The Cricket of N/A: When 'No Evidence' Gets Read as 'No Problem'

Two in the morning, and a spreadsheet with eleven columns is open on the screen — format, match, player, role, metric, benchmark, conclusion, source, time-sensitivity, risk, notes. Eleven columns, and the rightmost one repeats the same word down every row: N/A. Below the sheet sits a sentence I have read at least forty times in the last six months — "no serious issues found."

The sentence was not a lie. It was worse. The truth was that nothing had been examined. The structure was intact, the cells were clean, the headers sat where headers should sit — there was simply nothing underneath. And when a structure looks that tidy, the reader assumes the check has been run.

Cricket types that sentence somewhere every week. In a selection meeting, at an auction table, in a commentary box, and now on fantasy dashboards. A player has no data, so he is filed as "untested" — as though untested were a negative finding. A bowler has twelve death overs in empty grounds, so his price falls. A match has no available footage, so the scout report says "no red flags."

All three mistakes are the same mistake: reading "no risk found" as "no risk exists." In analytics language, the first is unrated; the second is low-risk. The gap is not subtle. It is the whole system. And cricket's information economy is built on erasing it.

Where the data comes from, and where it breaks

Cricket's data no longer arrives from one place. Ball-tracking cameras log the ball's path, the broadcast feed pushes score and field positions, and a coder at a desk at two in the morning tags every delivery — line, length, shot, fielding position, outcome. All of it runs through a two-stage pipeline. Stage one deconstructs and tags. Stage two interprets.

There is a rule every data desk knows and every publication ignores: stage two cannot manufacture evidence that stage one failed to capture. If extraction returns an empty shell — a JavaScript-rendered page, a paywall, a parser timeout — the entire downstream argument stands on an empty container. And the container does not announce itself. It arrives with metadata. It has a headline, a date, and a column that says N/A.

I have seen this pipeline from the inside. After the 2026 Ranji final I rebuilt Parthiv Patel's field settings from broadcast screenshots at Indore — fourteen diagrams, one question each: what did Gujarat change inside the thirty-yard ring? Mumbai lost that game in the ring, and the conclusion came from screenshots, not scorecards. But the exercise carried a flaw I only understood later. The field I drew matched the field I assumed must have been set. Broadcast gives one angle at a time, and the mind fills the rest. That is chalkboard retro-fitting, my oldest habit. The only defence is to check the actual field at the actual ball — and if the source is memory or a highlights package, to say so in the copy.

In 2026 the grounds were empty and my MS was online. I coded 312 death overs from the 2026 World Cup and the UAE-held IPL — slower-ball usage, yorker length, field changes. Teams bowled 14 percent more slower balls at the death in empty grounds, and defending captains brought third man up earlier. I still cite that number, but only alongside its format, its phase and its sample. Putting a Test average, an ODI strike rate and a T20 economy into one sentence is the most common and most expensive error in cricket analytics.

Selection: bench strength is a role problem, not a talent problem

The bench-strength debate returns after every series in the same shape — who should have played, who should not. The frame is pre-set, so the conclusion is pre-set too.

I look at it from the other end. The second XI is a role-definition problem, not a talent problem. Take a middle-order batter: of his last sixty innings, thirty-eight came at number five and twenty-two at number six. The database calls him inconsistent, because his average at five is 41 and at six is 39. The real story is that at six he was sent in with six overs left every time, while at five he was allowed to bat from the twentieth over. The averages look identical. The roles are not.

An average does not measure a role. It measures an outcome, and outcomes are made by roles. So when a selection panel says his record at five is poor, the right question is not why the record is poor. It is how often he was allowed to bat before the twentieth over at that position.

The other end of the same argument is more dangerous. A domestic batter has fourteen first-class innings, none against elite bowling. He is dropped from consideration because the sample is small. But dropping him is a conclusion, and conclusions need evidence. Here there is none. The correct status is unrated — neither good nor bad. When you convert unrated into "untested" and drop him, you have not measured anything; you have turned an absence of measurement into an opinion.

This is why I distrust milestone framing. Fastest to ten thousand runs is a broadcast convenience, not an analytical unit. In which phase were those runs made, against which field, against which attack? Without that, the number is a date stamp.

The auction: price is not what you think it is

The transfer window is a market for other people. In cricket it is called an auction, and its founding rule is that price and sporting value are different objects.

A franchise bids on a death bowler with twelve death overs last season, two of which went for four. Twelve overs. Seventy-two balls. Forty of those balls came in dead rubbers. He goes for three crore because the last two overs looked good in a highlight reel.

That is not valuation. It is an inference from a sample the broadcaster already edited — what was shown was the most dramatic, not the most representative. A highlights package is a sampling bias nobody edited, because editing is invisible.

The other end is just as expensive. An uncapped player has no televised data. His price is set not by his cricket but by the absence of his data. The scout says "no red flags," and that sentence takes him to base price — when all it actually said was that nobody looked.

I am not calling the auction irrational. I am saying it is a market where the price of missing information is real, and it does not always move in one direction. What never happens is recording the missing information as information.

DRS: one word, two different questions

This is one of my oldest complaints and it is linguistic, not technical. DRS is a system that answers two entirely separate questions with one word — umpire's call.

First question: would the ball have hit the stumps? Second: does this footage contain enough evidence to overturn? Those are not the same question. One is about the ball; the other is about resolution. Broadcast renders both in a single graphic, and the viewer reads both as one verdict.

