Recalculate tenure from one preserved banzuke row per tournament. Keep former rank, participation, absence, demotion, return and active-career states instead of overwriting only the consecutive, career-total and competed-tenure totals.
Each new banzuke either extends or interrupts an active wrestler’s tenure. A latest total cannot reveal which tournaments were counted or when a historical row was repaired. Treat the rank path as the raw table and tenure summaries as derived releases.
Fields in the tournament-level source table
Store tournament ID, wrestler ID, rank, number, side, banzuke version, participation state, first and last competed day, official URL and retrieval time. A tournament not held is not a fictional banzuke row; it belongs in a separate schedule-state table.
Consecutive tenure derives from links between adjacent banzuke tournaments. Career total counts every row at the specified rank. Competed tenure uses the participation field. Give all three outputs the same raw-table hash and queue recalculation whenever that source changes.
Separate routine updates from corrections
Distinguish a newly announced banzuke, following an official display update, wrestler-ID or rank transcription repair, participation-state correction and tenure-definition change. Repairing a historical rank can alter every later consecutive-tenure value, so record the affected tournament interval.
Update ranking tables, profile articles and structured data together. Copyediting that changes no value does not advance the dataset. A definition change blocks direct comparison with the former release unless both definitions are shown.
Current public baseline
- 3 September 2026, v1.0: one banzuke row per tournament is the raw source for three tenure measures.
- 3 September 2026, v1.0: competed, absent, not held and active-continuing are distinct states.
- Next checkpoint: after the next banzuke announcement, append source rows and recalculate only tenures still in progress.
A monitoring check with no difference creates an audit entry but does not move the public modified date. When a correction changes an old ranking, readers retain access to the former position and reason.
A recalculation can reveal one forgotten basho
When a continuous line breaks, watch one bout on either side. Between the numbers for demotion and return, a wrestler was still testing solutions on the dohyo. The tenure history is not merely protection for totals; it is an index that keeps the middle of the career visible.
A Different Denominator Produces a Different Answer
Keep the population and denominator fixed to the stated period, rank, division and treatment of participation. Do not generalise the rate or average to another period or division. When an update changes the denominator, show the old and new conditions and counts instead of silently overwriting one as if they were identical.
Freeze the Question Before Reading the Result
The research question is whether the defined metric differs under the article’s stated conditions. “There should be a difference” is a hypothesis, not a conclusion entered before counting. Preserve a contrary result, and treat a small difference, heavy missingness or incomparable conditions as grounds to hold the conclusion.
Sources
- Japan Sumo Association banzuke, wrestler profiles and annual schedule (checked 3 September 2026).
Source links used for verification
Each link was used for the corresponding claim in this article; inclusion is not a blanket endorsement of the whole page.
Article verified:
- Japan Sumo Association banzuke tablePrimary source (as checked for this article)
- Japan Sumo Association wrestler searchPrimary source (as checked for this article)
- Japan Sumo Association annual schedulePrimary source (as checked for this article)