Rawalpindi's 26/6 and the Number Behind Bangladesh's Real Win

It was three in the morning at my home in Rangpur, and I was sitting in front of the television. Rawalpindi, first Test. Pakistan had declared on 448/6, and Bangladesh had just lost six wickets for 26 runs. The number that flashed through my head was not 26. It was 196. Mushfiqur Rahim made 191, Mehidy Hasan Miraz 77. Their seventh-wicket stand of 196 runs is the highest for that wicket in Bangladesh's Test history. The innings closed on 565. Pakistan folded for 146 in their second innings, and Bangladesh won by 10 wickets. In the second Test, which began on August 30, Litton Das made 138, Mehidy Hasan Miraz turned the match with the ball, and Bangladesh won by six wickets to take the series 2–0. It was their first series win over Pakistan on Pakistani soil. That night I did not write about 2–0 in my notebook. I wrote about 26/6. Because 26/6 is the oldest and most familiar pattern in Bangladesh's Test batting. The 196 that followed is the largest exception to that pattern. The job of data is not to suppress emotion; it is to learn to recognise exceptions. When I started a Bengali-language data newsletter called Expected Goal from Rangpur in 2026, I had no expensive database. I had a notebook, an old laptop, and match descriptions I had heard from local coaches. I built Expected Goal in Rangpur, and the numbers started praying back — they began talking to each other, but only once I started placing them inside the same structure again and again. That is the structure I still look for in cricket. Domestic cricket in Bangladesh has thin data. There is no ball-by-ball record of the National Cricket League, no fielding maps, no log of dropped catches, no continuity in spell lengths. Where data is missing, one thing remains possible: place what exists inside a fixed framework. In Test cricket I track four things. The innings floor — the 25th percentile of a team's innings totals. The top-order collapse rate — what share of innings lose the top six under 100 runs. Partnership distribution — which wicket pairs produce how much. And bowler workload, measured through spell length. The first of those is the least discussed and decides the most. Test matches are not won with centuries; they are saved with a floor. Bangladesh's Test floor sat around 150 to 180 runs for years. One innings in four fell below 150. A team with a floor of 150 can never squeeze a side that makes 450 on a flat pitch, because it cannot hold its own innings together. In Rawalpindi Bangladesh made 565, and 262 in the second Test. Both sit well above the floor. The series was about the floor, not the ceiling. How the floor rose is the real question. The most visible answer is Mushfiqur Rahim. Making 191 in Test cricket at 37 is not just an innings; it is a template. On that Rawalpindi surface he did what Bangladesh's batsmen had not been doing: he left the ball, he waited, and he kept the scoreboard moving before reverse swing arrived. His strike rate per session was not dazzling. That was the point. When the top order has fallen to 26/6, the first job of the man walking in at seven is not to score; it is to survive the session. The more important answer is Mehidy Hasan Miraz, and this is the biggest structural signal of the series. A spinner who bats at seven plays two distinct roles in the same match. Cricket usually prices that kind of player in two separate numbers — a batting average and a bowling average. That pricing is wrong, because the two roles are correlated. Bowl more overs and the batting loses edge; bat longer and the spell loses bite. When I build a combined expected value for a player like Miraz, the arithmetic runs like this: his expected runs in an innings, plus his expected wicket impact — how far the opposition floor is pushed down — minus the gap created in the bowling unit by the extra overs he has to send down. In the second Test both sides of that equation worked. That was not coincidence. It was role design. Litton Das's 138 is a different case, and there is a trap here I try to avoid. When a batsman goes a long time without a century, we say he has lost form. But a Test century is a low-frequency event. For an aggressive batsman like Litton, the gap between centuries is naturally wide, because he has more routes to getting out. Going a long time without a hundred does not mean the skill has declined; often it means variance has not yet fallen his way. Litton's 138, to me, is not form returning. It is variance normalising. The difference is not small — one reading helps you pick a future team, the other helps you tell a story about the past. The least discussed and, I think, most future-relevant piece is Nahid Rana. Bangladesh's Test bowling unit ran on the same economy for years — pressure through accuracy, mistakes extracted through slow-medium and spin. Nahid Rana's pace changes that economy. A 145kph ball is not just extra speed; it is a recalculation of slip, gully, leg slip and short leg. On a pitch where a fast bowler is genuinely fast, a captain can create more pressure with fewer fielders. That is a mathematical advantage Bangladesh