Empty Data, Full Verdicts: The Silent Failure of Cricket Analysis

Last week a file landed on my desk in Sylhet. No headline, no source, no publication date. The information-point field was blank. No format — Test, ODI, T20, or the 100-ball Hundred — was stated. The nature of the match was unknown: bilateral series, ICC event, or franchise league. No venue, no scoreline, not a single player named. An analytical file whose raw material was zero.

The deadline was six in the evening.

You cannot write from zero. But you can write confidence from zero — and confidence is the easiest product to sell in cricket journalism. Had I written from that file that "this bowler's death-overs economy is a concern", nobody could have caught that I invented the number. The sentence would have been grammatically perfect, tonally assured, and hollow inside.

Take the number 140. A 140 strike rate is elite in ODI cricket; in a T20 top order it is ordinary, and for an IPL finisher it is inadequate. One number, three different verdicts. The number did not change; the frame around it did. This essay is about that frame — not about a match lost, but about a verdict formed before the match was understood.

Four anchors of a number

My first fifteen years of work were spent in Sylhet's regional press, writing match reports. After being dropped from a television panel in 2026, I stopped writing reports and moved into zone-based analysis. One thing became clear slowly: no number in cricket can stand alone. Every number needs at least four anchors — format, match nature, venue, and sample size. If any one of the four is missing, the number stops being information and becomes inference.

Format comes first because it fixes the benchmark for every metric. An economy of 4.5 is the mark of a good bowler in ODI cricket; at the T20 death it is almost miraculous. An economy of 3.5 in Tests means control; 3.5 in T20 means the opposing batters have surrendered. Batting average works the same way: 35 is acceptable in Tests and nearly meaningless in T20, where the question is not how many runs per dismissal but how many runs per ball. An analyst who delivers a verdict without knowing the format is not delivering a verdict; he is dressing an assumption in a verdict's clothes.

The 100-ball format is the cleanest illustration of this confusion. The Hundred has no overs, only five-ball sets. Anyone who calculates "economy per over" there is making a multiplication error, not an interpretation error. Small mistake, large consequence: with the wrong benchmark, a correct player can be proven incompetent.

Match nature is the second anchor. The same scoreline does not carry the same meaning in a bilateral dead rubber and a World Cup knockout. The pressure differs, the opponent's intent differs, the team's freedom to make decisions differs. A young batter's 70 in a dead match is an opportunity; the same 70 in a knockout is proof.

Venue is the third. Mirpur's spin-friendly pitch and Rawalpindi's flat deck can pull the same bowler's career in two directions. And the fourth, most ignored anchor — sample size.

Sample size: the myth of three matches

Bangladesh cricket's real crisis is not talent but narrative velocity. A player does well in three innings, the press declares him a "discovery", the selectors pick him, he fails in two matches, and the same machine discards him. That cycle does not test merit; it is a random number generator with a press office.

I have sat at Mirpur many times and watched a young batter look technically composed in first-class cricket while the numbers showed an abnormal reliance on the cover drive. It works in one match, works in two, fails in the third — because the bowlers have watched the video. The question here is not "who is a good player"; the question is "how large a claim is being made on how much data".

My minimum standard is this: any claim about a player's form needs at least a twelve-month trend, a comparison against career baseline, home-and-away splits, and a pace-versus-spin split. Without those four, the claim can still be made — but it will be an opinion, not analysis.

This is where the Rawalpindi test of August 2026 becomes useful. Bangladesh travelled to Pakistan and won the two-Test series 2-0 — the first Test by 10 wickets, the second by 6. That result is true and verifiable, because all four anchors are present: format Test, nature bilateral series, venue Rawalpindi, sample two matches. That last anchor is the warning — two matches are not a trend, they are an event. Nahid Rana's pace genuinely started a conversation in that series, but the claim that "Bangladesh's fast-bowling problem is solved" cannot be carried by two matches of data. We can speak about process; not about career.

