The Testimony of a Null Input: Cricket's Data Audit Trail and the Quiet Work of Blockchain

Last Tuesday, at twenty minutes to midnight, I posted a screenshot in the Rajshahi xG Circle Facebook group. The screen held a table — eight columns, twenty-one rows, and in every cell the same sentence: insufficient information. For four minutes nobody commented. Then Saiful bhai wrote: brother, the table is completely empty, what did you build this with? I answered that the empty table was the subject.

I did not sleep well that night. What had happened was not the story of a blank article. An analysis pipeline had run, passed through two stages, and then admitted its own incapacity by handing back a table whose every cell said it did not have enough to go on. On paper that is a failure. In the language of evidence it is a confession, and a confession is usually worth more than a failure.

My group knows I do not run from empty spaces. When I started it in 2026 with forty-three members, the rule was already the same. That season Cristiano Ronaldo scored twelve Champions League goals from an expected-goals figure of 10.1 — a surplus of 1.9. The number alone said little. Three hundred comments added the human story, and I learned that numbers do not live only in tables; they live in people's mouths.

So this piece circles that empty table. The question is simple. When a system genuinely does not know something, what does it do? And what can blockchain — a technology that advertises itself as an unalterable ledger of truth — actually offer in the face of that emptiness, and what can it not?

Let me be clear about one thing first, because it is an old habit. Before the table speaks, let the sample size breathe. Today's sample is a single event: one null output. Drawing a conclusion from one event is the cardinal sin of my trade. I will not claim cricket's data systems are collapsing. I will only show where the fracture hides, and whether an audit trail can throw light into that hiding place.

Let me explain the two-stage pipeline in cricket language. Stage one breaks an article or match report into atomic information points — who, what, how many runs, which over, which decision. Stage two builds deep analysis on those points. It is the scorer's book and the commentator's interpretation. Without the book, interpretation becomes storytelling, and however beautiful the story, it is not proof.

Those information points are the bricks beneath every claim. If a sentence cannot point to one, the sentence is somebody's guess. Last Tuesday's output contained zero bricks. You cannot raise a wall with zero bricks, and anyone who pretends to is not a craftsman but a conjuror — and journalism has no room for conjurors.

I learned this in 2026, covering the Wills Cup in Dhaka for Prothom Alo, with a notebook in hand and a rush to reconcile the score before deadline. An error in that notebook meant a correction the next day. It was embarrassing, but the correction existed, because once something is printed it cannot be erased, only acknowledged.

In 2026 I became The Daily Star's Bangladesh correspondent and began covering the national side home and away. That taught me that the conditions of data production live outside the field as well. The scorer at Mirpur, the person who records fielding positions, the club registers in Rajshahi gone yellow with age — these people are the real source. From their hands the data travels to television, then to a phone screen, then to our table, and at every junction something is lost while nobody answers for it.

At the 2026 World Cup I live-posted France's pressing numbers. France beat Croatia 4-2 in the final. I shared that their PPDA was 14.3 and that N'Golo Kante covered 6.9 kilometres before being substituted in the 55th minute. Three hundred comments argued about whether Kante was overrated. That day I understood that raw statistics need a human story. I began adding a section before the numbers: what the group actually saw.

On 16 May 2026 the Bundesliga returned to empty stands. I tracked home advantage. Before lockdown the 2026-20 home win rate was 43.3 per cent. Across the first three empty-stadium rounds it fell to 33.3 per cent. Members said the game felt incomplete without a crowd. I organised Zoom watch parties for twelve of them. The data showed the game had changed; the community needed connection.

Those three experiences — the discipline of the notebook, the people at the source, and the confession of the empty stadium — form the base of this argument. One principle emerges from them. Data never arrives by itself; someone carries it. A system that cannot identify its carriers cannot verify truth, only claim it.

This is where blockchain enters, though I do not treat it as a miracle. I treat it as a ledger that records who wrote what and when, and whether the writing was later altered. In cricket terms it is the paper version of the third umpire's review: the event does not change, but proof that the event occurred survives.

Now let us examine the body of that empty table. The problem has three layers. One, the source article may have been fetched but unreadable — broken encoding, a paywall, a malformed feed. Two, the article arrived but the extraction engine could not recognise any information points. Three, everything worked and the result was lost in storage. All three produce the same output: zero.

First lesson: a data supply chain fails silently. When a fielder drops a catch in front of forty thousand people, there is a roar. When data is dropped, there is no sound at all. The pipeline quietly emits an empty output and notes that there was insufficient information. That silence is dangerous because silent failure never demands repair.

The most ordinary and most useful blockchain application sits here: preserving the raw source. I now propose that every analysis keep a raw copy of its origin alongside a hash — a digital fingerprint — stored separately. Change one character in the document and the fingerprint changes. If the fingerprint stays intact, nobody has touched the body of the source.

Let me explain a hash to a cricket fan. Photocopies of a scorecard are easy to forge. But if you build a code out of every letter and number on that card, and write the code in a separate diary, then altering a single run breaks the match. The hash is that code. The blockchain is that diary, which no single person can erase once written.

