The Empty Ledger: In Cricket, the Record Nobody Files Speaks Loudest

A file landed on my desk last week. In the line where a headline should sit there was an empty bracket, below it a date, and after that eleven pages — all blank. A one-line note came attached: "data upload failed." A colleague on the desk said it was the most honest cricket report of the day: nothing had happened in it.

I did not throw the file away. In January 2026, in that small cabin at the Holkar Stadium in Indore, I sat in the TV umpire's chair and watched the same thing happen on a monitor. The Ranji Trophy final was on — Gujarat, captained by Parthiv Patel, chasing their first title. A review came up. One feed on the split screen was running two seconds behind. I had one angle, one button, and a protocol that said nothing about what to record when the feed is late.

The decision was given that day. The file recorded only the outcome. Those two seconds were never written down.

After that day I began reviewing every comfortable verdict — and six months later I resigned the chair. (— Root: Experience 1, the Indore TV umpire) This piece is about that empty space.

Six layers, each in somebody's hand

Cricket information reaches the record through six layers, and every layer has a door somebody can close.

The first layer is the scorer's app. The ball-by-ball feed is public within seconds, but if a connection drops mid-over, that over is rebuilt from the scorer's memory. The second layer is the third umpire's console, which logs which frame synced first, how long the process took, which angle was opened in what order, and when the soft signal was used. None of that is broadcast. The viewer gets the final graphic. The third layer is the match referee's report: code-of-conduct breaches, pitch rating, over-rate, sanctions. It is written, filed, and stays private until someone is fined. The fourth layer is the broadcast truck, where it is decided when the split screen appears and which angle leads. The fifth layer is the auction and contract sheet, published in full — as price, not as value. The sixth layer is the board's press note, which arrives when arriving is convenient.

None of those six layers ever prints "data unavailable." When a layer fails, the output is nothing — and nothing gets read as nothing happened.

Every category error in cricket commentary starts here.

The category error of format

A Test average, a T20 strike rate and the ODI middle ground are three different instruments, and swapping them is the most common error in the business. A middle-order batter striking at 180-plus in the death overs is meeting a benchmark; the same number in a Test is not a benchmark but often a warning, because it tells you he does not want to defend. The day a player's Test average starts getting "explained" by his T20 league numbers is the day analysis ends and publicity begins.

The auction sheet is the biggest factory of this error. The price paid at an auction is a player's market value, not his sporting value. A large contract does not prove a player will hold up internationally; it proves that on that day, in that room, one franchise had an empty purse and few alternatives. Any report that does not separate those two values is a prediction that will never be audited, because nobody will remember it three years later.

The model is the thing the graphic hides

Ball-tracking is not measurement. It is a projection from a model — an extrapolation forward from a handful of captured frames, with its own error band. Cricket has accepted that band, and the acceptance has a name: umpire's call. In an LBW review, when the ball is clipping the outside of the stump within half a ball's width, or the point of impact is itself marginal, the projection's uncertainty is not considered sufficient to overturn the on-field decision. Umpire's call is a design feature: the technology is declaring its uncertainty as loudly as it declares its conclusion.

The trouble is that uncertainty is the least televised part of the broadcast. On screen the ball travels as a clean, bright line and hits the stumps; no error bar appears. No model name, no margin, no clipping window. The viewer is shown measurement and not inference. That is why the argument after every contested review stops at "the decision was wrong" and never reaches the more useful question of which model made it, on what margin, from how many frames.

The soft signal: the law that admitted the uncertainty

Until a few years ago cricket carried a provision called the soft signal. Where the third umpire lacked conclusive evidence, the on-field umpire's tentative call stood. It was unpopular, because the viewer watching the graphic felt the technology knew and was refusing to say. It was removed from the playing conditions in 2026. The reasoning was that the technology was now complete.

That is a strange turn for the health of cricket's record. With the soft signal in place, the law itself was admitting it did not have enough. Without it, every decision arrives with total confidence even when a single angle is struggling behind it.

Set that against Duckworth-Lewis-Stern. Twenty overs get revised after rain and nobody calls it robbery, because the formula is published. A hidden model invites conspiracy theories and quiet fixes; a printed model invites argument about its flaws, which is the healthier outcome. Cricket could have published what it was uncertain about. It chose to delete the rule instead.

