No ODI fifty in ten years — Bangladesh's selection call indicts its own benchmark, not just Sohan

The fastest-moving number in Bangladesh cricket this week is a zero. No ODI half-century in ten years — pasted next to Nurul Hasan Sohan's name and presented as if it were the verdict itself. I read it as a question, not a verdict, because a fifty count means nothing until you publish the innings count, the batting position, and the era of pitches those innings came on. None of those three numbers appears anywhere in the report. A statistic that travels without its provenance is not analysis; it is a broadcast graphic.

That gap is the actual story. The sentence used to drop Sohan also exposes the foundation it stands on. The chief selector's argument, in sequence: the top order loses two or three wickets inside the first ten overs, which leaves the middle order permanently under pressure; when the opening pair does give a start, the team reaches 260 to 280; therefore the problem sits at the top, not in the middle; and Sohan, who has no ODI fifty in a decade of white-ball cricket and failed against Zimbabwe in the tri-nation event while playing in Liton Das's place, is the casualty.

The weakest link here is not the one drawing all the attention. It is the 260–280 benchmark. Across fifty overs that is 5.2 to 5.6 runs per over. In the current ODI cycle, at most venues, that is a sub-par first-innings total and chasing sides overhaul it routinely. So the selector is telling us, without meaning to, that when Bangladesh's opening pair succeeds, the team still arrives at a total designed to be chased. If a successful start produces 260–280, the defect is not simply the powerplay wicket cluster — it is tempo and intent through the middle overs, which the selector never addressed. That is a team-strategy disclosure smuggled out in a conversation about one player.

The spreadsheet started as a cricket scorecard and became a football argument here — meaning a model's conclusion was borrowed without its assumptions. A 260–280 template belongs to slow, turning, home conditions where spin strangles the middle overs. Treating that template as a universal standard, then evaluating batters against it, is metric imperialism in its plainest form. I did not port the words; I ported the assumptions hiding inside them.

Now the Sohan file. Hold two statements together. One: no ODI fifty in ten years. Two: the same selector concedes that Sohan got few opportunities and that expectations ran higher precisely because he is experienced. Those two sentences, side by side, tell you the sample is small and the demanded attribute — consistency — is structurally unavailable in that role. A keeper-batter who comes in for Liton Das receives block selections, not runs of matches, and consistency is the one thing block selection cannot manufacture. Punishing a player for failing a test he was never allowed to sit is not data use; it is the appearance of data use.

This is a role-redundancy decision dressed as a form decision. Bangladesh carries a surplus of wicketkeeper-batters and a scarcity of dependable openers. Three or four names compete for one slot while the top two positions stay unsettled. Sohan lost to positional redundancy, not to a talent ranking. That distinction should have been stated publicly, because it locates the fault in the supply pipeline rather than in a player's game.

The sample question is straightforward. One failed innings against Zimbabwe in a tri-nation fixture used to close an ODI career is a decision taken at n = 1, even if the decision turns out right. And note the inversion: Zimbabwe is historically Bangladesh's most favourable matchup, which makes failure there a warning rather than a measurement. The selector used the most forgiving fixture on the calendar as evidence. My formulation is this — the outcome may be correct, the reasoning is unverified.

Here is what the debate skipped entirely. If Sohan is a lower-middle-order keeper-hitter, and he is, the correct evaluation set is runs added in the final ten overs, strike rate in that phase, and keeping efficiency. Not one of those three numbers entered the public argument. On the age curve, ten-plus years of white-ball cricket places him past the batting peak band of roughly 27 to 33, so the conclusion is probably directionally right while the argument behind it would not survive scrutiny.

Then there is the claim nobody is touching: the middle order is not at fault. State it in its strongest form. Bangladesh's middle order is genuinely doing rescue work, repeatedly cleaning up after two or three wickets inside the powerplay, so blaming it misreads the situation. That is the fast version of the argument and it is partly true. But the same middle order is where Bangladesh's most experienced, most heavily contracted batters sit. Exonerating the senior bloc while sanctioning a fringe keeper is an accountability allocation, not a technical finding.

The Yasir Rabbi thread gives the contradiction away. If the middle order is fine, why is a T20 middle-order batter being floated for future ODI inclusion? The selector's non-committal phrasing signals the panel has not resolved its own question. The format transfer is not clean either: T20 middle-order batting rewards immediate strike rate, while ODI middle-order batting in overs 15 to 40 rewards strike rotation against spin and innings construction. Those are different skills, and any table that treats them as one turns data into decoration.

Years of watching Bangladesh's innings from the scorecard up keep teaching me the same thing: powerplay wickets are a symptom, not a cause. Every top order loses two or three early wickets sometimes. The question is why it recurs here — technique against quality pace with the new ball, constant batting-order churn, or intent. The report answers none of the three. A diagnosis that repeats the number rather than explaining it is a press conference, not a diagnosis.

One omission runs through the entire discussion: why was Liton Das unavailable? Injury, rest, form, personal reasons — that single fact changes the shape of the story. If Liton was being workload-managed, Sohan's opportunity was a one-match cameo, and judging an ODI future off a cameo weakens the case further. The absence of that fact is the biggest hole in the piece, and it belongs to the people asking the questions, not only to the man answering them.

Every clean dashboard begins with a dirty question nobody wanted to ask. Here it is: is Bangladesh's opening crisis a selection-balance problem or a supply problem? Domestic fifty-over cricket has not produced openers who can survive a new-ball spell and hold modern tempo. Until supply improves, deciding who bats where is furniture rearrangement. And I will say plainly that on this point the evidence does not contradict the consensus: the top order really is the fault line. When the data agrees with the commentary box, I say so. My objection is to the benchmark, not the diagnosis.

Watch three things in the next series. First, does the panel present the same statistic again, this time with the innings count attached? If yes, that is a genuine move toward transparency. Second, does the average powerplay wickets lost fall in the next bilateral assignment, especially against quality pace? If it does not, Sohan was never the problem. Third, and most importantly, is the 260–280 target publicly revised?

My verdict: dropping Sohan is probably the right call on age curve and limited opportunity, but it was justified with a counting exercise rather than an understanding of the numbers. A team whose successful template is 260–280 does not need new batters; it needs a new benchmark. Bangladesh needs an opener who survives the tenth over and keeps scoring after it far more than it needs another wicketkeeper. Sohan was never part of that equation — and that is the truth the statement worked to obscure.

Disclosure: this piece rests on public scorecards and quoted selection comments. I do not have independent verification of Sohan's innings-by-position or format splits, so the 'no fifty in ten years' claim should be read as unverified. I have no access to the BCB's internal selection data, and nothing here should be read as if I did.