The Empty Column: Cricket Analysis Doesn't Lack Data, It Lacks Provenance

April 2026. On a digital cricket desk in Kolkata I spent four months inside 2026 IPL ball-by-ball files. The target was an Expected Runs Added model and the question was single: why is KKR spending Sunil Narine in the middle order? The pre-season piece ran. That same April, KKR sent Narine out to open. He finished the season with 224 runs at a strike rate of 172.3. The piece drew roughly 400,000 reads. The franchise's own analyst emailed asking for my methodology sheet.

Now the part nobody asked about. The model's sample size was one season. One franchise. And opening was a role Narine had barely played in the IPL before that. What produced 400,000 readers was not the model. It was the verdict. The model stood behind that verdict without showing its papers. That is the oldest habit of this trade, and in the data era it has become an expensive one.

Context

Before any cricket story gets written, six things have to be on the table. Title and source, so the claim can be traced. Information points — scores, overs, players, dates, venues, quotes. Core viewpoint — where the writer stands. Entities — teams, players, coaches, leagues. Time sensitivity — whether this is a live match or series. And source quality — how much the originating source can be trusted. Lose any one of the six and everything downstream looks fine on paper and hollow on the ground.

I learned that reading scorecards, then understood it sitting on a football desk. The spreadsheet started as a cricket scorecard and became a football argument. I did not port the words; I ported the assumptions hiding inside them. That gap is now the most visible thing in cricket coverage.

Format comes first, because the tactical logic of Test, ODI and T20 cannot be poured into one another. A strike rate of 140 is par in T20, good in ODI, and close to irrelevant in a Test. Venue comes second: home-ground tilt, pitch behaviour, dew and DLS bend a result the way a toss does. DRS comes third — how many decisions were overturned, in which over, and whether that reversal changed the direction of the match. Without all three the interpretation stays incomplete, and incomplete interpretation still gets published, because the deadline to publish is shorter than the deadline to collect.

Core

Player analysis starts with role — batter, bowler, all-rounder, keeper. It then demands that average, strike rate or economy be placed against the league and era benchmark. A batting average of 30 in 2026 and 30 in 2026 are not the same object; pitches, bats and T20 have shifted the entire benchmark. Third comes situational splits — powerplay, middle, death; home and away; against spin and against pace. Fourth is recent trend, and fifth the age curve: a fast bowler crossing 30 and a spinner crossing 30 are not the same event.

The hundred is the cleanest illustration. The century is a counting convention inherited from print-era scorecards, not a measure of match value. Travis Head's 137 in the Ahmedabad final of 19 November 2026 and Rohit Sharma's 76 in the Dubai chase of 9 March 2026 both shift their centre of gravity away from the round number and onto the situation once you read them with phase splits, required rate and wickets in hand.

At team level four things matter — batting depth, bowling balance, bench depth, age structure. Here is the unfashionable line. India's bench strength is a selection problem, not a talent problem. The hardest team in world sport is India's second XI, and the proof sits not in the scorecard but in auction arithmetic.

IPL purse maths, retention calls and franchise scouting have quietly replaced the zonal selector. That is now the most honest public signal of what Indian cricket values. It is also where the inherited error lives: the highest price at an auction is the product of a brand race, not of sporting value. The real value signings happen at smaller clubs and among uncapped players, where purse pressure is low and scouting hunger is high. A franchise that reads the auction as an auction wins money; one that reads it as a market wins headlines.

The league-versus-national-team conflict grows out of exactly this soil — NOCs, workload, calendar density. At governance level five checks apply: power and revenue distribution, playing-rule controversies, integrity and anti-corruption monitoring, eligibility and selection, and geopolitics. The share of ICC revenue that settled around 38.5 per cent for India in the 2026-27 cycle is not a finance line. It decides whose needs build the bilateral calendar and whose politics shape neutral venues.

The risk matrix carries six categories — sporting, personnel, commercial, rules and integrity, public opinion, systemic. The least priced of these is systemic, and it is the most expensive: injury load, calendar congestion and cross-format conversion look like three separate problems and emerge from one place.

Narrative heat has to be measured too. Every claim faces three questions: how strong the underlying number is, how large the sample is, and how many days it will last. Where the gap between market expectation and ground assessment widens, the largest profit and the largest loss both hide.

Then transmission. Upstream sits youth development and talent supply; midstream the national teams and leagues; downstream broadcast, commercial markets and fantasy. An upstream event — a coaching method in junior cricket, a state association decision — takes three years to reach the bottom. A downstream event — a broadcaster's cheque — reaches the top in three weeks. Predict anything without that asymmetry of direction and duration and you are guessing.

Contrarian

Now the argument I hold against my own camp.

The consensus position is that cricket journalism improves as data grows. Stated at its strongest, that is also correct: DRS raised accountability on decisions, auction strategy professionalised scouting, and phase splits exposed batters that averages were hiding. That is a real gain.

But the danger is not missing data. The danger is unlabelled data. A number without sample size, without definition and without era is more damaging than a missing number, because a missing number invites doubt and an unlabelled number invites trust. In 2026 my model's sample was one season, and it convinced 400,000 readers that we knew something. We knew one decision and printed it as a rule. Every clean dashboard begins with a dirty question nobody wanted to ask. When an analysis returns an empty list of information points, the honest conclusion is not that the framework failed. The honest conclusion is that the framework is intact and the input is absent — and the published article built on that absent input is the same disease at a different scale.

Every piece of mine should carry a standing disclosure line: which dataset, whose access, and what I am not in a position to say. The day that line becomes routine, half the noise in cricket coverage cuts itself.

Takeaway

The next signal is not a new metric. The signal will be a publication printing its sample size and source beneath the piece — and, on the day it has nothing to say, having the nerve to print an empty column. The question is not how much data we hold. The question is whether we are willing to say where it came from.