Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Aoi vs Dural
Matches counted241
Win rate49.4%
Widest swing-32.7 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Aoi's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Open | 66.7% | +17.3 | 0.127 | 18 · | |
| Half fence | 64.5% | +15.1 | 0.071 | 31 · | |
| Rectangle | 57.7% | +8.3 | 0.369 | 26 · | |
| Low fence | 55.6% | +6.2 | 0.586 | 18 · | |
| Full fence | 54.5% | +5.2 | 0.611 | 22 · | |
| Single wall | 48.4% | -1.0 | 0.906 | 31 · | |
| Breakable full fence | 47.8% | -1.6 | 0.876 | 23 · | |
| Octagon | 33.3% | -16.1 | 0.124 | 21 · | |
| Breakable half fence | 26.7% | -22.7 | 0.008 | 30 · | |
| Fence and open | 16.7% | -32.7 | 0.105 | 6 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Aoi's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 17.18 on 9 df, p 0.046
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 54.5%22 | +6.70.652 | -10.00.465 | +27.90.041 | -1.00.949 | +6.20.659 | +37.90.099 | -12.10.436 | -3.10.827 | +21.20.161 |
| Breakable full fence | -6.70.652 | 47.8%23 | -16.70.220 | +21.20.111 | -7.70.623 | -0.60.967 | +31.20.168 | -18.80.228 | -9.90.490 | +14.50.329 |
| Half fence | +10.00.465 | +16.70.220 | 64.5%31 | +37.90.003 | +9.00.535 | +16.10.200 | +47.90.030 | -2.10.879 | +6.80.598 | +31.20.027 |
| Breakable half fence | -27.90.041 | -21.20.111 | -37.90.003 | 26.7%30 | -28.90.045 | -21.70.080 | +10.00.606 | -40.00.007 | -31.00.019 | -6.70.607 |
| Low fence | +1.00.949 | +7.70.623 | -9.00.535 | +28.90.045 | 55.6%18 | +7.20.628 | +38.90.098 | -11.10.494 | -2.10.888 | +22.20.163 |
| Single wall | -6.20.659 | +0.60.967 | -16.10.200 | +21.70.080 | -7.20.628 | 48.4%31 | +31.70.151 | -18.30.215 | -9.30.483 | +15.10.281 |
| Fence and open | -37.90.099 | -31.20.168 | -47.90.030 | -10.00.606 | -38.90.098 | -31.70.151 | 16.7%6 | -50.00.033 | -41.00.070 | -16.70.430 |
| Open | +12.10.436 | +18.80.228 | +2.10.879 | +40.00.007 | +11.10.494 | +18.30.215 | +50.00.033 | 66.7%18 | +9.00.548 | +33.30.038 |
| Rectangle | +3.10.827 | +9.90.490 | -6.80.598 | +31.00.019 | +2.10.888 | +9.30.483 | +41.00.070 | -9.00.548 | 57.7%26 | +24.40.096 |
| Octagon | -21.20.161 | -14.50.329 | -31.20.027 | +6.70.607 | -22.20.163 | -15.10.281 | +16.70.430 | -33.30.038 | -24.40.096 | 33.3%21 |
Better on the row's stage — deeper at 5 and 10 pointsBetter on the column's stagep < 0.05 on its ownStill there after correcting for how many were tested· fewer than 80 decided matches on one side of the comparison
How significance is decided
- Each gap is two win rates compared with a pooled two-proportion z test, over decided matches only.
- The 95% interval is Newcombe's hybrid score method, which keeps its shape near 0% and 100% where a symmetric interval does not.
- q is p corrected for how many tests were asked at once, by Benjamini–Hochberg at a 5% false discovery rate. A figure is called real on q, never on p alone.
- Standard errors are widened by a measured design effect, because one player appears in many matches and those matches are not independent of each other.
- Stages are assigned at random, so a gap between them is the stage's own effect and not a difference in who chose to play where.
