Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Aoi vs Sarah
Matches counted1,745
Win rate51.5%
Widest swing-6.2 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 |
|---|---|---|---|---|---|
| Rectangle | 57.3% | +5.9 | 0.115 | 164 | |
| Single wall | 56.2% | +4.8 | 0.180 | 176 | |
| Low fence | 55.2% | +3.8 | 0.297 | 172 | |
| Full fence | 52.2% | +0.7 | 0.838 | 184 | |
| Octagon | 51.1% | -0.3 | 0.928 | 176 | |
| Breakable full fence | 48.3% | -3.1 | 0.375 | 178 | |
| Fence and open | 48.2% | -3.3 | 0.477 | 110 | |
| Breakable half fence | 47.8% | -3.7 | 0.332 | 157 | |
| Half fence | 47.3% | -4.2 | 0.230 | 184 | |
| Open | 45.2% | -6.2 | 0.090 | 168 |
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 10.66 on 9 df, p 0.299
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.2%184 | +3.90.463 | +4.90.348 | +4.40.418 | -3.10.563 | -4.10.438 | +4.00.508 | +6.90.194 | -5.20.336 | +1.00.844 |
| Breakable full fence | -3.90.463 | 48.3%178 | +1.00.844 | +0.50.921 | -6.90.195 | -7.90.135 | +0.10.983 | +3.10.567 | -9.00.096 | -2.80.595 |
| Half fence | -4.90.348 | -1.00.844 | 47.3%184 | -0.50.928 | -8.00.134 | -9.00.089 | -0.90.881 | +2.00.701 | -10.00.061 | -3.90.465 |
| Breakable half fence | -4.40.418 | -0.50.921 | +0.50.928 | 47.8%157 | -7.50.176 | -8.50.122 | -0.40.947 | +2.50.647 | -9.60.087 | -3.40.540 |
| Low fence | +3.10.563 | +6.90.195 | +8.00.134 | +7.50.176 | 55.2%172 | -1.00.848 | +7.00.247 | +10.00.065 | -2.10.700 | +4.10.444 |
| Single wall | +4.10.438 | +7.90.135 | +9.00.089 | +8.50.122 | +1.00.848 | 56.2%176 | +8.10.183 | +11.00.041 | -1.10.843 | +5.10.336 |
| Fence and open | -4.00.508 | -0.10.983 | +0.90.881 | +0.40.947 | -7.00.247 | -8.10.183 | 48.2%110 | +2.90.630 | -9.10.137 | -3.00.627 |
| Open | -6.90.194 | -3.10.567 | -2.00.701 | -2.50.647 | -10.00.065 | -11.00.041 | -2.90.630 | 45.2%168 | -12.10.028 | -5.90.274 |
| Rectangle | +5.20.336 | +9.00.096 | +10.00.061 | +9.60.087 | +2.10.700 | +1.10.843 | +9.10.137 | +12.10.028 | 57.3%164 | +6.20.253 |
| Octagon | -1.00.844 | +2.80.595 | +3.90.465 | +3.40.540 | -4.10.444 | -5.10.336 | +3.00.627 | +5.90.274 | -6.20.253 | 51.1%176 |
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 counted1,596
Win rate51.1%
Widest swing-6.7 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 |
|---|---|---|---|---|---|
| Rectangle | 55.8% | +4.7 | 0.216 | 156 | |
| Low fence | 55.1% | +4.1 | 0.285 | 156 | |
| Single wall | 53.9% | +2.8 | 0.464 | 156 | |
| Full fence | 52.3% | +1.2 | 0.730 | 174 | |
| Octagon | 50.6% | -0.5 | 0.907 | 160 | |
| Breakable full fence | 49.4% | -1.7 | 0.652 | 160 | |
| Fence and open | 48.0% | -3.1 | 0.525 | 98 | |
| Breakable half fence | 47.7% | -3.4 | 0.383 | 153 | |
| Half fence | 47.2% | -3.8 | 0.302 | 163 | |
| Open | 44.4% | -6.7 | 0.084 | 151 |
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 7.76 on 9 df, p 0.558
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.3%174 | +2.90.593 | +5.10.353 | +4.60.408 | -2.80.607 | -1.60.779 | +4.30.492 | +7.90.154 | -3.50.528 | +1.70.760 |
| Breakable full fence | -2.90.593 | 49.4%160 | +2.10.701 | +1.70.769 | -5.80.306 | -4.50.427 | +1.40.825 | +5.00.377 | -6.40.255 | -1.20.823 |
