The Empty Evidence Base: A Protocol for Not Turning 'N/A' Into a Conclusion

One page in my notebook carries no score. It carries eight columns — format, player technique, team landscape, league and commercial ecosystem, rules and governance, risk, public narrative, and cricket-industry transmission. Every column ends in the same word: N/A. I counted twice, because the first time I assumed I had skipped a row. Eight out of eight. Not one dimension reaches a name, a date, a ball, an innings or a contract.

That page arrived from a Stage-1 deconstruction. Its job is simple: pull factual points out of the source material, isolate the core viewpoints, and identify the players, teams, organisers and institutions involved. It did none of the three. What came back was a scaffold of eight dimensions, each carrying the same sentence — assessment not possible.

Within twenty minutes of opening that file I had a headline in my head. That is the actual danger, and it is what this piece is about.

Cricket analysis runs in three layers. The first is extraction — pulling facts out of raw material. The second is analysis — weighing those facts on the scales of format, sample and context. The third is writing. When the first layer comes back empty, the second does not fail; it simply cannot start. The two failures need different corrections, and treating them as one is the most expensive mistake in the system.

Take the laboratory comparison. If a test report comes back blank, no doctor says the patient is healthy. He says the sample has to be taken again. In the cricket content industry this ordinary understanding breaks down constantly, because here a column gets filed on a blank report.

Across thirteen years of watching and writing matches, that pressure is not new to me. At Russia 2026 I filed nineteen pieces across one tournament while football was three per cent of that site's traffic and my editor kept asking for cricket. In a twenty-four-hour cycle, saying 'I have no opinion' is the hardest sentence available. Saying 'there is no fact here' is harder still — and the entire trust in the work rests on that price.

Format has to be settled first. This is not a formality. Test, ODI, T20 and The Hundred carry different constraints: overs, field restrictions, bowling quotas, squad depth. Where the format itself is unsettled, applying the verticality lens to the match is borrowing a lens. Where the format is unnamed, every comparison is a debt — and in cricket analysis the most expensive debt is settled by mixing formats. The depth of a Test second innings and the depth of a T20 death block cannot be measured with the same word.

After format comes the player, and here the sample question must be asked first. A batter's average, a batter's strike rate, a bowler's economy — all three are meaningless numbers without format-specific benchmarks. Without home and away splits, there is no way to tell whether a figure is ability or a ground's gift. Without a twelve-month trend, separating improvement from decline becomes guesswork. This is where the bubble I have written about repeatedly inflates: paying a three-figure sum for a player with fewer than fifty first-class matches is not analysis, it is gambling. But even that claim needs a name, an age, a format and a fee. In an empty file the claim collapses into an opinion.

At team level the first condition is that ICC rankings be read on three separate scales. Test ranking is not ODI ranking, and T20 ranking says nothing about strength in any other format. Squad depth, bowling combination, bench capacity and age structure — only together do these four say whether a side can win a series or sustain a decade. Without them, team analysis is a list of names.

On league and commercial ecosystem my method is fixed: retention arithmetic, purse allocation, and the slow effect of the Impact Player rule that quietly rewrites XI construction. I read a franchise roster the way an engineer reads a load path — which part is carrying weight, which part is decoration. Seen from a South African formation, one comparison comes to hand: the IPL's closed shop against the SA20 pipeline that actually exports players. That comparison is real, but it stays real only when the league, the season and the contract structure are present. In an empty file it is a template, not analysis.

At the governance level I carry one standing question, and I apply it to every official statement: who benefits from this version, and what would the numbers look like if it were false? Revenue distribution, DRS interpretation, Duckworth-Lewis calculation, NOC regulations, the selection committee's language of 'process' — each of these is a field where the institution's version and the facts' version run separately. The way lengthy video reviews cut into a match's rhythm is part of the same question for me: a wait beyond two minutes cools the celebration itself, and when rhythm breaks, the reading of the result changes with it. But this entire section requires knowing which institution, which rule and which match the question belongs to.

On the risk matrix the biggest error is the plainest: mistaking unknown for low. A matrix where every row says 'unknown' is not a green light. Missing information does not reduce risk; it makes risk invisible. Injury load, workload, availability, contract status, schedule density — not one of these can be measured without a player's name.

Public narrative can only be analysed once the narrative is identified. Signals of frenzy and panic, the gap between expectation and actual capacity, and how long that gap holds — all three are built from media, social feeds and audience numbers. Without them, saying 'the market overrated him' or 'the market underrated him' is directionless guesswork.

And the last layer — industry transmission. Upstream sits youth development and talent supply, midstream the national teams and leagues, downstream broadcast, sponsorship and derivative markets. That map is a general framework, not a description of any event. Imposing it on a news item is the same as drawing pressing lines on an empty ground.

Which brings me back to my own habits. My biggest weakness is that the chart is both my method and my identity — and the labour of counting starts to feel like proof in itself. Reading this file, for a moment I thought 'eight N/A out of eight' might become my thesis. It would have been a beautiful number and a useless argument. Any structural claim has to stand in one sentence with no numbers in it; if it cannot stand alone, the chart is decoration.

The second habit is institutional deference. Protocol is my instinct, so accepting any statement about 'process' as professional is easy for me. In this case the Stage-1 output arrived like an official statement, and I was about to accept it. The question is unchanged: who benefits from this version? Here the answer is plain — the process that wants to keep supplying an article even after an empty extraction.

The third point matters most: an empty file is not dangerous. Dangerous is the article that is born from that empty file and reads better for it. A system that fills gaps will always read better than one that shows them — and that is both its whole advantage and its whole cost. I opened the Kochi notebook and found the match had already drawn its own arrows; I counted the 1,029 passes until control stopped being a statistic and became a structure; and in Goa's empty stadiums I found the empty seats had not erased atmosphere, they had exposed the pressing lines underneath it. All three times the argument came before the number, not after.

So what comes out of this file is not a cricket decision. It is a process decision: set four triggers before moving. First, the original material or its full text becoming available; that unlocks everything. Second, players, teams, leagues and organisers being named; that fixes the scope. Third, format confirmation; that supplies the correct benchmark. Fourth, independent corroboration; an official announcement or credible report that leaves the claim standing on more than one leg.

The next time a deconstruction hands you eight empty columns out of eight, the question is not what you will write. The question is how quickly you will accept that there is nothing yet worth writing — and whether you will publish that refusal as cleanly as you publish a finished conclusion.