Empty Input, Empty Report: The Data-Quality Crack in the Cricket Analysis Pipeline

Every Monday a file lands on my desk — the 'Stage-1' deconstruction report of the cricket pipeline. It is the sheet my whole week's analysis stands on: the match's information points, core viewpoints, entities. This time I opened it and every box was empty. Title 'N/A', source 'N/A', and the information-points list contained zero entries. I have a habit of watching the same eleven seconds again, but here there was no footage to watch at all. The rulebook has a number for everything; the footage only has questions — and this time even the questions didn't arrive. In cricket we log every statistic with its source, timestamp and method. That is a discipline, not a hobby. A batter's strike rate, a bowler's economy, the margin on a DRS review — if any of those numbers doesn't show its working, we don't print it. That is precisely the rule of Stage-2 analysis: when Stage-1 is empty, you don't fill the boxes with imagination, you write 'insufficient information, cannot assess.' In today's report that rule was broken outright — or rather, the pipeline had already stalled before it could be tested. Three different boxes told three different stories, so I stopped trusting the one I liked best: the box that says 'there was a match.' I opened all eight possible dimensions one by one — format and match, player technique and data, team landscape and ranking, league and commercial ecosystem, rules and governance, risk, public narrative, and industry transmission. Inside each dimension there were tables, rows, but against every row the same four words: 'N/A — insufficient information.' No format could be identified — not Test, ODI, T20, The Hundred, nothing. No player's name, role, average or strike rate. No team, no ICC ranking, no batting depth. No league, no broadcast rights, no auction value. No rule controversy — not DRS, not DLS, not over-rate. I tried to rebuild the over rate from the scorecard and the ball-by-ball log, but when there is no scorecard, what do I reconcile it against? The temptation to fill this emptiness visits every analyst. Imagine a match, suggest a player's name, invent a controversy — the report fills up and ships on time. But this is where my accumulated habit earns its keep: after that IPL match in Rajkot in 2026 I built a spreadsheet of sixty DRS reviews and found fifteen were 'umpire's call' and twelve were overturned. I published that 20 percent error rate only when I had a frame-by-frame record of every review. Had I filled the boxes with guesses then, that blog post would not have been 'The Umpire's Call Illusion' but just another rumour. The real story here is not about cricket; it is about the pipeline. Stage-1's job is to break the source article into information points and viewpoints; Stage-2 builds deep analysis on those. If Stage-1 returns empty, Stage-2 has only two roads — quietly fill with imagination, or openly refuse. The first is fast, the second is honest. In cricket we treat the broadcast feed as neutral truth because it arrives pre-timed and pre-graphicked; the same mistake is being made with the data feed here. The Stage-1 report is also a pre-formatted feed. It too should be triangulated against a second source — the original article, its publication date, its source grade. In this case all three of those boxes were empty as well. Which means the problem is not one field but the entire upstream extraction. There is a contrarian angle here too, and it stings the most. In cricket we keep accounts of fines on players, code-of-conduct hearings and slow over-rate penalties — because there is a human being who has to bear the weight of the rule. The data pipeline has a 'human' too: the journalist or analyst who must sign off on an incomplete report, and the reader who takes that report as truth and moves on. When Stage-1 arrives empty and Stage-2 fills it with imagination, the penalty falls not on a player but on the reader's trust. I keep a folder of calls that were correct and still felt wrong; today it gained a new entry — a report that was in the right format and still empty. So what next? The most immediate task is to re-run Stage-1 on the original article and confirm that information points, core viewpoints and entities are populated. Second, install a 'null gate' in the pipeline — if Stage-1 returns empty, Stage-2 should stop automatically and raise a warning before publication. Third, every report should carry a confidence statement: where confidence is high, where it is low, and what evidence could change the verdict. Every system leaks somewhere; the job is knowing which leak matters. This time the leak was at the very top — in the source itself. And when the source is empty, the bravest thing a reporter can write is not a complete article but a clean sentence: 'This report is incomplete because the input is incomplete.' Do our quality systems give that sentence as much space as they give a century?