The Empty Cell Is a Lie: Null-Handling in the Cricket Data Pipeline

Last month, at a desk in Dhaka, I opened the coding sheet for a T20 series. One column was meant to hold powerplay economy figures. The cell was blank — no zero, nothing. Two people looked at that blank and reached two different decisions. The video analyst said, "No data, drop him." The selector said, "No data means no problem, he is economical." Both were wrong, and the shape of the error differed. For one, a blank meant unknown. For the other, a blank meant harmless.

That same week a larger blank landed on my desk. A deconstruction report in which every field read the same thing: insufficient information. No title, no source, no entities, no time sensitivity. All eight analytical dimensions empty. The most expensive lesson in cricket analytics sat in that file: the costliest thing on a spreadsheet is a blank cell, because a blank cell is the only cell a human fills with memory alone.

Geometry before adjectives

In 2026 I started a one-man blog from a dorm room in Dhaka called The Half-Space. The first post mapped Abahani Limited Dhaka's 4-2-3-1 against Sheikh Jamal Dhanmondi on a 5x6 grid I drew myself in Excel. By December there were fourteen posts and 412 subscribers. A piece on Antonio Conte's 3-4-3 at Chelsea was shared roughly 3,000 times. The habit that came out of it still shapes every match piece I write: start with geometry, not adjectives — a shape, a distance, a coordinate.

I built this from a Dhaka dorm room, so I trust patterns more than press boxes. Trusting patterns is not blind trust. To see a pattern you first have to know which column is missing from your sheet.

In 2026, working as a junior video analyst for Bashundhara Kings, I coded all 26 matches of their title-winning debut BPL season. That summer I watched all 64 World Cup matches across 21 nights and tagged more than 1,100 set pieces, confirming that dead balls produced a record share of Russia's 169 goals. In September I coded Bangladesh's SAFF Championship matches at Bangabandhu National Stadium.

Twenty-one sleepless nights in Russia taught me that fatigue is a dataset, not a badge. They taught me something else too: you can only write "14 of 22" when the tagger counted every ball even while exhausted. That number becomes the spine of a tactical claim, because a reader can check it.

A two-stage pipeline and one hard rule

My current work runs on a two-stage structure. Stage one breaks the source document or match into facts — who, when, which format, which source, which entity, how time-sensitive. Stage two analyses those facts across eight dimensions: format and match, player technique and data, team and ranking, league and commercial ecosystem, rules and governance, risk, public narrative, and industry transmission. The rule fits in one line: every judgement must be grounded in a stage-one information point, and baseless speculation is prohibited.

This is where an audit trail matters. A coding log that is traceable, verifiable and reusable is the real asset, because six months later someone will ask which ball carried which tag. A log nobody can verify is not a log, it is a claim.

So when stage one comes back empty, stage two has nothing solid to stand on. This is where something familiar and rarely admitted happens: the empty space never stays empty.

Three ways to fill a blank cell, all wrong

First — treating blank as zero. A bowler whose powerplay economy was never coded drifts into the "safe" column of the depth chart. Missing data and good data are different things. In a Dhaka domestic league I watched a spinner's powerplay spell coded for three of seven overs, with four overs blank; on the table he became the cheapest option available. The coach picked him for the powerplay on the strength of that blank, and the match changed pace inside two overs.

Second — treating blank as neutral. A batter has an average, but no column for how many catches went down. The average then gets read as proof of technique, when a large slice of it is the opposition's hands. A cricket scorecard never weights fielding errors, and we quietly write a zero into that blank ourselves.

Third, the most expensive — filling the blank with narrative. No data means no analysis, and who fills the vacuum of analysis? Authority. An ex-player's memory, the press box consensus, "I have watched him play." Authority is a null-filling device. Where numbers are absent, voices win.

A cheap habit: the null audit

My sheets now carry one rule. Under every match table there is a separate block titled "what is missing". How many overs of powerplay data were dropped, how many drop catches were never logged, which bowler's death-overs spell was never coded. I read that block once a week. That single habit has changed my analysis more than any new metric I have adopted.

Cricket already built its own null value

Consider umpire's call. Under DRS, when the ball's pitching or the fraction of stump struck falls inside a defined margin, the on-field decision stands. The system deliberately refuses to decide — a null value cricket built with its own hands. DLS does similar work: after rain the target is revised, leaving a gap in the original match record that nobody later fills.

The existence of both rules tells you the game already knows some cells stay blank. The problem starts when an analyst fills that blank to taste — taking one team's umpire's call and drawing a format-neutral conclusion from it.

