The Empty Scoreboard: Cricket's Most Dangerous Data Failure Is the One That Passes

Hook

The Mumbai data analyst who has worked on three episodes with me sent a file last Tuesday with one line attached — “Sir, it validated.” I opened it. The schema at the top was complete. Every field sat in its slot, every bracket closed, every comma in place. And inside there was nothing. N/A where the title should be. N/A where the source should be. The one-sentence summary blank. The core viewpoints blank. And the list of information points — the list the entire analytical framework stands on — entirely empty. Article type: “Unclassified.” Where the entities should have been, an instruction sat stranded: “identify from the information points above” — points that did not exist.

I scrolled it twice, closed it the third time, and then did what I have done for thirty years — went looking for the thing that was missing. The museum kept a scoreboard; this time the scoreboard itself was gone, and a green tick glowed in its place.

Context

Now it matters to explain what that file actually was, because this is not a match story — it is the story of the system on which cricket’s match stories now get printed. Over the last few years the spine of cricket coverage has changed. There was a time when a match report began with the reporter who sat in the stadium. Now it begins with a pipeline. Stage one — a brutal, mechanical extraction of facts from raw text. Its product is the “information point”: an atomic, verifiable fact tied to a source — a score, a date, a selection call, a price. Stage two — an eight-dimension analysis built on those points: format and match, player technique, team landscape, league and commercial ecosystem, rules and governance, risk, public narrative, industry transmission.

One guard sits ahead of all of them — format. Test, ODI, T20. That is not a stylistic preference; it is a mandatory gate. A T20 strike rate and a Test average are not the same animal; putting them on one table is not merely an error, it is a dishonesty. Each dimension likewise carries its own armour — sample size, source quality, and a confidence tag fixed to every conclusion: high, medium, low.

I am no stranger to this framework. From 2026 to 2026, 41 first-class matches, a wicketkeeper-batsman for Karnataka. In 2026, nine months after that career ended, I launched Deep Midwicket from a one-room studio in HSR Layout. Episode three, recorded two days after Karnataka’s Ranji quarterfinal exit, argued that the Ranji Trophy had become “a museum with a scoreboard” and that the real selection trial was now the IPL auction. In it I repeated a number I still use: at the February 2026 auction, Rising Pune Supergiant paid ₹14.5 crore for Ben Stokes while a Karnataka batsman with 4,000 first-class runs spent the winter unemployed. The episode drew 4,000 listens in a week and two furious calls from former teammates.

My claim then was that the auction had replaced the maidan as the selection gate. The gate has since slid further forward — inside the pipeline. We put the scoreboard in the museum, and the new scoreboard is a file no spectator will ever see.

Core

To the report that came back, I owe one fairness: it was honest. On its very first page it wrote that this input contained no analysable cricket information. It refused to produce a cricket-specific conclusion. All eight dimensions returned, one by one, “N/A — insufficient information.” And one line in it deserves reading by every board secretary in the country: “absence of signal here reflects absence of input, not absence of risk.”

That is safe degradation. The system broke, and breaking, it did not lie. It deserves credit.

But safe degradation is not the story. The story is the pass. A file that clears every automated check while being empty inside is far more dangerous than a file that is plainly broken. A broken file makes noise. It trips an alarm, calls an engineer, halts the line. An empty file slips through. The instant it exits schema-valid, every downstream system reads it as “fine” — and the next human who opens it finds a report that looks ready with air where the evidence should be.

This is where the real risk sits, and the report named it itself: narrative analysis is the most tempting channel for fabrication, because plausible cricket narratives lie scattered in abundance. Redemption stories, revenge stories, the coronation of a new star — all of it comes free. The easiest work you can hand an analyst is to build a convincing cricket narrative on top of an empty file. And that narrative will publish under the name of a real match, a real player, a real team — under a real byline.

