When the Scorecard Is Incomplete, Who Speaks for the Numbers

Sitting at a local match in Rajshahi last month, I noticed the scoreboard filling up with runs and wickets, but the boy recording bowling figures left one column blank—nobody had counted how many runs came in which over. After the match I asked a coach what the bowler's economy was. He laughed. "I don't need to know," he said. "I watched him bowl well."

That laugh has followed me for weeks. Because the same sentence, in slightly different clothing, I have heard in the corridors of the national selection committee, in franchise auction rooms, even in television studio panels. Only the scale changes. At a local match nobody notices the blank column. At the national level, that blank column becomes the hinge of a career—because story fills the blank space, and ownership of the story belongs to whoever has the loudest voice.

Let me state my claim up front, because I know the reader's thumb is already scrolling: a large share of Bangladesh's selection controversies is not about a lack of data, but about the narrative used to cover up incomplete data. And unless we understand where that narrative factory is, who runs it, and in whose interest it operates, we will repeat the same mistake in every debate.

Context: the data foundation we assume exists

When we talk cricket, we make a silent assumption—the numbers exist, they are reliable, and they speak the same language. Ball-by-ball Test data, List A scorecards, domestic first-class figures—all usable as evidence of equal weight. At international level this assumption largely holds. But drop below and the picture begins to fracture.

Enter Bangladesh's domestic structure and the density of information changes tier by tier. BPL matches carry per-ball tracking, data vendors, shot-angle recording. National bilateral series follow ICC-mandated protocols. But many Dhaka Premier League matches, some rounds of the National Cricket League, large parts of age-group tournaments—here you go back to paper score sheets, where the bowler's name has wickets and runs beside it, but no economy split, no phase-based breakdown, no boundary percentage.

From years of watching, I have noticed this pattern: where data is thin, decisions don't stop—only the language of decision changes. Impression replaces numbers. "He has hunger", "match-winner", "cool head under pressure"—these sentences sound harmless, but they are evaluation methods by another name, and this method has no stated error margin. Data has flaws, so it can be corrected. Where impression's flaws lie, nobody says.

This asymmetry has a concrete consequence rarely discussed. Players who get franchise leagues or televised matches have a large information base. Those outside that light have little. Yet both face the same selection glare. A bad series for a player we watch weekly registers instantly; the consistency of a player scoring runs on the domestic circuit stays invisible. This information gap is filled in two ways—either someone has gone and watched him personally, or a story has been attached to his name.

Core analysis: who fills the blank columns

Here I owe a confession. From the 2026 Germany call I learned something: confidence is the story we tell ourselves before the data arrives. Germany's midfield averaged 27.8 years, their oldest squad since 2026—that fact was in front of me, and I buried it under the confidence story. The same process runs silently through cricket, except here Germany is replaced by a domestic bowler about whom someone has already formed a story.

The process of filling blank columns runs on three levels.

First level: the recording gap. Where there is no ball-by-ball data, analysis cannot run. This is a technical problem, and it is solvable—install cameras, use scoring apps, archive match video. As a BCB advisor on digital and media affairs I have a direct interest here, and I will say progress has been made—but unevenly. Lower-tier BPL venues do not have the coverage quality that exists at Mirpur.

Second level: the comparability gap. This is the real problem, and the least discussed. Say one domestic bowler takes 40 wickets in 12 matches. Another takes 28 in 9. Who is better? The two numbers are not on the same scale, because pitches differ, opposition quality differs, fielding standards differ, even ball brands differ. But at the selection table these two numbers sit side by side, and the decision that emerges is called "data-driven". It is not data-driven, it is data abuse—and that abuse is the most dangerous because it arrives dressed as analysis, making it hard to question.

An example. Suppose a franchise auction is pricing two spinners. One has a low economy in a domestic tournament, but his matches came on spin-friendly pitches. The other concedes slightly more on uneven pitches, but in death overs. If we look only at economy, the first wins. If we look at context, the second is actually more valuable. In our leagues that context layer is often absent from the table, and its place is taken by auction-room chatter.

Third level: the narrative gap. However much data exists, one final gap remains—which question we ask determines which answer we get. Narrative fills this gap. And narrative is produced under a specific economy.

That economy needs understanding. When a story forms around a player, three parties are involved. First, the journalist—who wants a frame, because without a frame there is no story. Second, the fan—who wants a hero or a villain, because neutrality gives no emotional fuel. Third, the decision-maker—who wants an argument supporting what he already believes. When these three demands meet at the same point, a story is born and it sounds like truth.

I am myself part of this machine, and this piece is not honest unless I admit it. After the 2026 Champions Trophy I posted a 14-tweet thread showing Bangladesh were 0-for-6 in knockout matches since 2026. The thread went viral because the number was true and the frame was sharp. But looking back today, the number created a pattern that did not actually exist—because six knockout matches were played against six different opponents, in different conditions, with different squad constructions. The data was true, but the narrative placed beside it was not. And because it went viral, the narrative became the memory.

This is why I now timestamp every claim, rate my confidence, and state in writing where I could be wrong. This is not modesty, it is procedural protection—because I know my own instinct is also at risk of building a story in front of the same blank column.

