Analysis22
Being strong against one rikishi can mislead
Conclusion
From Natsu 2003 onward, only 46.7% of all-division bouts involved opponents who had met before; 53.3% were first meetings. A handful of bouts rarely separates a durable matchup effect from chance, so head-to-head history needs to be read alongside sample size and current strength.
How to read this number
From Natsu 2003 onward, only 46.7% of bouts across all divisions involved opponents who had met before. More than half were first meetings, which undercuts the premise of talking about matchups at all.
Even where a history exists, it is usually a handful of bouts. Calling 2-1 a good matchup is indistinguishable from tossing a coin three times and seeing two heads. A real matchup effect would have to survive adjustment for strength and form across a meaningful number of bouts.
The head-to-head pages on this site show career records, but a winning record there does not necessarily mean a favourable matchup. It is worth reading alongside the sample size and each rikishi's current strength.
Methodology
- Period
- 2003 Natsu Basho〜2026 Natsu Basho
- Sample
- Bouts of all classes are always determined based on the match history before the first match.
- Sample size
- 336,981 bouts
- Calculation
- The number of bouts in the past where the same combination was 1 or more was 157,330 bouts, 46.7%
- rest
- First meeting: 179,651 bouts, 53.3%
How to read this
If you have played against someone even once, you have experience. The percentage of matches with enough matches to discuss compatibility is even smaller.