Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Shun
Matches counted1,849
Win rate53.1%
Widest swing10.5 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 |
|---|---|---|---|---|---|
| Octagon | 63.5% | +10.5 | 0.006 | 159 | |
| Single wall | 55.2% | +2.2 | 0.501 | 210 | |
| Fence and open | 53.5% | +0.4 | 0.930 | 116 | |
| Full fence | 53.1% | +0.1 | 0.981 | 175 | |
| Rectangle | 52.2% | -0.9 | 0.793 | 205 | |
| Breakable half fence | 52.1% | -0.9 | 0.794 | 186 | |
| Open | 51.8% | -1.2 | 0.735 | 166 | |
| Low fence | 51.5% | -1.6 | 0.661 | 171 | |
| Half fence | 50.5% | -2.6 | 0.443 | 200 | |
| Breakable full fence | 49.7% | -3.3 | 0.342 | 183 |
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 9.13 on 9 df, p 0.425
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 53.1%175 | +3.40.518 | +2.60.609 | +1.00.850 | +1.70.754 | -2.10.681 | -0.30.959 | +1.30.805 | +0.90.854 | -10.40.055 |
| Breakable full fence | -3.40.518 | 49.7%183 | -0.80.880 | -2.40.641 | -1.70.744 | -5.50.275 | -3.70.530 | -2.10.698 | -2.50.627 | -13.80.010 |
| Half fence | -2.60.609 | +0.80.880 | 50.5%200 | -1.60.746 | -1.00.853 | -4.70.337 | -3.00.613 | -1.30.803 | -1.70.733 | -13.00.013 |
| Breakable half fence | -1.00.850 | +2.40.641 | +1.60.746 | 52.1%186 | +0.70.897 | -3.10.539 | -1.30.826 | +0.30.949 | -0.10.993 | -11.40.033 |
| Low fence | -1.70.754 | +1.70.744 | +1.00.853 | -0.70.897 | 51.5%171 | -3.80.462 | -2.00.741 | -0.30.949 | -0.70.887 | -12.10.027 |
| Single wall | +2.10.681 | +5.50.275 | +4.70.337 | +3.10.539 | +3.80.462 | 55.2%210 | +1.80.756 | +3.40.508 | +3.00.534 | -8.30.109 |
| Fence and open | +0.30.959 | +3.70.530 | +3.00.613 | +1.30.826 | +2.00.741 | -1.80.756 | 53.5%116 | +1.60.786 | +1.20.829 | -10.10.093 |
| Open | -1.30.805 | +2.10.698 | +1.30.803 | -0.30.949 | +0.30.949 | -3.40.508 | -1.60.786 | 51.8%166 | -0.40.941 | -11.70.033 |
| Rectangle | -0.90.854 | +2.50.627 | +1.70.733 | +0.10.993 | +0.70.887 | -3.00.534 | -1.20.829 | +0.40.941 | 52.2%205 | -11.30.030 |
| Octagon | +10.40.055 | +13.80.010 | +13.00.013 | +11.40.033 | +12.10.027 | +8.30.109 | +10.10.093 | +11.70.033 | +11.30.030 | 63.5%159 |
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 counted1,538
Win rate52.3%
Widest swing10.2 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 |
|---|---|---|---|---|---|
| Octagon | 62.5% | +10.2 | 0.016 | 128 | |
| Single wall | 54.3% | +2.0 | 0.555 | 186 | |
| Fence and open | 52.6% | +0.3 | 0.951 | 97 | |
| Breakable half fence | 51.9% | -0.4 | 0.926 | 156 | |
| Open | 51.8% | -0.5 | 0.912 | 137 | |
| Low fence | 51.4% | -0.8 | 0.833 | 140 | |
| Rectangle | 51.2% | -1.1 | 0.779 | 162 | |
| Full fence | 51.1% | -1.2 | 0.762 | 141 | |
| Breakable full fence | 49.4% | -2.9 | 0.430 | 168 | |
| Half fence | 48.5% | -3.8 | 0.304 | 161 |
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 7.39 on 9 df, p 0.597