The silent cost of format mixing

Cricket analysis's most common error is not complex, often it is tediously simple: Test statistics used for a T20 decision, or franchise-league form used to justify a Test place. This mixing never shouts. Nobody announces, "I am using the wrong format's data." Instead a clean table is shown, and the table itself does not lie — its heading lies.

My habit is to write one small question before every match analysis: "In which format is this claim true, and in which format does it break?" If the second question cannot be answered, the first answer is incomplete. This is not formality — it is the kind of discipline without which a number is merely an ornament.

Home and away: cricket's largest variable

The gap between a player's home average and his away average is often the most important fact of his career, and often the least quoted. The reason is understandable: the home number is flattering, the away number is uncomfortable.

For a bowler like Taskin Ahmed the question is never "is he good"; the question is in which environment his new-ball line and length are most effective, and in which environment he drifts to a short length. The answer lives not in the aggregate average but in venue splits, pitch type, and the phase of the innings in which he is bowling.

Home advantage is not only the pitch. The seam of the ball, the light, the crowd, the travel, even the umpire's familiarity — together they form an ecosystem. An analyst who reduces that ecosystem to a single number called "home advantage" is not analysing; he is labelling.

Transfer window: price and sporting value are not the same thing

The transfer market trades in narratives before it trades in players.

On 24 November 2026, at the IPL mega auction in Jeddah, Rishabh Pant went to Lucknow Super Giants for ₹27 crore — the highest price in IPL history. Shreyas Iyer went to Punjab Kings for ₹26.75 crore. Both numbers are true, verifiable, and nearly useless for analysis if you read them as evidence of sporting value.

An auction price is the sum of three things: the player's sporting capability, the intensity of a franchise's specific need, and auction mechanics — purse size, right-to-match position, who bids when. If a team is looking for a wicketkeeper-batter and there are two in the market, the price is the price of scarcity, not of talent. Miss that distinction and the auction number is mistaken for cricketing truth.

This is where the agent enters. Agents do not win matches or create stars — they control the flow of information. Who hears the news first, which franchise is made to look interested, which video clip is released in which week — all of it builds an invisible layer on the price. In my long experience, much of the "sourced" news circulating in the four weeks before an auction exists for one purpose: to set the price. Our job as journalists is not to join that flow but to verify its origin.

And the board holds a lever here. The No Objection Certificate — the NOC — is the mechanism by which a board decides which player plays in which league. That power can change a player's market value, his rest plan, even the timeline of his international career. An analysis that ignores NOC politics and looks only at performance is seeing half the picture.

Fatigue is a formation, not a feeling

Fatigue is a formation, not a feeling.

I first used that sentence in a football context, but in cricket its application is more specific. A fast bowler's fourth-innings economy rises above his first-innings economy — true in general, and almost useless when stated in general. The working question is: where does fatigue stand?

The answer is visible on the field. A tired bowler's yorker becomes a full toss, so the captain pushes the fielder at long-on further back. The slower ball's grip loosens, the batter plays early and goes leg side, and the field rotates. These small relocations are fatigue's signature — not on the scoreboard, on the field map.

But here is the caution. Not every death-over collapse is fatigue. If the same bowler's 16-20 economy is stable across three seasons and jumps only in the fourth innings of back-to-back Tests, that is workload. But if it rises only against left-handers, that is a matchup, not fatigue. Separate the two or the analysis stops being analysis and becomes grievance.

My own test is simple: before claiming fatigue, rule out three things — injury, technique (grip, release point), and match state (target, wickets fallen, Duckworth-Lewis pressure). If the drop persists after removing those three, then fatigue can be named. Otherwise we are simply choosing a comfortable explanation.

Data doesn't erase pressure; it relocates it. The pressure on a batter walking in at number four does not shrink because we show his "finisher strike rate"; the pressure merely changes its language.

The half-space: where the game hides its intentions

The half-space is where the game hides its intentions.

