Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Aoi vs Taka
Matches counted484
Win rate41.3%
Widest swing14.8 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 |
|---|---|---|---|---|---|
| Breakable full fence | 56.1% | +14.8 | 0.045 | 41 · | |
| Breakable half fence | 50.0% | +8.7 | 0.161 | 56 · | |
| Fence and open | 50.0% | +8.7 | 0.421 | 20 · | |
| Single wall | 45.5% | +4.1 | 0.509 | 55 · | |
| Half fence | 41.2% | -0.1 | 0.986 | 34 · | |
| Rectangle | 38.1% | -3.2 | 0.577 | 63 · | |
| Low fence | 37.2% | -4.1 | 0.533 | 51 · | |
| Octagon | 37.1% | -4.2 | 0.602 | 35 · | |
| Open | 37.1% | -4.2 | 0.469 | 62 · | |
| Full fence | 29.2% | -12.2 | 0.072 | 48 · |
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 10.67 on 9 df, p 0.299
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 29.2%48 | -26.90.010 | -12.00.259 | -20.80.031 | -8.10.394 | -16.30.089 | -20.80.101 | -7.90.383 | -8.90.326 | -8.00.444 |
| Breakable full fence | +26.90.010 | 56.1%41 | +14.90.198 | +6.10.552 | +18.90.071 | +10.70.302 | +6.10.654 | +19.00.058 | +18.00.071 | +19.00.099 |
| Half fence | +12.00.259 | -14.90.198 | 41.2%34 | -8.80.416 | +3.90.716 | -4.30.693 | -8.80.529 | +4.10.694 | +3.10.767 | +4.00.731 |
| Breakable half fence | +20.80.031 | -6.10.552 | +8.80.416 | 50.0%56 | +12.80.185 | +4.50.632 | +0.01.000 | +12.90.158 | +11.90.191 | +12.90.230 |
| Low fence | +8.10.394 | -18.90.071 | -3.90.716 | -12.80.185 | 37.2%51 | -8.20.392 | -12.80.326 | +0.10.986 | -0.80.927 | +0.10.992 |
| Single wall | +16.30.089 | -10.70.302 | +4.30.693 | -4.50.632 | +8.20.392 | 45.5%55 | -4.50.727 | +8.30.359 | +7.30.418 | +8.30.436 |
| Fence and open | +20.80.101 | -6.10.654 | +8.80.529 | -0.01.000 | +12.80.326 | +4.50.727 | 50.0%20 | +12.90.306 | +11.90.346 | +12.90.352 |
| Open | +7.90.383 | -19.00.058 | -4.10.694 | -12.90.158 | -0.10.986 | -8.30.359 | -12.90.306 | 37.1%62 | -1.00.908 | -0.00.996 |
| Rectangle | +8.90.326 | -18.00.071 | -3.10.767 | -11.90.191 | +0.80.927 | -7.30.418 | -11.90.346 | +1.00.908 | 38.1%63 | +1.00.926 |
| Octagon | +8.00.444 | -19.00.099 | -4.00.731 | -12.90.230 | -0.10.992 | -8.30.436 | -12.90.352 | +0.00.996 | -1.00.926 | 37.1%35 |
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 counted418
Win rate40.0%
Widest swing13.0 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 |
|---|---|---|---|---|---|
| Breakable full fence | 52.9% | +13.0 | 0.107 | 34 · | |
| Fence and open | 47.4% | +7.4 | 0.499 | 19 · | |
| Breakable half fence | 47.1% | +7.1 | 0.269 | 51 · | |
| Single wall | 43.8% | +3.8 | 0.568 | 48 · | |
| Half fence | 42.9% | +2.9 | 0.745 | 28 · | |
| Rectangle | 38.6% | -1.4 | 0.822 | 57 · | |
| Low fence | 38.1% | -1.9 | 0.796 | 42 · | |
| Octagon | 35.5% | -4.5 | 0.598 | 31 · | |
| Full fence | 31.8% | -8.1 | 0.244 | 44 · | |
| Open | 28.9% | -11.1 | 0.109 | 45 · |
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 8.14 on 9 df, p 0.520
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 31.8%44 | -21.10.060 | -11.00.342 | -15.20.131 | -6.30.542 | -11.90.239 | -15.60.239 | +2.90.764 | -6.80.481 | -3.70.740 |
| Breakable full fence | +21.10.060 | 52.9%34 | +10.10.429 | +5.90.595 | +14.80.196 | +9.20.412 | +5.60.697 | +24.10.030 | +14.30.182 | +17.50.157 |
