Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Eileen
Matches counted646
Win rate49.1%
Widest swing11.6 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 60.7% | +11.6 | 0.068 | 56 · | |
| Breakable full fence | 52.4% | +3.3 | 0.580 | 63 · | |
| Full fence | 51.5% | +2.4 | 0.676 | 68 · | |
| Open | 50.0% | +0.9 | 0.884 | 56 · | |
| Fence and open | 47.8% | -1.2 | 0.903 | 23 · | |
| Low fence | 47.8% | -1.2 | 0.796 | 92 | |
| Breakable half fence | 46.9% | -2.2 | 0.711 | 64 · | |
| Single wall | 46.9% | -2.2 | 0.711 | 64 · | |
| Rectangle | 45.9% | -3.2 | 0.603 | 61 · | |
| Octagon | 41.4% | -7.7 | 0.219 | 58 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 5.43 on 9 df, p 0.796
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 51.5%68 | -0.90.917 | -9.20.302 | +4.60.598 | +3.60.649 | +4.60.598 | +3.60.763 | +1.50.871 | +5.60.528 | +10.10.258 |
| Breakable full fence | +0.90.917 | 52.4%63 | -8.30.360 | +5.50.535 | +4.50.577 | +5.50.535 | +4.50.708 | +2.40.795 | +6.50.471 | +11.00.226 |
| Half fence | +9.20.302 | +8.30.360 | 60.7%56 | +13.80.130 | +12.90.128 | +13.80.130 | +12.90.293 | +10.70.254 | +14.80.109 | +19.30.039 |
| Breakable half fence | -4.60.598 | -5.50.535 | -13.80.130 | 46.9%64 | -0.90.907 | +0.01.000 | -0.90.938 | -3.10.733 | +1.00.913 | +5.50.542 |
| Low fence | -3.60.649 | -4.50.577 | -12.90.128 | +0.90.907 | 47.8%92 | +0.90.907 | +0.01.000 | -2.20.797 | +1.90.815 | +6.50.440 |
| Single wall | -4.60.598 | -5.50.535 | -13.80.130 | -0.01.000 | -0.90.907 | 46.9%64 | -0.90.938 | -3.10.733 | +1.00.913 | +5.50.542 |
| Fence and open | -3.60.763 | -4.50.708 | -12.90.293 | +0.90.938 | -0.01.000 | +0.90.938 | 47.8%23 | -2.20.861 | +1.90.875 | +6.50.597 |
| Open | -1.50.871 | -2.40.795 | -10.70.254 | +3.10.733 | +2.20.797 | +3.10.733 | +2.20.861 | 50.0%56 | +4.10.658 | +8.60.356 |
| Rectangle | -5.60.528 | -6.50.471 | -14.80.109 | -1.00.913 | -1.90.815 | -1.00.913 | -1.90.875 | -4.10.658 | 45.9%61 | +4.50.619 |
| Octagon | -10.10.258 | -11.00.226 | -19.30.039 | -5.50.542 | -6.50.440 | -5.50.542 | -6.50.597 | -8.60.356 | -4.50.619 | 41.4%58 |
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 counted453
Win rate47.7%
Widest swing-14.3 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Full fence | 57.5% | +9.8 | 0.157 | 47 · | |
| Half fence | 56.4% | +8.7 | 0.254 | 39 · | |
| Breakable full fence | 53.2% | +5.5 | 0.424 | 47 · | |
| Single wall | 52.2% | +4.5 | 0.520 | 46 · | |
| Fence and open | 50.0% | +2.3 | 0.860 | 14 · | |
| Breakable half fence | 48.8% | +1.2 | 0.873 | 43 · | |
| Open | 48.6% | +0.9 | 0.913 | 35 · | |
| Low fence | 42.2% | -5.4 | 0.319 | 71 · | |
| Rectangle | 40.0% | -7.7 | 0.277 | 45 · | |
| Octagon | 33.3% | -14.3 | 0.042 | 45 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 9.61 on 9 df, p 0.383
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.5%47 | +4.30.678 | +1.00.923 | +8.60.413 | +15.20.106 | +5.30.609 | +7.50.622 | +8.90.425 | +17.40.094 | +24.10.020 |
| Breakable full fence | -4.30.678 | 53.2%47 | -3.20.765 | +4.30.680 | +10.90.244 | +1.00.922 | +3.20.834 | +4.60.679 | +13.20.205 | +19.90.055 |
| Half fence | -1.00.923 | +3.20.765 | 56.4%39 | +7.60.493 | +14.20.155 | +4.20.696 | +6.40.679 | +7.80.500 | +16.40.133 | +23.10.034 |
