From BPL Powerplay to Death Overs: Coding the Mirpur Pitch in a Five-Lane Grid

Mirpur, Sher-e-Bangla National Cricket Stadium, the 17th over. A left-arm spinner at the top of his mark. A left-handed batter at the crease, set, 41 off 34. The field is worth memorising: deep long-off inside the rope, deep midwicket on the boundary, nobody at cover, a point fielder seven yards in from the circle. The ball lands roughly seven metres from the stumps, just outside Lane 3, with eleven centimetres more bounce than the surface had offered all evening. The batter closes his stance, the ball comes in, takes the inside edge, runs fine to square leg. One run.

I rewatched that delivery eight times. The television commentary at the time said 'excellent spell', 'building pressure'. I would say something else. The line was Lane 3. The length was seven metres. The bounce was eleven centimetres above the mean. The front foot had crossed the crease line by six centimetres. The batter did not misplay the ball. The pitch did half the work before the ball reached him. That difference is the only reason I write: not who won, but which variable won it.

I sat in the corner of my room and re-ran fourteen matches. I tagged every delivery into five vertical lanes, every reception into a zone number, every fielder's position in yards from the boundary. Two thousand three hundred and seventeen deliveries, forty-one spells, twenty-six field-setting frames. The eye said 'excellent spell'. The code said something else: four of the six successful balls in that spell landed on Lane 3 at seven metres, the other two were slower balls on Lane 4. Not the bowler's magic. The consistency of length. I coded the Bangladesh Premier League before I trusted the eye test, because in this league the eye is deceived more than anywhere else.

Context: small grounds, small samples, big decisions

The first BPL season began in February 2026 under the Bangladesh Cricket Board. Since then, seven or eight franchises each year, and a handful of venues. That narrow venue pool turns the league into a coded laboratory. A bowler who works in Dhaka knows that surface is black soil, somewhat two-paced; a bowler working in Chattogram or Sylhet keeps a separate page. I regularly put the same bowler's Mirpur spell and Chattogram spell side by side. Usage is almost identical. Outcomes are not. The variable is not the bowler.

My grid is simple, and I start every article with it. Five lanes along the length of the pitch, and five depth zones measured from the stumps.

Lane 1 is well outside off, almost an eighth stump — a ball you can cut but cannot drive. Lane 2 is the off-stump corridor, the line that forces a batter to play without rewarding him. Lane 3 is over the stumps, top-of-off to top-of-leg. Lane 4 runs from leg stump to the pads — for a left-hander this is a death trap, because the ball enters behind his angle. Lane 5 is down the leg side.

Depth: Zone 1 is two metres from the stumps, the yorker. Zone 2 is four metres. Zone 3 is six metres, what we call hard length. Zone 4 is eight metres, short of a good length. Zone 5 is ten metres and beyond.

Twenty-five cells make my map. Which lane and which depth a ball lands in decides more than the type of ball, because the pitch touches it twice — once at the pad, once at the bat.

One honest admission here, because without it the rest of the arithmetic is meaningless. A BPL season is twelve to fourteen matches, seven teams. A given bowler meets a given batter three or four times in a season — eighteen to twenty-four balls. Calling that a 'matchup' is not statistics. It is poetry. I am happy to write poetry, but I am not willing to pass it off as proof. My notebook therefore has two separate columns: 'pattern' and 'evidence'. Most of the claims in this piece sit in the first column.

Dew is another massive variable, and at Mirpur it is brutal. In the March-April evenings, after the ball sweats, seamers lose grip and spinners lose turn. In my coding, left-arm spin average break was 2.9 degrees in the first innings and 1.7 in the second. So I keep second-innings spin data in a separate file. Average the two innings together and you invent a bowler who never existed. When the game shut down in 2026, my only work in Barishal was hunting these errors in my own notebooks.

Core: four phases, one door

One. Powerplay geometry: runs come from gaps, not shots

In the first six overs the fielding law does more work than the bowling. Only two fielders may stand outside the circle, and the split between inside and outside the thirty-yard ring creates a trust. Across those fourteen matches I divided every powerplay run by shot type. The largest share came from balls pushed into the ring just inside the boundary — into the dead zone where no fielder stands.

In my coding, first-innings powerplay scoring was 7.4 an over; second innings, 8.6. In Dhaka today that gap is not form. It is the fact that in the second innings the target is known.

Here the first adjustment appears. Batting first, you can wait. Batting second, you must track. But not every side takes the same track, and that is where a decision lives. The batter who leaves a Lane 2 ball alone is producing the most valuable commodity in the powerplay: a dot ball. That dot becomes eight runs two overs later, because it forces the bowler to pull his length back, and the pulled-back Lane 3 ball can be hit.

In my sample the cheapest shot for a left-hander is the Lane 1 cut. The most expensive is the inside-out drive over cover — from Lane 2, front foot thrown. That second shot demands waiting for bounce, and bounce is at its highest in the first six overs. The clock itself is a weapon.

Two. The door in the middle overs

Overs seven to fifteen. Five fielders in the ring. Here the game is moved by patience, by closing gaps. I have watched enough to say the fielders in this phase do not shut a window. They open a door.

Take the left-arm spinner's release angle. He bowls from the right corner of the crease, left arm extended, so at release the line of his body is nearly parallel to the batter's. The ball travels from outside off towards the stumps at about nine degrees. If it lands near Lane 4 at six metres, the seam keeps its direction and, after pitching, the ball drifts across into the left-hander's pads. The left-hander, meanwhile, is closed from off stump; his bat face opens outside the line and the ball goes behind his angle.