| Half fence | +11.00.342 | -10.10.429 | 42.9%28 | -4.20.720 | +4.80.690 | -0.90.940 | -4.50.760 | +14.00.221 | +4.30.706 | +7.40.562 |
| Breakable half fence | +15.20.131 | -5.90.595 | +4.20.720 | 47.1%51 | +9.00.385 | +3.30.741 | -0.30.982 | +18.20.068 | +8.50.375 | +11.60.304 |
| Low fence | +6.30.542 | -14.80.196 | -4.80.690 | -9.00.385 | 38.1%42 | -5.70.586 | -9.30.495 | +9.20.363 | -0.50.960 | +2.60.819 |
| Single wall | +11.90.239 | -9.20.412 | +0.90.940 | -3.30.741 | +5.70.586 | 43.8%48 | -3.60.788 | +14.90.137 | +5.20.593 | +8.30.465 |
| Fence and open | +15.60.239 | -5.60.697 | +4.50.760 | +0.30.982 | +9.30.495 | +3.60.788 | 47.4%19 | +18.50.155 | +8.80.500 | +11.90.405 |
| Open | -2.90.764 | -24.10.030 | -14.00.221 | -18.20.068 | -9.20.363 | -14.90.137 | -18.50.155 | 28.9%45 | -9.70.305 | -6.60.543 |
| Rectangle | +6.80.481 | -14.30.182 | -4.30.706 | -8.50.375 | +0.50.960 | -5.20.593 | -8.80.500 | +9.70.305 | 38.6%57 | +3.10.773 |
| Octagon | +3.70.740 | -17.50.157 | -7.40.562 | -11.60.304 | -2.60.819 | -8.30.465 | -11.90.405 | +6.60.543 | -3.10.773 | 35.5%31 |
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 counted66
Win rate50.0%
Widest swing50.0 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 |
|---|---|---|---|---|---|
| Fence and open | 100.0% | +50.0 | 0.314 | 1 · | |
| Breakable half fence | 80.0% | +30.0 | 0.163 | 5 · | |
| Breakable full fence | 71.4% | +21.4 | 0.230 | 7 · | |
| Open | 58.8% | +8.8 | 0.398 | 17 · | |
| Single wall | 57.1% | +7.1 | 0.689 | 7 · | |
| Octagon | 50.0% | +0.0 | 1.000 | 4 · | |
| Half fence | 33.3% | -16.7 | 0.392 | 6 · | |
| Low fence | 33.3% | -16.7 | 0.282 | 9 · | |
| Rectangle | 33.3% | -16.7 | 0.392 | 6 · | |
| Full fence | 0.0% | -50.0 | 0.039 | 4 · |
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 11.09 on 9 df, p 0.270
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 0.0%4 | -71.40.022 | -33.30.197 | -80.00.016 | -33.30.188 | -57.10.058 | -100.00.025 | -58.80.034 | -33.30.197 | -50.00.102 |
| Breakable full fence | +71.40.022 | 71.4%7 | +38.10.170 | -8.60.735 | +38.10.131 | +14.30.577 | -28.60.537 | +12.60.562 | +38.10.170 | +21.40.477 |
| Half fence | +33.30.197 | -38.10.170 | 33.3%6 | -46.70.122 | +0.01.000 | -23.80.391 | -66.70.212 | -25.50.283 | +0.01.000 | -16.70.598 |
| Breakable half fence | +80.00.016 | +8.60.735 | +46.70.122 | 80.0%5 | +46.70.094 | +22.90.408 | -20.00.624 | +21.20.387 | +46.70.122 | +30.00.343 |
| Low fence | +33.30.188 | -38.10.131 | -0.01.000 | -46.70.094 | 33.3%9 | -23.80.341 | -66.70.197 | -25.50.216 | +0.01.000 | -16.70.569 |
| Single wall | +57.10.058 | -14.30.577 | +23.80.391 | -22.90.408 | +23.80.341 | 57.1%7 | -42.90.408 | -1.70.939 | +23.80.391 | +7.10.819 |
| Fence and open | +100.00.025 | +28.60.537 | +66.70.212 | +20.00.624 | +66.70.197 | +42.90.408 | 100.0%1 | +41.20.412 | +66.70.212 | +50.00.361 |
| Open | +58.80.034 | -12.60.562 | +25.50.283 | -21.20.387 | +25.50.216 | +1.70.939 | -41.20.412 | 58.8%17 | +25.50.283 | +8.80.748 |
| Rectangle | +33.30.197 | -38.10.170 | -0.01.000 | -46.70.122 | -0.01.000 | -23.80.391 | -66.70.212 | -25.50.283 | 33.3%6 | -16.70.598 |
| Octagon | +50.00.102 | -21.40.477 | +16.70.598 | -30.00.343 | +16.70.569 | -7.10.819 | -50.00.361 | -8.80.748 | +16.70.598 | 50.0%4 |
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.