In football the half-space is the corridor between full-back and centre-back. Cricket has no direct equivalent, but it has a working one — and I define it separately for each format.

In T20 the half-space is the corridor between two orthodox ring stations: for a right-hander, the angle between cover and point, where the captain usually places a fielder to stop the cover drive and another at point — but the ball exactly between them belongs to nobody. That is where the batter builds his run flow. In ODI the half-space sits deeper, between sweeper cover and deep point, where fielders thin out after the powerplay. In Tests it is an entirely different thing — the area between slip cordon and gully, where the new ball's character decides who wins.

Why do these definitions matter? Because once the phrase becomes magic, it stops working as analysis. Without a separate geographical address for the half-space in each format, we get a beautiful sentence and not a decision.

I have a habit of drawing fielding maps in my notebook — not before play, during it, every six overs. A fielder who shifts one step is the most honest signal of a captain's changing plan. The scoreboard can lie; fielders' feet usually cannot.

The merit gate: unstated criteria cannot be audited

Discontent about selection in Bangladesh cricket is permanent, and most of it is personal. My objection is not personal, it is procedural. The problem is not that some selector has a bias; the problem is that the criteria are never published.

Suppose the criterion is recent first-class average. Then write it down, publish it, and let everyone verify who qualifies. Suppose the criterion is "horses for courses". Then say which course, and why this player's specific skill is needed on that course. Unstated criteria cannot be audited — and where audit is impossible, merit and identity cannot be told apart.

The second layer is quieter: access to information. A domestic player who scores every season but has no agent, no media manager, no two-minute highlight reel is effectively invisible. That is not bias; it is a failure of discoverability. And that failure is distributed unequally by gender, region, and club connection.

In women's cricket I see this error repeatedly. Discussion of Nigar Sultana's Bangladesh side is often trapped in the "progress of women's cricket" frame and never enters the tactical frame. Yet the same questions apply: who breaks the fielding ring in the powerplay, how much control does the spinner hold through the middle, who can bowl the yorker at the death. My notebook has no separate page. Same map, same questions, same standard.

The counter-angle: more data has made analysis more confidently wrong

Here an uncomfortable admission. Over the last decade the volume of data in cricket has exploded — tracking, ball-by-ball logs, field maps, workload monitoring. The expectation was that analysis would become more accurate. What happened is different: analysis became more confident, and accuracy did not rise by the same amount.

The reason is that the danger is not empty data — it is unlabelled data. An empty file forces nobody to lie; but a full file with no format tag, no sample size, no source grade makes the analyst feel safe. He fills the gaps with his own pattern memory, and the output looks like experience. The most dangerous report is not the one with many blank fields; the most dangerous report is the one with no blank fields and not a single source.

The second counter-angle is subtler. Rigour about sample size can itself become gatekeeping. The sentence "there isn't enough data" can silence players from small boards, new formats, under-broadcast women's tournaments — while decisions must be made today and the data arrives in three years.

The answer is not to abandon rigour; the answer is to separate types of claim. A two-Test sample cannot carry the claim "what will this bowler's career average be". But the same sample can carry another claim: "this bowler's release point and new-ball line work in these conditions". The first is an outcome claim, the second a process claim. Process claims survive small samples, provided you say clearly that they are about process.

My rule, then, is not "no data, no opinion". The rule is: attach to every claim its confidence level and its falsification condition. A claim that cannot be broken is not a claim; it is a belief.

Takeaway: three questions before the next match

The next time a squad list is printed "citing sources", or a star's "grievance" surfaces before an auction, or a youngster's three innings produce a "discovery" headline — ask three questions. One: which format, and what is the benchmark there? Two: how big is the sample, and is the claim about outcome or process? Three: if this news is true, who gains most — the agent, the board, the franchise, or the analyst under deadline pressure?

If the three answers do not line up, hold the verdict. In cricket you can afford to be late; a wrong decision cannot be recalled.