The column that was never on the sheet

A blank cell and a missing column are different diseases, and the second is worse. In 2026 the BPL stopped in March and in June my contract was not renewed. I did not apply for work for five weeks. Laid off in June, saved by empty stadiums — I re-watched the 92 remaining Bundesliga matches of Project Restart and logged every result. Home teams' points per game fell from 1.62 to 1.28, while away wins rose from 29 percent to 37 percent. "The Silence Effect" ran in October.

The thing to notice is that crowd presence was never a column on any sheet. Nobody ever placed a "spectators" column beside match data, because it was a constant — present at every match. Constants do not enter models, because their variance is zero. Then variance arrived, and it turned out part of what we had been measuring as home advantage was, in fact, human noise.

Transfer window: blank cells are the fuel of rumour

The same disease shows up in the transfer window wearing different clothes. The real story is a club's wage bill and the structure of its release clauses, and it never reaches a headline, because contract gaps surface in private. The blank cells are exactly what fuels rumour — and rumour velocity beats the speed of truth.

On July 3, 2026, after a single season at Benfica, João Félix moved to Atlético Madrid for 126 million euros, for a player with fewer than 50 senior appearances. That transfer was a decision without a column: potential was priced, its variance was never logged. The young-player premium was a one-way spread, and that spread is now compressing. Rules like Right to Match and the No Objection Certificate create league-versus-country tension, yet the rumour table has no cell for either.

Coded columns versus inference

On July 11, 2026, eighteen hours after the Euro 2026 final at Wembley, I published a twelve-page breakdown of Italy's build-up — Jorginho dropping between the centre-backs, Spinazzola carrying forty metres into the left half-space. It was translated into four languages and read roughly 300,000 times.

The thing to notice there is structure, not tactics. That analysis was possible only because the build-up phase had been coded as its own column. Without a "build-up" cell on the tagging sheet, Jorginho's position and Spinazzola's carry would have been visible to the eye but unprovable. Eyes notice; sheets persuade.

Pedri's case says the same thing from the other direction. In the 2026-21 season he played 52 matches for Barcelona, then Euro 2026 for Spain, then the Tokyo Olympics — more than sixty matches across club and country on a teenager's legs. The club's workload column existed; nobody joined it to the injury table. The lesson from Russia applies here: fatigue is a dataset, not a badge. Praising a young player as hard-working while leaving his workload column blank are two acts performed at the same table.

Risk flags are really a column checklist

My stage-two framework carries five risk flags: mixing formats, over-extrapolating from a small single-match sample, ignoring home-ground advantage, failing to strip out luck factors such as the toss or DLS, and DRS controversy. Look closely and all five are one question in five costumes: which column did you forget?

Blending a Test average with a T20 strike rate, deciding a series on three matches of form, hiding away weakness behind home-made centuries — each is a judgement standing on a blank cell. Isolating the luck factor means admitting some results sit outside your model, and that is the most honest column of all.

The press box vacuum and the transmission chain

Public narrative is another blank-filling machine. A win produces "a dynasty begins", a loss produces "crisis". Meanwhile a gap always exists between fundamentals and market expectation, and nobody accounts for it. Panic and frenzy signals ring loudest exactly where information is thinnest.

Seen through industry transmission, the picture is clean. Upstream sits youth development and talent supply, midstream the national team and franchise leagues, downstream broadcast, commercial deals, fantasy and betting — and that last segment fills blank cells fastest, because speed there is profit. The same rule applies: every claim has to rest on a traceable information point.

The blind spot nobody audits

Analysts audit their numbers and never audit their columns. That is the execution gap. You can verify averages, strike rates and economies to the decimal — nobody will catch the error in a column that is not on the sheet, because the error is invisible.

There is a more uncomfortable possibility, and I will admit it plainly. A null result has two explanations: the source document really was empty, or something broke silently in the intake pipeline. Both point to one action — verify the pipeline. Speculation is the expensive option, because speculation means pouring imagination into a blank cell, and nobody ever checks the source of an imagined number.

Counter-intuitive takes are themselves a null-filling trap. Showing an odd pattern is easy; the hard part is offering a prediction that can be proven wrong. If a contrarian view produces no testable claim, it is not analysis — it is a blank cell dressed in adjectives.

What to watch in the next match

Before you read the next scorecard, do not read the scorecard. Count the empty cells first. Ask which column is missing from the sheet. A model that cannot tell you which of its columns is absent is not a model, it is a press release. Watch which cell fills first in the next series, and which stays blank forever — because the blank nobody fills is the real subject of your next piece.