The Blank File and the Filled Page: An Evidence Filter for Cricket's Transfer-Season Noise

The file looked entirely ordinary. Inside, there was exactly one usable line: domain label, cricket_asia. Title N/A. Source N/A. Publication date N/A. The list of information points, empty. And the field that should have named the teams, players and boards was itself instructing me: "identify from the information points above." The points from which to identify did not exist. The system returned a status: successful.

Sitting in front of that screen at a quarter past eleven at night in Kolkata, the first thing that struck me was not technical. It was that this kind of blank file is not built only on servers. It is built on every agent's phone call where the name is fixed and the numbers are found afterwards. It is built on every breaking-news banner where the headline is decided first and the fact is decided later. In cricket's transfer coverage the most expensive error is not a lie. The most expensive error is a void that has already been stamped "successful."

I have been watching cricket coverage for the better part of four decades, and one thing has become clear: the volume of noise has risen, not the volume of signal. The transfer window is the cleanest example. Fifty stories arrive, two are true, and neither of the two tells you the structure of the deal — the length, the release clause, the escalation, the agent's cut. Only the fee is reported, because the fee is the easiest number and the least informative one.

Structure first, fee later. That sentence sits on the first page of my notebook, and it came from the day I first opened a ball-by-ball log and found that the scorecard and the deliveries tell two different stories.

Format is the first filter nobody applies

Every cricket judgment must begin with one question: which format is this? Test, ODI and T20 numbers cannot be stacked on top of each other, because risk is priced differently in each. In Test cricket the greatest skill is protecting a wicket; in T20 it is spending one. A batter with a strike rate of 95 in the 50-over game is doing an entirely different job at 145 in T20 — and that same 145 is low for a finisher and excellent for an anchor.

Without role, the number is meaningless. That is my second condition. When a player's average or economy is quoted, the prior question is which slot he occupied. An opener's 40 and a number six's 40 are not the same 40. A fast bowler who bowls the powerplay and one who bowls the death overs cannot have their economy rates compared, any more than a bat can be compared to a ball.

The third condition is circumstance — home or away, the pitch, dew, and the toss. In India, dew in the second innings rewrites an entire bowling plan, and analysis that does not count dew is half an analysis. DLS (Duckworth-Lewis-Stern) revises a target after rain, and numbers built against a revised target belong in a different drawer from the original match. DRS (Decision Review System) runs on ball-tracking and UltraEdge, and in a match where two or three decisions were overturned, the accounting between luck and skill has to be redone.

All of which means one thing: if a report carries no format, that report can travel in any direction. And that was the real problem with the blank file — it said "cricket_asia," which is a label for a geography, not a game. From that label you cannot tell whether the subject was a Test match or a franchise auction.

I stopped writing match reports and started drawing failure maps

June 18, 2026, The Oval. Pakistan made 338 for 4 in the Champions Trophy final — Fakhar Zaman 114 off 106 balls. India were bowled out for 158 in 30.3 overs, a 180-run defeat. I filed the match report in forty minutes. At midnight I pulled the ball-by-ball log and charted where the fielders actually stood during Fakhar's first thirty deliveries: point and third man pushed square, no catching midwicket, mid-on deep. The picture that emerged was not a 180-run loss. It was a field setting that had vacated three catching positions inside the first ten overs and forced the bowler onto a line where the claim of control was false.

What I wrote that night was not a match report. It was a failure map. I stopped writing match reports and started drawing failure maps — I have repeated that sentence for seventeen years, and each year it proves more correct. A match report tells you what happened. A failure map tells you where the decision was made, and that place is usually not in the main column of the scorecard.

The match report died the moment we counted all 180 runs. Because 180 is not one event. It is three innings stapled together: the first ten overs, the next fifteen, the last ten. Each carried a different price of risk and each had a different field. Until you un-staple them, you cannot understand 180 — you can only pronounce it.

Name the metric first, defend it later

The ratio was born from a spreadsheet and a suspicion. The suspicion was that half the runs scored in a powerplay are a reward for empty space, not for skill. So I built something simple: the ratio of deliveries a batter left or defended in the first six overs to the deliveries he attacked. But the metric was not the point. The point was that I named it before the match began and wrote its kill-switch alongside it: if this ratio showed no clear relationship to winning across the next ten matches, I would publish it anyway — same place, same prominence as if it had worked.

Falling in love with your own metric is the most common disease in this trade. Once you have named a ratio, it becomes your child, and you stop asking the data what is true and start asking it for an alibi. I have made that mistake. I built a "pressure index" that combined runs conceded at the death, the value of wickets taken, and the state of the game. After four matches it looked beautiful; in the fifth it collapsed. I published it, and publishing it was correct, because the alternative was quietly sliding it five overs forward until it started to look right.

