Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Sarah vs Shun
Matches counted1,902
Win rate49.0%
Widest swing8.9 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 57.9% | +8.9 | 0.026 | 145 | |
| Single wall | 54.8% | +5.8 | 0.087 | 197 | |
| Octagon | 53.0% | +4.0 | 0.215 | 213 | |
| Fence and open | 51.1% | +2.1 | 0.689 | 90 | |
| Breakable full fence | 48.6% | -0.5 | 0.899 | 179 | |
| Full fence | 48.0% | -1.1 | 0.736 | 225 | |
| Breakable half fence | 47.2% | -1.9 | 0.581 | 195 | |
| Open | 47.0% | -2.0 | 0.582 | 168 | |
| Rectangle | 46.8% | -2.3 | 0.513 | 186 | |
| Low fence | 42.3% | -6.7 | 0.039 | 208 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.46 on 9 df, p 0.143
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.0%225 | -0.60.904 | -9.90.062 | +0.80.867 | +5.70.234 | -6.80.162 | -3.10.618 | +1.00.848 | +1.20.804 | -5.00.291 |
| Breakable full fence | +0.60.904 | 48.6%179 | -9.30.094 | +1.40.783 | +6.30.215 | -6.20.228 | -2.50.698 | +1.60.768 | +1.80.727 | -4.50.380 |
| Half fence | +9.90.062 | +9.30.094 | 57.9%145 | +10.80.050 | +15.60.004 | +3.10.567 | +6.80.307 | +10.90.054 | +11.20.044 | +4.90.362 |
| Breakable half fence | -0.80.867 | -1.40.783 | -10.80.050 | 47.2%195 | +4.90.326 | -7.60.130 | -3.90.537 | +0.20.976 | +0.40.937 | -5.90.236 |
| Low fence | -5.70.234 | -6.30.215 | -15.60.004 | -4.90.326 | 42.3%208 | -12.50.012 | -8.80.161 | -4.70.360 | -4.50.373 | -10.70.027 |
| Single wall | +6.80.162 | +6.20.228 | -3.10.567 | +7.60.130 | +12.50.012 | 54.8%197 | +3.70.559 | +7.80.137 | +8.10.115 | +1.80.719 |
| Fence and open | +3.10.618 | +2.50.698 | -6.80.307 | +3.90.537 | +8.80.161 | -3.70.559 | 51.1%90 | +4.10.531 | +4.30.499 | -1.90.757 |
| Open | -1.00.848 | -1.60.768 | -10.90.054 | -0.20.976 | +4.70.360 | -7.80.137 | -4.10.531 | 47.0%168 | +0.20.963 | -6.00.243 |
| Rectangle | -1.20.804 | -1.80.727 | -11.20.044 | -0.40.937 | +4.50.373 | -8.10.115 | -4.30.499 | -0.20.963 | 46.8%186 | -6.30.211 |
| Octagon | +5.00.291 | +4.50.380 | -4.90.362 | +5.90.236 | +10.70.027 | -1.80.719 | +1.90.757 | +6.00.243 | +6.30.211 | 53.0%213 |
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 counted1,622
Win rate49.7%
Widest swing7.8 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Fence and open | 57.5% | +7.8 | 0.170 | 73 · | |
| Half fence | 55.5% | +5.8 | 0.191 | 119 | |
| Octagon | 55.0% | +5.3 | 0.132 | 182 | |
| Breakable full fence | 51.0% | +1.3 | 0.738 | 155 | |
| Single wall | 50.9% | +1.2 | 0.743 | 169 | |
| Rectangle | 49.3% | -0.3 | 0.928 | 152 | |
| Breakable half fence | 47.9% | -1.8 | 0.628 | 169 | |
| Full fence | 47.3% | -2.4 | 0.462 | 201 | |
| Open | 46.4% | -3.3 | 0.419 | 140 | |
| Low fence | 44.9% | -4.8 | 0.179 | 178 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.42 on 9 df, p 0.492
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 47.3%201 | -3.70.488 | -8.20.156 | -0.70.898 | +2.30.651 | -3.60.487 | -10.30.133 | +0.80.879 | -2.10.699 | -7.70.133 |
| Breakable full fence | +3.70.488 | 51.0%155 | -4.50.460 | +3.00.585 | +6.00.272 | +0.10.988 | -6.60.354 | +4.50.436 | +1.60.776 | -4.00.466 |
| Half fence | +8.20.156 | +4.50.460 | 55.5%119 | +7.50.208 | +10.50.076 | +4.60.444 | -2.10.779 | +9.00.147 | +6.10.317 | +0.50.930 |