The inside-angle channel of a left-hand batter is not a fashion; it is a door. The door stays open exactly as long as the bowler holds Lane 4 at six metres. Push the length to seven or eight metres and the door shuts — the batter goes back and cuts.

My coding shows that in middle-over spin spells, a left-arm bowler sends roughly nineteen per cent of deliveries against left-handers into Lane 4; the rest go to Lane 3 or Lane 2. Lane 3 at six metres can be pushed for ones. Lane 4 at six metres cannot be played; it can only be survived. Two feet separate those two lanes, and those two feet are the whole of the middle overs.

The field setting must sit with it. Deep midwicket on the rope, long-on inside — that is the natural set. But where the gap is left between midwicket and long-on, the batter sweeps or late cuts, and both shots are easiest against exactly the Lane 4 ball. That gap is a shout: the bowler knows the door is open, so he does not want the batter coming in. In my data from Mirpur, in ten overs where the midwicket-long-on gap was left open, more than a quarter of the runs conceded came from the sweep. To sweep you must get onto the pad line. The pad line is the door.

One more thing worth remembering: a fielder's position sells the plan. A fielder standing seven yards outside the wicket means the spinner is not bowling a dot ball; he is inviting the batter to attack. One camera frame showed six fielders forming a triangle between cover and mid-off. That is a shout, not a reassurance.

Three. The death-over myth

Overs sixteen to twenty. To me this is the most misunderstood phase, because memory floats up most freely here.

I coded those forty-one spells into a dry list: what did deliveries of ninety centimetres or less concede, and what did deliveries of 2.5 to 4 metres — short of a good length, plus the slower ball — concede. The result went against my own assumption. The one-and-a-half to three metre length — not the yorker but slightly longer, the far edge of Zone 1 — produced the lowest strike rate, and the attempted yorkers produced the most square-leg and fine-leg boundaries.

The reason is geometric. To bowl a yorker, a bowler must release the ball roughly two metres in front of the crease, and a fraction of an error turns it into a full toss, which on Lane 3 is the worst ball in cricket. A one-and-a-half to three metre ball that goes wrong becomes a low full toss, which jams under the bat. The price of a mistake is far lower. In the death overs a bowler can survive inconsistency, provided he does not err on the most expensive line.

Something else needs saying boldly about Mirpur. In the second innings, dew plus scoreboard pressure push the two-outside-the-circle setting to Lane 5. In my coding, the fielder inside fine leg averaged 3.1 saves per death over in the second innings, against 2.2 in the first. Two yards of repositioning can save two runs an over. It looks like luck. It is setting.

Four. Cross-code translation

The way a 4-2-3-1 and the left half-space describe attacking shape in football, cricket's field coordinates describe attacking shape too. In the left half-space one player pulls two opponents across, freeing a one-on-one on the right. T20 spin fielding does the same job: the attacking side over-loads the off-side square with two extra fielders and thins the leg side. Off-side lines are blocked. The opportunity sits on the leg side, behind square, through the wicket.

At Barishal I use both languages in training. I tell the trainees: before you move a fielder, ask what invitation you are sending. If you do not know the answer, you have not moved a fielder. You have opened a door — for the opposition.

Contrarian: the matchup trap, and the scorecard's lie

The most numerically confident claim in this piece is the role of Lane 4 for a left-arm spinner against a left-hander. This is exactly where I must be careful. A small sample invites a second error: we see two variables correlate and skip the third.

Say a left-arm spinner has dismissed a left-hander three times. The headline writes: the matchup worked. Watch the video, please. How many of those three balls were six metres? How much was hitting the scoring field? Yorker, slower ball, carrom ball, wide — line them up and we call it a plan, but on the frame it sometimes happened at eight metres, sometimes it became a catch because the out-fielder was two yards inside. The report sees 'skill'. The code sees distance.

My central caution is this: at BPL scale, no matchup analysis can decide anything. It can only build a question for the next match. If a bowler takes a left-hander's wicket three games running, that is not a career priority for me, it is a clue. I follow the clue into the footage, then I code. I do not believe first.

One more thing. We credit the pitch only under the name of bowling quality, and nobody can see it. How much of a turning short-length ball is the bowler's fingers, and how much is the pitch's slowness? Of those forty-one spells I watched eleven in a frame-behind view — how much the ball turned off the surface after pitching. My estimate: roughly forty per cent. Two-fifths of the decision belongs to the soil, not the bowler. The studio gentleman who says 'the bowler broke him mentally' has forgotten the track that let him do it.

Takeaway: what to watch next match

I usually end with something you can write in your own notebook. Do this: take the first spin over of the middle phase at the next Mirpur match, and draw five lanes on paper. Track six balls — which lane, how many metres from the stumps. At the end of the spell, count how many reached Lane 4. If the number is under four, that bowler is surviving the batter, not dismissing him. If it is five or more and no wicket has fallen, believe me — the wicket comes next innings. Not form. The grid says so.

Mirpur's dew, its short square boundaries, a fourteen-match season: within those three I leave myself a claim. If that changes next season, I will throw out my own code. Until I see it, I will sit in the corner of my room and watch that ball eight more times, because to me those eight are eight different deliveries.