| MelanieOsakie vs King | 15–6 | 71.43% |
| MelanieOsakie vs Heihachi | 9–8 | 52.94% |
| MelanieOsakie vs Reina | 8–8 | 50.00% |
| MelanieOsakie vs Jin | 5–10 | 33.33% |
| MelanieOsakie vs Lili | 9–1 | 90.00% |
| MelanieOsakie vs Dragunov | 5–4 | 55.56% |
| MelanieOsakie vs Lidia | 5–3 | 62.50% |
| MelanieOsakie vs Paul | 4–3 | 57.14% |
| MelanieOsakie vs Yoshimitsu | 3–4 | 42.86% |
| MelanieOsakie vs Jack-8 | 6–1 | 85.71% |
| MelanieOsakie vs Hwoarang | 4–2 | 66.67% |
| MelanieOsakie vs Jun | 2–4 | 33.33% |
| MelanieOsakie vs Alisa | 2–3 | 40.00% |
| MelanieOsakie vs Lee | 3–2 | 60.00% |
| MelanieOsakie vs Steve | 1–3 | 25.00% |
| MelanieOsakie vs Devil Jin | 1–3 | 25.00% |
| MelanieOsakie vs Feng | 2–2 | 50.00% |
| MelanieOsakie vs Shaheen | 0–4 | 0.00% |
| MelanieOsakie vs Eddy | 2–2 | 50.00% |
| MelanieOsakie vs Xiaoyu | 2–1 | 66.67% |
| MelanieOsakie vs Bryan | 2–1 | 66.67% |
| MelanieOsakie vs Lars | 1–2 | 33.33% |
| MelanieOsakie vs Leroy | 1–2 | 33.33% |
| MelanieOsakie vs Victor | 2–1 | 66.67% |
| MelanieOsakie vs Raven | 1–2 | 33.33% |
| MelanieOsakie vs Asuka | 1–1 | 50.00% |
| MelanieOsakie vs Leo | 1–1 | 50.00% |
| MelanieOsakie vs Nina | 0–2 | 0.00% |
| MelanieOsakie vs Kuma | 0–2 | 0.00% |
| MelanieOsakie vs Law | 0–1 | 0.00% |
| MelanieOsakie vs Kazuya | 0–1 | 0.00% |
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.