Two Meanings of Hockey: Field vs Ice, and the Analyst's Trap

Last month a data file landed on my desk. Three columns — date, team, score. The header held a single word: hockey. I sat quietly for nearly ten minutes. The file never said whether this was field hockey or ice hockey. Yet behind that one word sit two different rule systems, two different calendars, two different tactical vocabularies. In October 2026, at sixteen, I took the night bus from Sylhet to Dhaka with 1,200 taka and a spiral notebook. At the Asia Cup at Maulana Bhasani Stadium I did not watch the ball; I watched India's left half's starting position, the injector's angle, the stopper's feet, the drag-flicker's run-up. I found the 41st corner in the Maulana Bhasani notebook, and the match changed shape. There, 'hockey' meant exactly one thing — field hockey. Outside Dhaka, that certainty dissolves.

International hockey has two governing bodies. Field hockey runs under the FIH; ice hockey runs under the IIHF, and in North America the NHL operates by its own rules. The playing surfaces differ — one on grass or astroturf, the other on ice. Squad size, match duration, disciplinary systems, even the logic of scoring all differ. Yet in English both carry one name. In South Asia this confusion is almost absent, because in Bangladesh or Pakistan 'hockey' always means field hockey. Bangladesh's 1-0 loss to Pakistan at the 2026 Asia Cup remains our brightest memory; ice hockey casts no shadow there.

The trouble begins when the sport enters a data pipeline. An analysis system usually runs in two stages. Stage one deconstructs the article into information points, core viewpoints, entities involved and time sensitivity. Stage two analyses that broken-down material across nine dimensions — tactics, data, competition system, global context, rules and governance, management, risk, public narrative and industry transmission. If stage one is empty, stage two cannot produce analysis at all — it can only fill the framework's blank cells, not deliver a verdict. This is the first rule of analysis, and I learned it from my own mistakes.

For me, evidence means four layers read together: the notebook, the video, federation records and club trajectories. Deschamps did not press; I translated his silence into a 3-3-3-1 trigger, because the video caught that empty space. But if an article names no team, no date, no player, no model can take its place. Dropping a model into an empty cell means invention, and invention is not analysis.

In field hockey this verification matters even more, because its central measurement is absent from ice hockey. The penalty corner — injector, stopper, drag-flicker, first runner — has no equivalent in ice hockey. Every figure in the report I wrote on Belgium's 2026 Tokyo Olympic gold and Alexander Hendrickx's tournament-leading drag-flick haul came from field hockey's own corner logic. Ice hockey's power play and field hockey's penalty corner are not the same; putting both in one column makes the arithmetic entirely wrong. So without determining which sport sits beneath the word 'hockey', not a single number can be interpreted.

The greatest risk hides here. If someone forces an analysis out of an empty input, it will look complete while its foundation is zero. No team, no ranking, no head-to-head, no form curve — yet the report will carry headings for all nine dimensions. This false completeness is dangerous, because readers trust the headings and never see the emptiness inside.

A counterintuitive truth hides here too, one I came to understand slowly. Empty years, full archive — from 2026 to 2026 the Bangladesh Premier Division was not held; Maulana Bhasani stood silent. That silence was itself data. I digitised the 2026 Asia Cup and the Mohammedan years of Shahbaz Ahmed and Tahir Zaman; the federation's own records showed only 13 league editions in 27 years. Absence is sometimes the clearest statement. The same holds in analysis — an empty input is also a signal of the analysis system's own weakness. Between missing data and mislabelled data, the second is more damaging. Mislabelled data quietly produces wrong conclusions, and nobody notices.

I can say this now: as long as the data pipeline keeps merging two sports under the word 'hockey', field hockey's corner counts will blur into ice hockey's power-play counts. In any future match analysis, my first question will be a single one: which hockey is this? If there is no answer, the analysis must stop — because the number that fills an empty cell is not information, it is assumption.