| Breakable half fence | +0.70.898 | -3.00.585 | -7.50.208 | 47.9%169 | +3.00.577 | -3.00.586 | -9.60.170 | +1.50.793 | -1.40.800 | -7.00.189 |
| Low fence | -2.30.651 | -6.00.272 | -10.50.076 | -3.00.577 | 44.9%178 | -6.00.268 | -12.60.070 | -1.50.792 | -4.40.425 | -10.00.058 |
| Single wall | +3.60.487 | -0.10.988 | -4.60.444 | +3.00.586 | +6.00.268 | 50.9%169 | -6.60.342 | +4.50.435 | +1.60.782 | -4.10.447 |
| Fence and open | +10.30.133 | +6.60.354 | +2.10.779 | +9.60.170 | +12.60.070 | +6.60.342 | 57.5%73 | +11.10.124 | +8.20.250 | +2.60.707 |
| Open | -0.80.879 | -4.50.436 | -9.00.147 | -1.50.793 | +1.50.792 | -4.50.435 | -11.10.124 | 46.4%140 | -2.90.619 | -8.50.130 |
| Rectangle | +2.10.699 | -1.60.776 | -6.10.317 | +1.40.800 | +4.40.425 | -1.60.782 | -8.20.250 | +2.90.619 | 49.3%152 | -5.60.307 |
| Octagon | +7.70.133 | +4.00.466 | -0.50.930 | +7.00.189 | +10.00.058 | +4.10.447 | -2.60.707 | +8.50.130 | +5.60.307 | 55.0%182 |
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 counted280
Win rate45.4%
Widest swing33.2 pts
Survive correction1
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Single wall | 78.6% | +33.2 | <0.001 | 28 · | |
| Half fence | 69.2% | +23.9 | 0.010 | 26 · | |
| Full fence | 54.2% | +8.8 | 0.365 | 24 · | |
| Open | 50.0% | +4.6 | 0.603 | 28 · | |
| Breakable half fence | 42.3% | -3.0 | 0.743 | 26 · | |
| Octagon | 41.9% | -3.4 | 0.685 | 31 · | |
| Rectangle | 35.3% | -10.1 | 0.209 | 34 · | |
| Breakable full fence | 33.3% | -12.0 | 0.216 | 24 · | |
| Low fence | 26.7% | -18.7 | 0.030 | 30 · | |
| Fence and open | 23.5% | -21.8 | 0.062 | 17 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 29.88 on 9 df, p <0.001
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 54.2%24 | +20.80.146 | -15.10.273 | +11.90.402 | +27.50.039 | -24.40.061 | +30.60.050 | +4.20.764 | +18.90.153 | +12.20.368 |
| Breakable full fence | -20.80.146 | 33.3%24 | -35.90.011 | -9.00.514 | +6.70.594 | -45.2<0.001 | +9.80.497 | -16.70.225 | -2.00.877 | -8.60.515 |
| Half fence | +15.10.273 | +35.90.011 | 69.2%26 | +26.90.051 | +42.60.001 | -9.30.434 | +45.70.003 | +19.20.151 | +33.90.009 | +27.30.039 |
| Breakable half fence | -11.90.402 | +9.00.514 | -26.90.051 | 42.3%26 | +15.60.218 | -36.30.006 | +18.80.207 | -7.70.571 | +7.00.580 | +0.40.977 |
| Low fence | -27.50.039 | -6.70.594 | -42.60.001 | -15.60.218 | 26.7%30 | -51.9<0.001 | +3.10.813 | -23.30.067 | -8.60.457 | -15.30.210 |
| Single wall | +24.40.061 | +45.2<0.001 | +9.30.434 | +36.30.006 | +51.9<0.001 | 78.6%28 | +55.0<0.001 | +28.60.026 | +43.3<0.001 | +36.60.004 |
| Fence and open | -30.60.050 | -9.80.497 | -45.70.003 | -18.80.207 | -3.10.813 | -55.0<0.001 | 23.5%17 | -26.50.079 | -11.80.393 | -18.40.202 |
| Open | -4.20.764 | +16.70.225 | -19.20.151 | +7.70.571 | +23.30.067 | -28.60.026 | +26.50.079 | 50.0%28 | +14.70.243 | +8.10.535 |
| Rectangle | -18.90.153 | +2.00.877 | -33.90.009 | -7.00.580 | +8.60.457 | -43.3<0.001 | +11.80.393 | -14.70.243 | 35.3%34 | -6.70.583 |
| Octagon | -12.20.368 | +8.60.515 | -27.30.039 | -0.40.977 | +15.30.210 | -36.60.004 | +18.40.202 | -8.10.535 | +6.70.583 | 41.9%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.