did not previously hold. The bowling-unit arithmetic deserves more detail. Bangladesh fielding three seamers across Pakistan's two innings was not only a selection call; it was a budget call. Trading a spinner for a seamer creates two separate risks: the over rate drops, and legs get heavy in the final session of an innings. What worked in Rawalpindi was the shape of Pakistan's batting line-up, where a large part of the top order is not comfortable against pace. The decision was not made with good bowlers alone; it was made with a map of the opponent's weakness. That distinction is something we underuse in domestic cricket. Now the question I force myself to ask every time: was 2–0 a skill breakthrough, or a compression of variance? I lean towards the second. The reasons are clear. Both Tests were played in Rawalpindi, on the same kind of flat pitch, where batting is easy and avoiding the follow-on is easy. Pakistan were passing through a long-running institutional instability that showed up in selection and match planning. And Bangladesh's two biggest innings came from their two most experienced batsmen, both in the final phase of their careers. What the process is telling us, then, is that Bangladesh's floor has risen while the ceiling has barely moved. A rising floor matters enormously in Test cricket, because the floor is what keeps you in the match. But if the main driver of that floor is two veteran batsmen, the obvious question follows: where does the floor sit once they are gone? At the 2026 World Cup in Russia I built a pressing-resistance model for Croatia for a London syndicate. In the group stage Croatia were allowing only 8.3 passes per defensive action, and Luka Modrić covered 72.3 kilometres across seven matches. The model said the path to the final was real. — Root: 2026 Croatia. Croatia reached the final, lost the final, but the process was proven, and that process held for years afterwards. I apply one condition when I import the Croatia lesson into Bangladesh. The Croatia case works because a small population, a limited domestic league and heavy talent export combine to create a specific kind of tournament variance. In Bangladesh's case the parallel is only half true. The population is enormous and the cricket pool is not small; what is limited is an organised pipeline and verifiable data. So the line worth taking from Croatia is that process reaches finals, not talent. The line not worth taking is small country, therefore small expectations — that framework does not apply here. In 2026, when stadiums emptied, I pulled data from 83 Bundesliga matches. Home advantage fell from 0.42 goals to 0.11 goals per game, and the home win rate dropped from 43 per cent to 33 per cent. In 2026, the empty stadium became a variable no one had trained for. That experience gave me a habit: one variable at a time. I learned to treat silence in the stands as a coefficient, not a backdrop. In Rawalpindi, that single variable was the pitch and the state of the opponent. The pitch was flat, the opponent was unstable, and Bangladesh touched the upper edge of its own floor. When those three align, the result is 2–0. The question is what happens when one of them is removed. The transfer market makes this mistake every day. After the 2026 World Cup, Enzo Fernández was priced on a single tournament sample, and Chelsea paid £106.8 million. The decision was not wrong, but it was an outcome bet, not a process bet. In domestic cricket we make exactly the opposite error: a young player who performs in one tournament is left ignored for five years, because we do not hold a verifiable record of his work. The disease is the same in both cases — decisions are being made exactly where the data is thinnest. In franchise cricket the same disease takes another form. Global leagues borrow bowlers that Bangladesh has built — injury replacements, short-term contracts, minimal responsibility. The best week of a bowler developed here over seven years gets sold to somebody else's play-off push. The club or board that built him holds no long-term return. In that arrangement, the smaller system always supplies half-finished products, and somebody else collects the benefit. In the next cycle, Bangladesh's real Test examination is not Pakistan. It is their own soil. Whether the floor holds on spin-friendly home pitches is the true measure. I am watching two numbers. The first is Nahid Rana's overs bowled per Test. A 145kph bowler's body keeps separate accounts, and Bangladesh currently has no direct replacement in the pipeline. If his workload rises by more than 20 per cent over the next twelve months, the benefits of this series win will be spent quickly. The second number is duller and more important: how many balls the No. 7 batsmen face per innings in the National Cricket League. Because what saved the match in Rawalpindi was the patience of a man batting at seven, and that quality is recorded nowhere in our domestic logs. A quality that is not measured does not survive in the team either. So the question is not about 2–0. The question is which number we learned from it, and which number has already turned into a story.