| Roshi vs Steve | 35–79 | 30.70% |
| Roshi vs Kazuya | 22–84 | 20.75% |
| Roshi vs Jin | 23–74 | 23.71% |
| Roshi vs Paul | 23–47 | 32.86% |
| Roshi vs Yoshimitsu | 18–48 | 27.27% |
| Roshi vs King | 13–48 | 21.31% |
| Roshi vs Reina | 15–44–1 | 25.42% |
| Roshi vs Hwoarang | 14–39 | 26.42% |
| Roshi vs Nina | 19–26 | 42.22% |
| Roshi vs Asuka | 15–26 | 36.59% |
| Roshi vs Alisa | 6–34 | 15.00% |
| Roshi vs Armor King | 2–38 | 5.00% |
| Roshi vs Bryan | 13–25 | 34.21% |
| Roshi vs Lili | 3–27 | 10.00% |
| Roshi vs Miary Zo | 8–21 | 27.59% |
| Roshi vs Eddy | 4–24 | 14.29% |
| Roshi vs Heihachi | 9–16 | 36.00% |
| Roshi vs Xiaoyu | 8–16 | 33.33% |
| Roshi vs Devil Jin | 3–21 | 12.50% |
| Roshi vs Dragunov | 9–13 | 40.91% |
| Roshi vs Lars | 5–16 | 23.81% |
| Roshi vs Panda | 6–14 | 30.00% |
| Roshi vs Law | 3–16 | 15.79% |
| Roshi vs Claudio | 5–11 | 31.25% |
| Roshi vs Lee | 0–15 | 0.00% |
| Roshi vs Lidia | 4–11 | 26.67% |
| Roshi vs Jun | 6–7 | 46.15% |
| Roshi vs Kuma | 4–8 | 33.33% |
| Roshi vs Azucena | 4–7 | 36.36% |
| Roshi vs Clive | 1–10 | 9.09% |
| Roshi vs Leo | 2–7 | 22.22% |
| Roshi vs Feng | 2–6 | 25.00% |
| Roshi vs Zafina | 2–5 | 28.57% |
| Roshi vs Victor | 1–3 | 25.00% |
| Roshi vs Fahkumram | 1–2 | 33.33% |
| Roshi vs Jack-8 | 1–1 | 50.00% |
| Roshi vs Leroy | 0–2 | 0.00% |
| Roshi vs Anna | 0–2 | 0.00% |
| Roshi vs Shaheen | 1–0 | 100.00% |
| Roshi vs Raven | 1–0 | 100.00% |
Limitations
This data is often requested to give insight into which characters you have more trouble with than others, but it is not particularly helpful for that. The main issue is that it is heavily skewed by how strong the opponents you play are.
For example, this data suggests my worst matchup is clearly vs Reina, but that's just because most of those games are vs Yagami.
There is a way to account for this being worked on. The central idea is to assign each matchup a rating vs you which adjusts based on the result, much like the regular rating but also based on the rating of each player. With this, it would give a better summary of how well you perform vs each character.
In the meantime, this page is here to present the data as requested.