| Half fence | -5.10.353 | -2.10.701 | 47.2%163 | -0.50.933 | -7.90.159 | -6.60.238 | -0.70.910 | +2.90.610 | -8.50.128 | -3.40.543 |
| Breakable half fence | -4.60.408 | -1.70.769 | +0.50.933 | 47.7%153 | -7.40.192 | -6.10.281 | -0.20.970 | +3.30.559 | -8.10.156 | -2.90.606 |
| Low fence | +2.80.607 | +5.80.306 | +7.90.159 | +7.40.192 | 55.1%156 | +1.30.820 | +7.20.265 | +10.80.059 | -0.60.909 | +4.50.423 |
| Single wall | +1.60.779 | +4.50.427 | +6.60.238 | +6.10.281 | -1.30.820 | 53.9%156 | +5.90.361 | +9.50.097 | -1.90.733 | +3.20.567 |
| Fence and open | -4.30.492 | -1.40.825 | +0.70.910 | +0.20.970 | -7.20.265 | -5.90.361 | 48.0%98 | +3.60.579 | -7.80.225 | -2.70.678 |
| Open | -7.90.154 | -5.00.377 | -2.90.610 | -3.30.559 | -10.80.059 | -9.50.097 | -3.60.579 | 44.4%151 | -11.40.046 | -6.20.270 |
| Rectangle | +3.50.528 | +6.40.255 | +8.50.128 | +8.10.156 | +0.60.909 | +1.90.733 | +7.80.225 | +11.40.046 | 55.8%156 | +5.20.360 |
| Octagon | -1.70.760 | +1.20.823 | +3.40.543 | +2.90.606 | -4.50.423 | -3.20.567 | +2.70.678 | +6.20.270 | -5.20.360 | 50.6%160 |
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 counted149
Win rate55.7%
Widest swing31.8 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 |
|---|---|---|---|---|---|
| Rectangle | 87.5% | +31.8 | 0.063 | 8 · | |
| Single wall | 75.0% | +19.3 | 0.062 | 20 · | |
| Low fence | 56.2% | +0.6 | 0.963 | 16 · | |
| Octagon | 56.2% | +0.6 | 0.963 | 16 · | |
| Open | 52.9% | -2.8 | 0.807 | 17 · | |
| Breakable half fence | 50.0% | -5.7 | 0.816 | 4 · | |
| Full fence | 50.0% | -5.7 | 0.707 | 10 · | |
| Fence and open | 50.0% | -5.7 | 0.678 | 12 · | |
| Half fence | 47.6% | -8.1 | 0.421 | 21 · | |
| Breakable full fence | 38.9% | -16.8 | 0.126 | 18 · |
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 9.31 on 9 df, p 0.409
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%10 | +11.10.569 | +2.40.901 | +0.01.000 | -6.20.756 | -25.00.171 | +0.01.000 | -2.90.883 | -37.50.094 | -6.20.756 |
| Breakable full fence | -11.10.569 | 38.9%18 | -8.70.584 | -11.10.683 | -17.40.311 | -36.10.024 | -11.10.547 | -14.10.404 | -48.60.022 | -17.40.311 |
| Half fence | -2.40.901 | +8.70.584 | 47.6%21 | -2.40.930 | -8.60.603 | -27.40.072 | -2.40.895 | -5.30.744 | -39.90.051 | -8.60.603 |
| Breakable half fence | -0.01.000 | +11.10.683 | +2.40.930 | 50.0%4 | -6.20.822 | -25.00.315 | +0.01.000 | -2.90.916 | -37.50.157 | -6.20.822 |
| Low fence | +6.20.756 | +17.40.311 | +8.60.603 | +6.20.822 | 56.2%16 | -18.80.236 | +6.20.743 | +3.30.849 | -31.20.126 | +0.01.000 |
| Single wall | +25.00.171 | +36.10.024 | +27.40.072 | +25.00.315 | +18.80.236 | 75.0%20 | +25.00.149 | +22.10.161 | -12.50.466 | +18.80.236 |
| Fence and open | -0.01.000 | +11.10.547 | +2.40.895 | -0.01.000 | -6.20.743 | -25.00.149 | 50.0%12 | -2.90.876 | -37.50.085 | -6.20.743 |
| Open | +2.90.883 | +14.10.404 | +5.30.744 | +2.90.916 | -3.30.849 | -22.10.161 | +2.90.876 | 52.9%17 | -34.60.093 | -3.30.849 |
| Rectangle | +37.50.094 | +48.60.022 | +39.90.051 | +37.50.157 | +31.20.126 | +12.50.466 | +37.50.085 | +34.60.093 | 87.5%8 | +31.20.126 |
| Octagon | +6.20.756 | +17.40.311 | +8.60.603 | +6.20.822 | -0.01.000 | -18.80.236 | +6.20.743 | +3.30.849 | -31.20.126 | 56.2%16 |
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.