Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Sarah vs Vanessa
Matches counted1,637
Win rate51.6%
Widest swing7.0 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 |
|---|---|---|---|---|---|
| Breakable half fence | 58.6% | +7.0 | 0.051 | 174 | |
| Octagon | 56.6% | +5.0 | 0.173 | 166 | |
| Half fence | 53.4% | +1.8 | 0.631 | 161 | |
| Breakable full fence | 53.1% | +1.5 | 0.688 | 160 | |
| Rectangle | 50.6% | -1.0 | 0.791 | 160 | |
| Full fence | 50.3% | -1.3 | 0.732 | 157 | |
| Open | 49.7% | -1.9 | 0.615 | 151 | |
| Fence and open | 48.0% | -3.7 | 0.521 | 73 · | |
| Single wall | 47.2% | -4.4 | 0.238 | 163 | |
| Low fence | 46.9% | -4.8 | 0.182 | 175 |
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 9.07 on 9 df, p 0.431
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.3%157 | -2.80.617 | -3.10.580 | -8.30.130 | +3.50.529 | +3.10.582 | +2.40.738 | +0.70.909 | -0.30.956 | -6.30.256 |
| Breakable full fence | +2.80.617 | 53.1%160 | -0.30.958 | -5.50.312 | +6.30.252 | +5.90.290 | +5.20.463 | +3.50.542 | +2.50.654 | -3.50.525 |
| Half fence | +3.10.580 | +0.30.958 | 53.4%161 | -5.20.338 | +6.60.230 | +6.20.266 | +5.50.438 | +3.80.508 | +2.80.617 | -3.20.560 |
| Breakable half fence | +8.30.130 | +5.50.312 | +5.20.338 | 58.6%174 | +11.80.028 | +11.40.036 | +10.70.123 | +8.90.106 | +8.00.142 | +2.00.710 |
| Low fence | -3.50.529 | -6.30.252 | -6.60.230 | -11.80.028 | 46.9%175 | -0.40.944 | -1.10.876 | -2.80.612 | -3.80.491 | -9.80.071 |
| Single wall | -3.10.582 | -5.90.290 | -6.20.266 | -11.40.036 | +0.40.944 | 47.2%163 | -0.70.920 | -2.40.667 | -3.40.543 | -9.40.088 |
| Fence and open | -2.40.738 | -5.20.463 | -5.50.438 | -10.70.123 | +1.10.876 | +0.70.920 | 48.0%73 | -1.70.809 | -2.70.704 | -8.70.215 |
| Open | -0.70.909 | -3.50.542 | -3.80.508 | -8.90.106 | +2.80.612 | +2.40.667 | +1.70.809 | 49.7%151 | -0.90.866 | -7.00.215 |
| Rectangle | +0.30.956 | -2.50.654 | -2.80.617 | -8.00.142 | +3.80.491 | +3.40.543 | +2.70.704 | +0.90.866 | 50.6%160 | -6.00.277 |
| Octagon | +6.30.256 | +3.50.525 | +3.20.560 | -2.00.710 | +9.80.071 | +9.40.088 | +8.70.215 | +7.00.215 | +6.00.277 | 56.6%166 |
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 counted1,344
Win rate51.0%
Widest swing-6.0 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 |
|---|---|---|---|---|---|
| Breakable half fence | 56.9% | +5.8 | 0.137 | 146 | |
| Octagon | 56.3% | +5.3 | 0.182 | 142 | |
| Breakable full fence | 53.3% | +2.2 | 0.604 | 122 | |
| Rectangle | 52.0% | +0.9 | 0.826 | 127 | |
| Fence and open | 51.7% | +0.6 | 0.921 | 60 · | |
| Half fence | 50.8% | -0.3 | 0.951 | 128 | |
| Full fence | 48.9% | -2.1 | 0.598 | 135 | |
| Open | 47.9% | -3.1 | 0.473 | 121 | |
| Single wall | 47.9% | -3.1 | 0.427 | 144 | |
| Low fence | 45.0% | -6.0 | 0.131 | 140 |
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 7.19 on 9 df, p 0.617
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.9%135 | -4.40.482 | -1.90.759 | -8.00.182 | +3.90.518 | +1.00.871 | -2.80.720 | +1.00.879 | -3.10.618 | -7.50.215 |
| Breakable full fence | +4.40.482 | 53.3%122 | +2.50.693 | -3.60.558 | +8.30.181 | +5.40.383 | +1.60.838 | +5.30.405 | +1.30.836 | -3.10.618 |