| Breakable half fence | -8.60.413 | -4.30.680 | -7.60.493 | 48.8%43 | +6.60.493 | -3.30.753 | -1.20.940 | +0.30.981 | +8.80.404 | +15.50.139 |
| Low fence | -15.20.106 | -10.90.244 | -14.20.155 | -6.60.493 | 42.2%71 | -9.90.293 | -7.80.593 | -6.30.538 | +2.20.810 | +8.90.337 |
| Single wall | -5.30.609 | -1.00.922 | -4.20.696 | +3.30.753 | +9.90.293 | 52.2%46 | +2.20.887 | +3.60.748 | +12.20.244 | +18.80.069 |
| Fence and open | -7.50.622 | -3.20.834 | -6.40.679 | +1.20.940 | +7.80.593 | -2.20.887 | 50.0%14 | +1.40.928 | +10.00.508 | +16.70.260 |
| Open | -8.90.425 | -4.60.679 | -7.80.500 | -0.30.981 | +6.30.538 | -3.60.748 | -1.40.928 | 48.6%35 | +8.60.443 | +15.20.168 |
| Rectangle | -17.40.094 | -13.20.205 | -16.40.133 | -8.80.404 | -2.20.810 | -12.20.244 | -10.00.508 | -8.60.443 | 40.0%45 | +6.70.512 |
| Octagon | -24.10.020 | -19.90.055 | -23.10.034 | -15.50.139 | -8.90.337 | -18.80.069 | -16.70.260 | -15.20.168 | -6.70.512 | 33.3%45 |
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 counted193
Win rate52.3%
Widest swing-19.0 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 70.6% | +18.3 | 0.115 | 17 · | |
| Octagon | 69.2% | +16.9 | 0.207 | 13 · | |
| Low fence | 66.7% | +14.3 | 0.164 | 21 · | |
| Rectangle | 62.5% | +10.2 | 0.395 | 16 · | |
| Open | 52.4% | +0.1 | 0.996 | 21 · | |
| Breakable full fence | 50.0% | -2.3 | 0.845 | 16 · | |
| Fence and open | 44.4% | -7.9 | 0.628 | 9 · | |
| Breakable half fence | 42.9% | -9.5 | 0.357 | 21 · | |
| Full fence | 38.1% | -14.2 | 0.166 | 21 · | |
| Single wall | 33.3% | -19.0 | 0.090 | 18 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 11.48 on 9 df, p 0.244
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 38.1%21 | -11.90.469 | -32.50.046 | -4.80.753 | -28.60.064 | +4.80.757 | -6.30.745 | -14.30.352 | -24.40.141 | -31.10.078 |
| Breakable full fence | +11.90.469 | 50.0%16 | -20.60.226 | +7.10.666 | -16.70.306 | +16.70.324 | +5.60.790 | -2.40.886 | -12.50.476 | -19.20.296 |
| Half fence | +32.50.046 | +20.60.226 | 70.6%17 | +27.70.087 | +3.90.796 | +37.30.028 | +26.10.192 | +18.20.254 | +8.10.622 | +1.40.936 |
| Breakable half fence | +4.80.753 | -7.10.666 | -27.70.087 | 42.9%21 | -23.80.121 | +9.50.542 | -1.60.936 | -9.50.537 | -19.60.236 | -26.40.134 |
| Low fence | +28.60.064 | +16.70.306 | -3.90.796 | +23.80.121 | 66.7%21 | +33.30.038 | +22.20.255 | +14.30.346 | +4.20.793 | -2.60.877 |
| Single wall | -4.80.757 | -16.70.324 | -37.30.028 | -9.50.542 | -33.30.038 | 33.3%18 | -11.10.573 | -19.10.232 | -29.20.089 | -35.90.048 |
| Fence and open | +6.30.745 | -5.60.790 | -26.10.192 | +1.60.936 | -22.20.255 | +11.10.573 | 44.4%9 | -7.90.690 | -18.10.383 | -24.80.245 |
| Open | +14.30.352 | +2.40.886 | -18.20.254 | +9.50.537 | -14.30.346 | +19.10.232 | +7.90.690 | 52.4%21 | -10.10.538 | -16.90.332 |
| Rectangle | +24.40.141 | +12.50.476 | -8.10.622 | +19.60.236 | -4.20.793 | +29.20.089 | +18.10.383 | +10.10.538 | 62.5%16 | -6.70.705 |
| Octagon | +31.10.078 | +19.20.296 | -1.40.936 | +26.40.134 | +2.60.877 | +35.90.048 | +24.80.245 | +16.90.332 | +6.70.705 | 69.2%13 |
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.