Nine Dimensions, Zero Data Points: The Failure Sits in the Harvesting Stage

In nearly fifty years on the analyst's chair, I have read thousands of reports. The one that stopped me longest was empty. It carried all nine professional dimensions: tactics, player data, operations and salary cap, league landscape, rules and governance, locker room, risk, media narrative, industry ripple. Each dimension was a tidy table with assessment, ranking, and risk columns, complete with conclusions and warnings. Yet when I traced cell by cell looking for one number, one name, one game, everything returned the same sentence: insufficient information. No source headline. No outlet. No summary. No author stance. No entity named. The only populated field was a routing label: basketball. That is the entire real content of the report.

Sports desks operate very differently from the days when I still carried a clicker. An article entering the system passes two stages. Stage one breaks the source text into discrete information points: headline, source, one-sentence summary, author stance, entity list, time sensitivity, source quality. Stage two takes that output as raw material and produces deep analysis across nine dimensions. Such a pipeline only runs when stage one has a body.

The report in my hand had no body. Headline blank. Source blank. Summary blank. Information point list empty. Entity list empty. Domain label: basketball. That label is the output of a classifier running on metadata — URL, category, content-management tags — not on the words of the article. It indicates topic, not content. A label cannot distinguish the NBA from FIBA, from the Chinese professional league, or from a US college competition, and each of those systems carries entirely different rules, salary structures, and calendars.

The signature of the failure is a populated label with an empty body. Had the fault been in the summarization stage, the report would still retain fragments: a name, a score, a quote. No fragment means the text never entered the system. Three plausible causes, ordered by frequency: the original sits behind a paywall, the harvesting layer was blocked, or the link pointed to an empty or mis-categorized page.

Another detail worth noting is a circular dependency in the extraction prompt. The related-entities field is instructed with precisely this line: identify from the information points above. When the list above is empty, the instruction ties its own hands. Whoever writes prompts for these systems needs a fallback branch, the same way a print editor needs a clear rule: no source, no page.

To me, the most notable part of the report is its refusal to speak. All nine dimensions were marked insufficient information rather than padded with guesses. In this trade, that is the hardest behavior to sustain. In analytics circles, people still call me the numbers saint, and that title is worth something only when I know to stay quiet. I once broke a story in 2026 after counting Manchester City under Pep Guardiola at 72% average possession in the Champions League with 1,040 completed passes, then declaring Monaco would eliminate them. They lost. In the summer of 2026 in Russia, I sat in the stands watching Spain hold 79% of the ball, complete over a thousand passes, and exit after losing the penalty shootout 3-4 to Russia. Possession is an illusion, goals are the naked truth — but I only dare say that line once I have counted.

On a blank sheet there is nothing to count. That is why the salary cap dimension is the most dangerous one. Every figure there — maximum contracts, luxury tax, mid-level exception — is tied to a specific moment and a specific collective bargaining cycle. Without a stated as-of date, a salary number still looks entirely real while being wrong. The rules, locker room, and industry ripple dimensions are even riskier, because they live on inference. Inference on an empty base produces entities that do not exist, causality that does not exist, and finally a smooth-reading report with nothing that can be verified.

The correct stage-two conclusion, therefore, must be a structured null value, not nine dimensions filled in with plausible-sounding prose. Numbers do not score, but numbers are quietly rewriting history — and empty numbers rewrite history in someone else's words.

Where could I be wrong? This may be a one-off technical glitch rather than a systemic defect. The threshold for calling it systemic has to be a rate: if fewer than 2% of records in a sufficiently long window come through empty, that is a scratch. Above that threshold it is a pipeline fault. In another direction, most Vietnamese sports desks do not yet run this two-stage pipeline, so the local risk is far lower than at units outsourcing high-volume content production. Yet outsourcing is exactly where the concern sits, because there payment follows article count, not source quality.

The more worthwhile counter-angle: the culprit may not be the system but the habit of readers and of ourselves — the habit of demanding an opinion on everything. A report that says insufficient information in all nine dimensions looks like a failure, while a report with every cell filled looks like a success. Newsroom culture rewards volume, not well-timed silence. An analytics professional who cannot stay silent eventually becomes a seller of belief.

A verifiable prediction for the next twelve months: at least one sports outlet will publish an analysis built from empty or near-empty input. The tell is easy: all nine dimensions present, tables intact, but no source headline, no publication date, no named source, and no named entity. Anyone who wants to check can open it and count. Only what can be counted can be spoken.