The review economy

The most useful thing my economics degree gave me is this: a review is a resource. In Tests each side has two unsuccessful reviews per innings, and they are retained when successful. That is a budget with a refund clause, and a refund clause distorts spending behaviour in every budget ever written. The captain knows it comes back if he is right, so he takes the risk; and when he is wrong, he has no insurance for the next thirty overs.

A review burned in the eighth over does not just cost a review; it changes the field for the next forty overs. A captain holding a review keeps an extra catcher at slip, because he knows that if the edge comes and the umpire misses it, he can retrieve the moment. A captain without one pushes that fielder to mid-off. None of it appears on the scorecard, and it is the cheapest, least examined strategic decision in the match.

It is why I file a protocol note before every fixture naming the three laws most likely to decide it. Twice I have refused to let a colleague on the desk cover a review-heavy match without one. That rigidity has cost me friends at work.

The small-sample deception

Three good innings are not a trend. Calling three good innings a trend is the oldest weapon of journalism that has nothing else to say. A batter who has twice cleared mid-off this month is not a changed player; he is the same player who twice received the ball in the slot. Testing change against coincidence requires the length, the line and the field to be on the record. None of that sits in a scorecard. So the trend claim is never falsified, because the instrument for falsifying it is never published.

The silence of governance

The match report is never the whole match; the minutes are where the pressure leaves fingerprints. The referee's report holds what no press conference will: who said what in which over, how the pitch behaved, which side burned time, which bowler was warned. It is written and it does not leave the building. A pitch rating surfaces only when somebody is sanctioned — that is, only when the institution needs to show it did something.

I should declare a limit here. I have access to a few referees and a few people inside the board, and that access produces detail no press release contains. My rule is fixed: in any piece where someone briefed me, the piece also prints which questions I sent and which went unanswered. A sourced detail and a verdict do not sit in the same paragraph unless the distance between them is stated. Anything touching integrity or anti-corruption goes first, in its own paragraph, with no sugar on it.

A board's silence is not a mystery; it is a process — and an unpublished process always looks like chaos from outside. A disorderly auction, a schedule changed overnight, a sanction with no visible paper trail: these are usually symptoms of one thing, a page that was written and never printed.

The transcript of silence

Between September and November 2026, sixty IPL matches were played in the UAE with no crowds and with stump microphones pushed as far open as cricket has ever pushed them. Sixty silent matches left a louder transcript than any crowd could produce — the edge, the no-ball, the keeper's warning, the bowler's breathing. That season I wrote a one-page silent-ground checklist for the desk: what a stump mic catches, when a no-ball is audible, how canned crowd noise buries a real edge. It became the template for every behind-closed-doors fixture we covered.

On 19 December that same year, India were bowled out for 36 in Adelaide, their lowest Test total. The scorecard was complete. Josh Hazlewood's five wickets, Pat Cummins' four, Virat Kohli's 74 in the first innings, Jasprit Bumrah's struggle — all of it was recorded. What was not recorded was how many balls in that hour were bowled on which length, and how often each batter pushed at a ball he could have left. The shorter a match's scorecard, the thinner the record of why it happened.

The sentence that gets written every quarter

Here is the line that appears in some column every three months: data has made cricket transparent.

It is half true. Scoring got faster; explanation did not. What got faster was confidence — the graphic now looks far more certain than it used to, and its uncertainty looks far smaller. The error band is the least broadcast part of the broadcast. In that sense data has made cricket confidently opaque rather than transparent.

I also have to rein in my second habit: the reflex to contradict the crowd. Flipping every contested call works at first, then becomes a habit, and then the method starts selecting conclusions. So on Adelaide I will also publish the piece where the crowd was right. The popular read that day — pink ball, fast pitch, one bad session — survives the protocol trail. The lengths Hazlewood and Cummins bowled were aimed at a weakness that existed before that tour. I do not argue with it. Sometimes the crowd is right, and that day it was.

What holds is this: a correct outcome and a correct process are two different things, and we only keep records of the first.

Three asks, and one question

Three demands are not difficult. Print the sync time and the margin on every review graphic — two lines, no argument. Publish every match referee's report within seven days, whatever it contains. And give every data table a column for unknown, because zero and unknown are not the same number and printing them alike is the largest lie available to us.

I resigned once from a chair in Indore, and I have reviewed every comfortable verdict since. Whether cricket has data is an old argument. The new one is this: on the day the data did not arrive, who writes that into the record?