The domestic numbers we do not read

Now to the real place—a map of what actually fills those blank columns.

First and most common filler: recency bias. Whatever a player did most recently gets heard most. Two centuries in the first three rounds of a domestic league and his name spreads everywhere, but an average of 22 across the middle eight rounds—nobody looks for that. It happens because memory holds only a few moments, not the full timeline. And since the full timeline is not easily available, memory becomes the evidence.

Second filler: profile confusion. Does "good strike rate" mean good in death overs, or in the powerplay? Many domestic records do not carry that split. So a batter is played in his weakest position, he fails, and that failure stands as evidence against him. In this situation I would argue that repeated failure at a misused position is driven more by a usage gap than by personal limitation. Yet we keep that as proof.

Third filler: sample-size neglect. A batter's 50-ball sample across a whole domestic season is treated as determining his capability. That number is so small it shows nothing better than noise. But in our culture there are talk shows about a five-ball strike rate, because the number is easy to grasp and easy to settle.

Fourth, and in my view the most damaging: memory-based continuity in selection committees. When a committee drops a player, that decision becomes part of board politics, and recalling him later means admitting the earlier call was wrong. In other words, data-based re-evaluation gets admitted as an administrative failure. Under this incentive structure, neutral use of data becomes impossible.

I will add something I have noticed repeatedly at matches: two people can use the same data set to reach opposite conclusions, and both can be honest. Because data does not give answers, data only changes the range of questions. If my question is "who is safest", I pick one player. If my question is "who can take the most risk", I pick another. Selection's real work is not data analysis, it is question selection. And the power to choose the question is the centre of power.

Information inequality and its politics

Now a direction some readers may find uncomfortable. I think we often frame the information-inequality question wrongly—we turn it into a progressive versus backward debate. That is not what it is.

The real thing is this: whoever cannot gather information exerts the most influence by selling story. Tracking data is expensive. Hiring professional analysts is expensive. Video archiving and tagging are expensive. A limited number of organisations invest; the rest operate on verbal reports. The result is that the less a verbal report is subject to institutional accountability, the more influence it carries. Where data exists, accountability is easiest—show it when questioned. Where only impression exists, there is no accountability, because impression has no fingerprint.

This structure has a practical consequence important in Bangladesh. Many domestic players' careers depend on one or two observers' reports, who may well be right, but whose reports have no documented source. This system is flexible on one hand and leading-prone on the other. And if that observer has a regional or club affiliation, the information system quickly becomes biased—even though everyone knows it is happening, it cannot be proven, because the very thing we call proof is absent.

I offer a sensitive proposal. If we want a single standard, we can say this: beside every verbal evaluation, write which match it came from, on what date, on what basis. This is not difficult and costs nothing. It is only a habit. And this habit alone can break a large part of the narrative economy. Because narrative survives on vagueness. Where time, place and testimony are written down, narrative does not hold.

The strongest opposing case: perhaps the data gap is a myth

I build my own strongest opposition, because beating a weak opponent is not victory—it is trickery.

The opposing argument runs like this: cricket's information infrastructure is expanding fast. Ball-tracking technology is cheaper, youth matches are recorded on phones, app-based scoring has gone open source. So the "information scarcity" I describe is becoming irrelevant. Rather, it is now used as a cheap excuse for weak decisions—saying "we have no data" means nobody demands accountability. And the opposition makes a stronger claim: bad decisions happen where data exists too. The problem is not lack of information but the capability of decision-makers. More information will not fix it; it will only make the wrapper of error smoother.

This argument genuinely hits part of my thesis, and I admit it. I will even go further: even if the data gap did not exist, my core claim survives—because my claim was never about lack of information, but about the monopoly over interpretation that narrative builds on top of information. Narrative problems are not solved by more data; they are solved by question-level accountability.

Still, in at least one place my critics are entirely right: I cannot keep using the data gap as a general-purpose blame machine unless I also guard the roof of my own house. This is why I always timestamp every claim, and ask the same of every side. If I demand clarity from others but do not press my own side, the credibility of this whole archive is merely rhetorical.

Final word: a testable prediction

At the end of all this I want to make a claim that can be verified within 12 months—and if I am proven wrong, I will admit it myself.

In the next two selection cycles, at least one player will be called up to the national side on the basis of domestic league performance where the strongest argument for selection is a number that is in fact context-free—total runs, total wickets, or a single series average. And when that selection triggers public debate, the argument presented will not be a phase split or an opposition-quality adjustment, but that number itself—and beside it a sentence that will almost always be there: "He knows how to turn it around under pressure."

I am not saying we are stupid. I am saying we are pragmatic. Telling a story takes less time than completing data, costs less, and requires fewer answers. Anyone who asks can be told "we have a different view". And that holds, as long as nobody demands accountability.

So the real question is not whether we need more data. The real question is whether we want a system where arguing for a decision means showing its source, date and context as a requirement. If we do not, then whether numbers rise or fall, the blank column will always be filled by a story. And on that story's last page will be written that laugh—the one I saw at the Rajshahi ground, and which still belongs to no one.