The same discipline applies to transfer stories. Before you accept a deal, decide what you will count as proof. A board's official announcement? An agent's briefing? A named journalist's report with two independent sources? Or the account that breaks forty trades a week and never files a correction? Without that decision you will weigh every story equally, and then conclude that the world is unreliable, when in fact your scale is broken.

The evidence ladder

At the top sits the official announcement — the contract, the retention list, the final auction ledger. Below it, the named journalist's report that describes the nature of the sourcing even when it cannot name it. Below that, general media, which is usually derived from the top two tiers and adds nothing. And at the bottom, the aggregator, whose business model is the number itself — clicks, views, reposts.

Once you know the ladder, the most useful question is never "is this true." It is: whose interest does this story serve right now? A player's contract is expiring, so his agent needs to display demand. A franchise has an empty overseas slot, so it needs the market to believe it is bidding. A board wants to hold a player back, so the NOC (No Objection Certificate) story leaks before the announcement.

The RTM card has changed the arithmetic of this entire economy. When the Right to Match card returned to the IPL auction, a player's market price and his value to his previous franchise became two different numbers. A team now has to know which player the opposing side will match, because the whole bidding structure rests on it. Stories that say only that "three teams are watching a player," with no mention of RTM, omit the most important part of the auction's politics.

That is why every major transfer story in my notebook carries three columns: fee, duration, structure. The fee makes the headline. The duration decides the team's budget for years. And the structure — release clause, season-by-season escalation, injury payment terms, the agent's percentage — tells you who actually won the deal. Often the youngest player in the transaction is the biggest winner, because his duration is long and his fee looks small.

The transmission chain, and where the truth dies in it

A rumour completes its journey in five stops. First, an agent tells a journalist that three teams are asking. Second, the story is printed and circulates in three places on social media that day. Third, the franchise's media department sees the market stirring and raises its estimate, because it believes there is competition. Fourth, that raised offer becomes the story — "the price has gone up" — and now the rumour has a piece of evidence, because two teams really did ask. Fifth, an analyst sits down and writes that "the market has recognised this player's value."

Nowhere in that chain did the truth die. It changed at the third stop. The story was false at the start and then made itself true, because it changed the behaviour of the person it was lying about. Media studies calls this reflexivity; cricket calls it "the market is hot" and prints it.

A system reveals itself in the passes nobody applauds. A franchise reveals itself in the deals nobody headlines — the spend on a third seamer, the wicketkeeping understudy, the three-year contract for a domestic spinner. When a team is called "balanced" after an auction, the question that matters is how full its wage bill is and how much room is left for retentions next season. That number never makes a headline, and it is the number that decides the next three years.

Where the Ranji stories go

The largest cost of this structure is the one that never appears in a headline. Domestic coverage thins before the knockouts, and that thinning is not a conspiracy — it is the market working. When one franchise-league match brings more viewers and more advertising than an entire Ranji season, the newsroom budget follows. The result is that we have T20 numbers to three decimal places and the same three first-class ratios we had fifty years ago.

That is not an absence of information; it is an unequal distribution of it. And that inequality produces the whole line of thinking that calls domestic cricket "weak." It is not weak. It is invisible. You cannot say a thing you are not watching is bad. This is the same error that produces a confident field map from a Ranji debutant's twelve balls. I have made that error and I have published it, and afterwards I set myself a rule: impose a minimum sample floor before any zone-based claim. Below the floor, write the observation as a question, not a finding.

The angle we all miss

We assume the problem with cricket coverage is false stories. It is not. False stories are a traffic problem, and traffic problems are solved by technology. The real problem is that verification is work without glory. Nobody wants their name on the report that says: "not one source for this could be found." And so when a blank file arrives, the pressure to fill it is far greater than the pressure to trim it.

The second thing this trade must accept: more data has not improved cricket coverage, because the data that grew is aggregate data. We have ten times the scorecard and the same amount of construction detail. That is my oldest complaint — strategy is born in the moment you separate the three innings inside one innings, and that separation never gets printed. You will find lists of hundreds of centuries; you will not find which fifteen-minute passage changed the field, and why.

And the gravest risk is the one nobody notices: the blank file marked "successful." If the system returns "nothing found," the next person checks. If it returns "all clear" with nothing inside, the next person assumes there is no risk. Cricket calls this the difference between "no information" and "no danger," and it matters as much as the difference between ball and bat. A player with no fitness data should not be treated as fit. He should be treated as a player whose fitness has not been checked.

What to watch next

The next time a major signing breaks, look at two things before the fee: the date on the story, and both squads' retention lists. The date, because cricket information decays in hours — a toss report by morning, an injury by evening, a rule change over months. The retention list, because the real ground is prepared there, not on the auction stage.

A contract whose structure can be sold needs no explaining — it is the model. A story that carries only a fee is the older game.