So what lives on chain? Not the score. The testimony of the score. Which source, which moment, which process produced the number. Immutability is a witness, not a judge. It can say the data existed in this exact form; it cannot say the data was true. Miss that distinction and you commit the most expensive error in sports data.

Then comes the oracle problem. Who writes the cricket score on chain? The stadium scoreboard cannot sign a transaction by itself. A person or a machine must carry the truth from the ground to the ledger. That carrier is the oracle, and if the oracle errs, the chain's immutability makes the error permanent.

DRS is the useful analogy. Ball-tracking tells us where the ball pitched, but the third umpire makes the call. Technology gives evidence; a human gives judgment. Cricket has already learned this two-tier balance — technological testimony, human verdict, and a public reason behind the decision. On-chain sports data needs the same architecture.

Second lesson: a minimum information threshold can be a contract. My proposal is simple. If stage one returns zero information points, stage two halts automatically. No human grants permission; the rule stops itself. This is the least discussed virtue of a smart contract — it does not analyse, it enforces a condition.

Why does the halt matter? Because if the pipeline keeps running after a null input, the analyst is forced to invent something, and a person forced to invent crosses the boundary of truth without noticing. In journalism that act is not reporting; it is fabrication. An automatic stop removes the temptation.

The third lesson concerns confidence tiers. I have long argued that decisions can be made on incomplete data provided the confidence level is printed beside the decision. A chain can carry that level — high, medium, low — so readers know which claim is settled and which is still soft. Hidden confidence is the safest shelter for a lie.

Consider a fan token. A club sells it, supporters buy it, the market prices it. But if the data underneath — results, player form, the club's financial health — is empty, what is the token? In 2026 empty stadiums taught me that when the crowd leaves, the numbers confess what we had avoided. With fan tokens it is the same: when the supporters leave, only code remains, not value.

Now let me raise the Kante question in a different direction. The Kante question was never about one man; it was about how we measure quiet work. Today I aim it at data workers. The scorer reconciling rows at midnight, the translator moving Bengali commentary into an English table, the archivist saving old files — none of them reach the highlights, and without them every table we own is empty.

The Bangladesh context matters here, because I was born in Australia and work in Bangladesh. That distance can trap me into treating Australian data habits as universal. Conditions of production differ. Heat, humidity, dew, handwritten club registers in Rajshahi, overs lost to rain — these demand different yardsticks.

So I localise every metric. If someone says fielding efficiency is low, I ask: at which ground, at what hour, in what light? If someone says strike rate is low, I ask: in what position did he bat? Asking those questions is a professional duty, because behind a number there is always a specific person in a specific circumstance.

And this is where my group matters. Rajshahi taught me that a circle of analysts can be a sanctuary. We began with forty-three people; there are many more now, but the method is unchanged. One raises a question, others verify, someone catches an error, and only then does a conclusion stand. That is the human oracle, and no chain replaces it.

Because the eye test and the model must sit together, or neither can see the whole match. The model supplies pattern, the eye supplies context, the chain supplies provenance. Without all three we get an audit trail but not the truth.

Now the counter-argument, because part of my identity demands it. If a blockchain is immutable and the input is garbage, the result is immortal garbage. A chain can stop forgery; it cannot stop foolishness. A wrong score written once cannot be corrected, only annotated beside.

More dangerous is a habit I would call provenance-laundering. Someone takes a weak claim, stamps a chain seal on it, and markets it as immutable proof. The audience sees the seal, not the weakness. A certificate does not improve the quality of what it certifies, just as an expensive frame does not improve a painting.

My old warning about correlation applies too. A timestamp is not a cause. A chain may record that an event happened at a certain moment; it does not prove the event happened for that reason. We routinely mistake sequence for causation, then present a tidy table as a verdict. Blockchain does not prevent that error; it lengthens it.

And a minority report, because my duty in the group is not to make everyone agree but to keep dissent alive. Perhaps that empty table was the system's most honest output. A system that can say it does not know has not broken; it has matured. The danger arrives when a system does not know and confidently invents anyway.

So let me propose a sunset clause: which habit should be retired? Mine is the habit of declaring a final verdict from a single match or a single innings. The null-input event taught us that missing information is also information. We count what is present; we rarely count what is absent.

Looking forward, I will watch three signals in the next round. First, whether a rerun of stage one returns information points — if it does, the fault lies in the source, not the process. Second, the integrity of the raw payload: encoding, paywalls, the health of the feed. Third, entity extraction, whether players, teams and leagues reappear.

Together those three will give a clear picture. If the points return, the full eight-dimension analysis can be rebuilt. If they do not, we will know the problem is deeper, and we will have to admit that we do not always know what we do not know. That admission should be blockchain's true companion, because the greatest virtue of a ledger that cannot lie is this: when it does not know, it records that it does not know.

I leave the final question to the group. If last Tuesday's empty table had been written on chain, what would we be arguing about today? Perhaps about blame, perhaps about proof. The larger question is whether we can build a cricket culture where nobody is ashamed of an empty table, and everybody asks together: where did the data go?