| Half fence | +1.90.759 | -2.50.693 | 50.8%128 | -6.10.315 | +5.80.344 | +2.90.637 | -0.90.910 | +2.90.653 | -1.20.850 | -5.60.361 |
| Breakable half fence | +8.00.182 | +3.60.558 | +6.10.315 | 56.9%146 | +11.80.045 | +8.90.128 | +5.20.497 | +8.90.146 | +4.90.419 | +0.50.930 |
| Low fence | -3.90.518 | -8.30.181 | -5.80.344 | -11.80.045 | 45.0%140 | -2.90.622 | -6.70.387 | -2.90.636 | -7.00.255 | -11.30.057 |
| Single wall | -1.00.871 | -5.40.383 | -2.90.637 | -8.90.128 | +2.90.622 | 47.9%144 | -3.80.625 | -0.00.998 | -4.00.506 | -8.40.154 |
| Fence and open | +2.80.720 | -1.60.838 | +0.90.910 | -5.20.497 | +6.70.387 | +3.80.625 | 51.7%60 | +3.70.636 | -0.30.969 | -4.70.542 |
| Open | -1.00.879 | -5.30.405 | -2.90.653 | -8.90.146 | +2.90.636 | +0.00.998 | -3.70.636 | 47.9%121 | -4.00.525 | -8.40.174 |
| Rectangle | +3.10.618 | -1.30.836 | +1.20.850 | -4.90.419 | +7.00.255 | +4.00.506 | +0.30.969 | +4.00.525 | 52.0%127 | -4.40.473 |
| Octagon | +7.50.215 | +3.10.618 | +5.60.361 | -0.50.930 | +11.30.057 | +8.40.154 | +4.70.542 | +8.40.174 | +4.40.473 | 56.3%142 |
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 counted293
Win rate54.3%
Widest swing-23.5 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 |
|---|---|---|---|---|---|
| Breakable half fence | 67.9% | +13.6 | 0.129 | 28 · | |
| Half fence | 63.6% | +9.4 | 0.251 | 33 · | |
| Full fence | 59.1% | +4.8 | 0.637 | 22 · | |
| Octagon | 58.3% | +4.1 | 0.676 | 24 · | |
| Open | 56.7% | +2.4 | 0.781 | 30 · | |
| Low fence | 54.3% | +0.0 | 0.998 | 35 · | |
| Breakable full fence | 52.6% | -1.6 | 0.828 | 38 · | |
| Rectangle | 45.5% | -8.8 | 0.281 | 33 · | |
| Single wall | 42.1% | -12.2 | 0.271 | 19 · | |
| Fence and open | 30.8% | -23.5 | 0.082 | 13 · |
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 8.79 on 9 df, p 0.457
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 59.1%22 | +6.50.628 | -4.50.734 | -8.80.522 | +4.80.722 | +17.00.278 | +28.30.105 | +2.40.861 | +13.60.322 | +0.80.958 |
| Breakable full fence | -6.50.628 | 52.6%38 | -11.00.349 | -15.20.214 | -1.70.887 | +10.50.454 | +21.90.173 | -4.00.740 | +7.20.546 | -5.70.660 |
| Half fence | +4.50.734 | +11.00.349 | 63.6%33 | -4.20.730 | +9.30.434 | +21.50.132 | +32.90.044 | +7.00.572 | +18.20.138 | +5.30.685 |
| Breakable half fence | +8.80.522 | +15.20.214 | +4.20.730 | 67.9%28 | +13.60.274 | +25.80.080 | +37.10.026 | +11.20.380 | +22.40.079 | +9.50.477 |
| Low fence | -4.80.722 | +1.70.887 | -9.30.434 | -13.60.274 | 54.3%35 | +12.20.393 | +23.50.147 | -2.40.847 | +8.80.467 | -4.00.758 |
| Single wall | -17.00.278 | -10.50.454 | -21.50.132 | -25.80.080 | -12.20.393 | 42.1%19 | +11.30.515 | -14.60.320 | -3.30.815 | -16.20.290 |
| Fence and open | -28.30.105 | -21.90.173 | -32.90.044 | -37.10.026 | -23.50.147 | -11.30.515 | 30.8%13 | -25.90.119 | -14.70.362 | -27.60.109 |
| Open | -2.40.861 | +4.00.740 | -7.00.572 | -11.20.380 | +2.40.847 | +14.60.320 | +25.90.119 | 56.7%30 | +11.20.374 | -1.70.902 |
| Rectangle | -13.60.322 | -7.20.546 | -18.20.138 | -22.40.079 | -8.80.467 | +3.30.815 | +14.70.362 | -11.20.374 | 45.5%33 | -12.90.337 |
| Octagon | -0.80.958 | +5.70.660 | -5.30.685 | -9.50.477 | +4.00.758 | +16.20.290 | +27.60.109 | +1.70.902 | +12.90.337 | 58.3%24 |
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.