Matches counted223
Win rate46.2%
Widest swing-29.5 pts
Survive correction1
Period2026/03/26 – 2026/09/23
Where it was fought
Aoi's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Open | 64.7% | +18.5 | 0.111 | 17 · | |
| Full fence | 57.9% | +11.7 | 0.285 | 19 · | |
| Rectangle | 57.7% | +11.5 | 0.211 | 26 · | |
| Low fence | 55.6% | +9.4 | 0.406 | 18 · | |
| Half fence | 52.2% | +6.0 | 0.543 | 23 · | |
| Single wall | 44.8% | -1.4 | 0.875 | 29 · | |
| Breakable full fence | 42.9% | -3.3 | 0.748 | 21 · | |
| Octagon | 33.3% | -12.9 | 0.214 | 21 · | |
| Breakable half fence | 21.4% | -24.8 | 0.005 | 28 · | |
| Fence and open | 16.7% | -29.5 | 0.141 | 6 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Aoi's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 16.27 on 9 df, p 0.061
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.9%19 | +15.00.342 | +5.70.711 | +36.50.011 | +2.30.886 | +13.10.376 | +41.20.078 | -6.80.676 | +0.20.989 | +24.60.119 |
| Breakable full fence | -15.00.342 | 42.9%21 | -9.30.537 | +21.40.107 | -12.70.429 | -2.00.890 | +26.20.241 | -21.90.180 | -14.80.312 | +9.50.525 |
| Half fence | -5.70.711 | +9.30.537 | 52.2%23 | +30.70.022 | -3.40.829 | +7.30.598 | +35.50.119 | -12.50.428 | -5.50.698 | +18.80.208 |
| Breakable half fence | -36.50.011 | -21.40.107 | -30.70.022 | 21.4%28 | -34.10.018 | -23.40.061 | +4.80.793 | -43.30.004 | -36.30.006 | -11.90.350 |
| Low fence | -2.30.886 | +12.70.429 | +3.40.829 | +34.10.018 | 55.6%18 | +10.70.474 | +38.90.098 | -9.20.581 | -2.10.888 | +22.20.163 |
| Single wall | -13.10.376 | +2.00.890 | -7.30.598 | +23.40.061 | -10.70.474 | 44.8%29 | +28.20.200 | -19.90.193 | -12.90.341 | +11.50.413 |
| Fence and open | -41.20.078 | -26.20.241 | -35.50.119 | -4.80.793 | -38.90.098 | -28.20.200 | 16.7%6 | -48.00.043 | -41.00.070 | -16.70.430 |
| Open | +6.80.676 | +21.90.180 | +12.50.428 | +43.30.004 | +9.20.581 | +19.90.193 | +48.00.043 | 64.7%17 | +7.00.646 | +31.40.054 |
| Rectangle | -0.20.989 | +14.80.312 | +5.50.698 | +36.30.006 | +2.10.888 | +12.90.341 | +41.00.070 | -7.00.646 | 57.7%26 | +24.40.096 |
| Octagon | -24.60.119 | -9.50.525 | -18.80.208 | +11.90.350 | -22.20.163 | -11.50.413 | +16.70.430 | -31.40.054 | -24.40.096 | 33.3%21 |
Better on the row's stage — deeper at 5 and 10 pointsBetter on the column's stagep < 0.05 on its ownStill there after correcting for how many were tested· fewer than 80 decided matches on one side of the comparison
How significance is decided
- Each gap is two win rates compared with a pooled two-proportion z test, over decided matches only.
- The 95% interval is Newcombe's hybrid score method, which keeps its shape near 0% and 100% where a symmetric interval does not.
- q is p corrected for how many tests were asked at once, by Benjamini–Hochberg at a 5% false discovery rate. A figure is called real on q, never on p alone.
- Standard errors are widened by a measured design effect, because one player appears in many matches and those matches are not independent of each other.
- Stages are assigned at random, so a gap between them is the stage's own effect and not a difference in who chose to play where.
Matches counted18
Win rate88.9%
Widest swing-55.6 pts
Survive correction1
Period2026/03/26 – 2026/09/23
Where it was fought
Aoi's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable full fence | 100.0% | +11.1 | 0.596 | 2 · | |
| Breakable half fence | 100.0% | +11.1 | 0.596 | 2 · | |
| Half fence | 100.0% | +11.1 | 0.180 | 8 · | |
| Open | 100.0% | +11.1 | 0.716 | 1 · | |
| Single wall | 100.0% | +11.1 | 0.596 | 2 · | |
| Full fence | 33.3% | -55.6 | <0.001 | 3 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Aoi's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 11.25 on 5 df, p 0.047
| Stage | Full | Break. full | Half | Break. half | Wall | Open |
|---|---|---|---|---|---|---|
| Full fence | 33.3%3 | -66.70.136 | -66.70.011 | -66.70.136 | -66.70.136 | -66.70.248 |
| Breakable full fence | +66.70.136 | 100.0%2 | +0.01.000 | +0.01.000 | +0.01.000 | +0.01.000 |
| Half fence | +66.70.011 | -0.01.000 | 100.0%8 | +0.01.000 | +0.01.000 | +0.01.000 |
| Breakable half fence | +66.70.136 | -0.01.000 | -0.01.000 | 100.0%2 | +0.01.000 | +0.01.000 |
| Single wall | +66.70.136 | -0.01.000 | -0.01.000 | -0.01.000 | 100.0%2 | +0.01.000 |
| Open | +66.70.248 | -0.01.000 | -0.01.000 | -0.01.000 | -0.01.000 | 100.0%1 |
Better on the row's stage — deeper at 5 and 10 pointsBetter on the column's stagep < 0.05 on its ownStill there after correcting for how many were tested· fewer than 80 decided matches on one side of the comparison
How significance is decided
- Each gap is two win rates compared with a pooled two-proportion z test, over decided matches only.
- The 95% interval is Newcombe's hybrid score method, which keeps its shape near 0% and 100% where a symmetric interval does not.
- q is p corrected for how many tests were asked at once, by Benjamini–Hochberg at a 5% false discovery rate. A figure is called real on q, never on p alone.
- Standard errors are widened by a measured design effect, because one player appears in many matches and those matches are not independent of each other.
- Stages are assigned at random, so a gap between them is the stage's own effect and not a difference in who chose to play where.