or reverse sweep. When they connect, the match changes pace; when they fail, the side drops four to five runs and must recover in the middle overs. Teams without strong strike rotation lose wickets trying to repay that debt.

How Bangladesh sits against the rest of Asia

I placed India, Pakistan, Sri Lanka and Afghanistan powerplay data in the same frame, under the same limitations. Bangladesh's dot-ball rate runs four to six percent higher, and its count of 'intended boundaries' — shots where no fielder was in the way and only the line was missed — is 1.8 lower per innings. The gap is not enormous, but 1.8 boundaries in T20 is seven to eight runs, and seven runs often decides a result.

Venue frays the picture further. On Dhaka's slow surface, chasing more than 45 in the powerplay is a risk, because the ball does not come onto the bat. Sylhet offers bounce but a slow outfield; there, hitting along the ground earns runs and hitting in the air earns catches. My ER6 model is incomplete without a venue variable — that is its largest limitation.

Conceding the gap between correlation and cause

Now the uncomfortable part. Every number above rests on one season, a handful of venues, and limited tracking data. 2,094 deliveries sounds large, but each innings contributes six overs — a sample of 349 powerplay innings. At that size, part of the difference is simply variance. I pre-registered the hypothesis and reported base rates before telling the story. No model can explain why a batter chose the slog sweep; it can only report what that decision produced last time. xG cannot describe form, illness, or the conversation with the physio. Data will never occupy that space.

Empty stadiums taught me that home advantage is a variable, not a law. The same holds in cricket. The belief that a fifty-run powerplay will stay a fifty for eighteen months is the largest trap available.

One more thing belongs here, and it never appears on a data table. I tag the first four or five innings of any player returning from injury separately. I do not measure their runs, or even their strike rate — I measure how often they shorten their footwork inside the first ten balls. The number climbs. The recovery line is not always straight, and the side that understands this and bats them lower scores more.

What I will watch across the next four matches

From this Rajshahi desk, one request: for the next four matches, log not the powerplay score but the dot-ball percentage and the count of intended boundaries. I will take the bet that sides with high powerplay scores and high dot-ball rates will fall below fifty twice in seven matches. A model does not rush to judgement; it keeps accounts, and accounts do not lie — they simply wait.