The system is not measuring the ball; it is measuring the distance between the ball and the evidence. When that distance collapses, the system retreats into umpire's call — technically honest, communicatively confusing.

There is a second cost, which is time. A long review cuts the rhythm of the match. Two minutes of waiting is enough to cool a celebration, and in cricket that celebration is a wicket. The sharpest moment of the game becomes a procedure. In my reading, two minutes is a sufficient ceiling; beyond that it is not accuracy, it is ceremony.

And here I have to guard against my own reflex. Anger at DRS is easy, and it usually curdles into a moral verdict on a player or an umpire. That is not my method. My question is what the system claims to measure, what it actually measures, and at which phase the gap between those two becomes widest.

Empty stadiums: when every field call becomes audio

The 2026 IPL in the UAE was an accidental experiment. Empty stands, live mics. It revealed things a full ground never shows.

Empty stadiums turn every voice into a tactical leak. The bowler's grunt, the keeper's appeal, the warning from slip — all of it lands in the audio. A full Eden Gardens hides a misfield the same way it hides a fielding call. That is why I treat crowd noise as an input, not as atmosphere.

But the easiest error sits right here. The observation is true, so it generalises far too quickly. Turning every misfield, every run-out and every wrong call into an acoustics story is easy — especially where the stands genuinely are sparse.

So the order matters. First decide whether this was a communication failure or a positioning failure. If the fielder was in the wrong place, noise is not the story. If he was in the right place and never heard the call, noise is the story. Merging the two is not analysis; it is laziness.

Phase architecture: a number means nothing without its situation

At Ahmedabad on 19 November 2026, Travis Head made 137, India were bowled out for 240, and Australia reached 241 for 4 in 43 overs (source: ICC match report). The number is true and, on its own, tells you almost nothing.

Break it into phases. How much did India lose in the powerplay, how did the middle-overs wickets fall, what did they add at the death. Then break Australia's innings the same way. Only then does the 137 mean what it should mean: an innings built in a particular phase at a particular risk tolerance, and the precise ball on which India's bowling plan stopped being followed.

That is the execution audit. The question is never whether the plan was good. It is at which over the plan stopped being followed, and who stopped following it.

Take the 2026 World Test Championship final at The Oval, won by Australia by 209 runs, or the 2026 final at Southampton, won by New Zealand by eight wickets. Read either as "India's Test batting failed" and you get a story and lose a mechanism. The chalk was right; the legs arrived late.

Preparation and delivery are two distinct failures with two distinct culprits. The culprit in preparation built the plan; the culprit in delivery abandoned it on the first difficult ball. Merging them is convenient, because then you only have to pick one name.

29 June 2026, Barbados. India 176 for 7, South Africa 169 for 8 (source: ICC match report). The margin is seven runs, and the seven runs are nowhere. They were made in the eighteenth, nineteenth and twentieth overs — a different decision, a different bowler, a different field in each. Jasprit Bumrah's over, then Hardik Pandya's, then the last one. And Suryakumar Yadav's catch on the boundary rope. That is death-phase architecture, and calling it "the better team under pressure won" is exactly the shortcut I refuse.

The contrarian turn: the problem is editorial, not technical

Everything above — the empty shell, N/A against low-risk, unrated against untested — sounds technical. The root is editorial. The publishing format has no slot for the conclusion "insufficient information."

A match preview is three hundred words. A team-news item is two hundred. You cannot write "we do not have enough information about this match and therefore offer no view." The sentence may be true; it cannot be printed. And if it cannot be printed, the schema will be filled. Filling the schema is a system pressure, not a personal dishonesty.

The direct consequence is what I call information-gain pressure. Every piece must carry something new. But not every match carries something new. Sometimes a match is just a match. When the format demands a fresh insight and the match supplies none, an insight gets stretched into place — a number that previously meant nothing becomes a "trend."

My second warning is for my own side. Number-skepticism is a reflex, and Indian coverage feeds it endlessly. But when skepticism curdles into number nihilism, no measurement survives in the argument. Point the skepticism at interpretation, not at measurement. Keep the phase split; discard the ranking. That is my own rule and I break it monthly.

The third point is the least comfortable. An empty shell is a diagnostic signal. When the metadata shell is intact and every content field is null, the likeliest explanation is not that the source contained nothing. It is that the source never arrived. "Insufficient information" can never become "no risk." It is unrated — not low-risk.

Which is why I do not read any of this as the story of one selection or one bad call, and I will not convert a defeat into one person's moral failure. The moment we do that, we stop reading structure. And reading structure requires the same discipline as reading an empty sheet as an empty sheet — against the pressure that every sheet demands a verdict.

Draws leave fingerprints. I learned that from an exercise where the result was nothing: a drawn match, an empty scorecard, and still fourteen diagrams showing where the field opened. Zero is information too. It only asks that you admit you do not know.

What to watch next

When the next number appears on screen, ask it four questions. Which phase? Which format? What sample? And most importantly — is this no risk found, or no evidence examined? The fourth answer never makes it onto a broadcast graphic, and it is the one that tells you the most.

My eye is on that in the coming weeks, as the selection debate returns, the auction table is laid out again, and the same DRS graphic rolls. Every time, an N/A will appear. The question is who reads it — a person who knows an empty cell is an empty cell, or a pipeline that will render it "clear."