I say this not because I am above it. I am a veteran of exactly this theft. Thirty years of watching the hype cycle makes inverting consensus feel like rigour. That is why I wrote myself a rule for the show: one uncomfortable claim per episode, defended with at least three numbers, and never a reaction episode inside twelve hours of a result. A machine has no such rule. Trained on the reflex, it can produce one in twelve seconds.

Look at how many doors this empty file left open. The format gate — the first, the most basic — never opened, because the key to it was never in the input. When format is undetermined, every metric hangs in the air: average of what, strike rate of which game, economy under which pressure. Then the qualification gate — no player’s name surfaced at all, so fixing a role became moot. No player, no team, no league, no board. And where there is no team there is no ICC ranking table; where no league, no broadcast-rights valuation; where no board, not one governance checklist item is live.

In my favourite dimension, governance, the report stated plainly that the Big Three revenue share, playing-rule controversies, anti-corruption vigilance, eligibility and selection — none of it was engaged by this input. Cronje 2026, Pakistan spot-fixing 2026, IPL 2026 — the precedents exist, but none is touched.

Now the part that stung most. The file carried the domain label cricket_asia, when the expected canonical label was Cricket. It looks like an innocent slip; its effect is direct. When a routing hint comes to sit in the place of a classification, content gradually reaches the wrong analyst — and the wrong analyst writes with full confidence too. Do not seat region above classification; region is a separate field, not an identity.

Same with the source-quality field. The instruction was “judge from the source fields” — and the source field was itself N/A. The very thing that caps the weight of every conclusion was missing. Without a source, no conclusion earns a high confidence tag. And the entities: “identify from the information points above” — an instruction with no possible route to completion, because the points are zero. This is the loop in which one stage hands the next a thing it never had, and every stage assumes someone upstream did the work.

Here the whole affair joins my real beat — the board, not the badge. The 2026 match-fixing rupture, the 2026 spot-fixing suspensions, the Lodha reforms — my beat has always been this: who signs, who sanctions, who gets paid. Apply it here. Who signed off on this empty payload? What status did it get? The answer is a generic “pass.” No distinct EMPTY_CONTENT status, no hold, no log, no red flag. Cricket learned twenty years ago that a game turning over crores has to keep receipts. Data pipelines have not learned that lesson yet.

And here is a human detail no structural indictment can clean up. The analyst who sent the file was not lazy. He did precisely what he was trained to do: he checked the machine, not the meaning. He knew the schema had validated; he could not know whether the story was true — and knowing that, honestly, was not in his job description. This is the thing I leave standing against my own framework. The indictment of the system is right, and the boy has not gone anywhere either.

The Other Side

Now I will state the consensus in its strongest form, so that knocking it over is not made to look easy. It can be argued: pipelines break every day. This is a case where the system did the correct, safe, honest thing — returned nulls, invented nothing. It is a QA document, not a crisis. Making it breaking news is exactly the exaggeration I accuse others of. The objection is fair, and I accept it fully.

So where could I be wrong? Two places.

First — the rate. I do not have the number for how often such empty files arrive and how often they slide through silently. If it happens once a week and is caught every time, this is noise, not a crisis, and I am turning housekeeping into breaking news. That is the largest hole in my argument, and I will not hide it.

Second — perhaps the file was not empty at all, but non-textual. A video package, an audio clip, a photo gallery — content with no text body. In that case the pipeline is not broken; it was sent through the wrong door. Not a missing room, a misdirected address.

So I write my own refutation for you to use: I want the count of empty files that received a generic “pass” instead of EMPTY_CONTENT. If that count is zero, I am wrong and I will say so.

Takeaway

The fix is not clever, and should not be. A minimum-content gate before dispatch: at least one information point, a non-null title, a resolved format tag. A distinct status on every empty file — EMPTY_CONTENT — not a generic pass. And a public count of it.

My prediction: within the next broadcast-rights cycle, cricket’s data vendors will face the audit cricket’s boards faced after 2026 and again after 2026 — because a deal worth thousands of crores cannot rest on a file nobody read. The hot take is the surface; the pattern is the story. And the pattern here is a green tick sitting where a scoreboard should be.