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 51.1%141 | +1.70.771 | +2.60.650 | -0.90.882 | -0.40.951 | -3.20.561 | -1.50.818 | -0.80.899 | -0.20.976 | -11.40.059 |
| Breakable full fence | -1.70.771 | 49.4%168 | +0.90.862 | -2.50.651 | -2.00.724 | -4.90.357 | -3.20.619 | -2.40.674 | -1.80.740 | -13.10.025 |
| Half fence | -2.60.650 | -0.90.862 | 48.5%161 | -3.50.536 | -3.00.606 | -5.80.277 | -4.10.520 | -3.40.561 | -2.80.616 | -14.10.017 |
| Breakable half fence | +0.90.882 | +2.50.651 | +3.50.536 | 51.9%156 | +0.50.932 | -2.40.661 | -0.70.919 | +0.10.987 | +0.70.902 | -10.60.073 |
| Low fence | +0.40.951 | +2.00.724 | +3.00.606 | -0.50.932 | 51.4%140 | -2.90.607 | -1.10.862 | -0.40.947 | +0.20.973 | -11.10.068 |
| Single wall | +3.20.561 | +4.90.357 | +5.80.277 | +2.40.661 | +2.90.607 | 54.3%186 | +1.70.783 | +2.50.659 | +3.10.568 | -8.20.149 |
| Fence and open | +1.50.818 | +3.20.619 | +4.10.520 | +0.70.919 | +1.10.862 | -1.70.783 | 52.6%97 | +0.80.910 | +1.40.834 | -9.90.135 |
| Open | +0.80.899 | +2.40.674 | +3.40.561 | -0.10.987 | +0.40.947 | -2.50.659 | -0.80.910 | 51.8%137 | +0.60.919 | -10.70.079 |
| Rectangle | +0.20.976 | +1.80.740 | +2.80.616 | -0.70.902 | -0.20.973 | -3.10.568 | -1.40.834 | -0.60.919 | 51.2%162 | -11.30.055 |
| Octagon | +11.40.059 | +13.10.025 | +14.10.017 | +10.60.073 | +11.10.068 | +8.20.149 | +9.90.135 | +10.70.079 | +11.30.055 | 62.5%128 |
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 counted311
Win rate56.9%
Widest swing10.8 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 |
|---|---|---|---|---|---|
| Octagon | 67.7% | +10.8 | 0.199 | 31 · | |
| Single wall | 62.5% | +5.6 | 0.565 | 24 · | |
| Full fence | 61.8% | +4.8 | 0.545 | 34 · | |
| Half fence | 59.0% | +2.1 | 0.781 | 39 · | |
| Fence and open | 57.9% | +1.0 | 0.929 | 19 · | |
| Rectangle | 55.8% | -1.1 | 0.875 | 43 · | |
| Breakable full fence | 53.3% | -3.6 | 0.774 | 15 · | |
| Breakable half fence | 53.3% | -3.6 | 0.677 | 30 · | |
| Open | 51.7% | -5.2 | 0.553 | 29 · | |
| Low fence | 51.6% | -5.3 | 0.530 | 31 · |
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 3.07 on 9 df, p 0.961
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 61.8%34 | +8.40.580 | +2.80.808 | +8.40.496 | +10.20.409 | -0.70.955 | +3.90.782 | +10.00.422 | +6.00.599 | -6.00.615 |
| Breakable full fence | -8.40.580 | 53.3%15 | -5.60.707 | +0.01.000 | +1.70.913 | -9.20.571 | -4.60.790 | +1.60.919 | -2.50.868 | -14.40.343 |
| Half fence | -2.80.808 | +5.60.707 | 59.0%39 | +5.60.639 | +7.40.538 | -3.50.781 | +1.10.938 | +7.20.552 | +3.20.773 | -8.80.451 |
| Breakable half fence | -8.40.496 | -0.01.000 | -5.60.639 | 53.3%30 | +1.70.893 | -9.20.498 | -4.60.754 | +1.60.902 | -2.50.834 | -14.40.249 |
| Low fence | -10.20.409 | -1.70.913 | -7.40.538 | -1.70.893 | 51.6%31 | -10.90.419 | -6.30.665 | -0.10.993 | -4.20.721 | -16.10.196 |
| Single wall | +0.70.955 | +9.20.571 | +3.50.781 | +9.20.498 | +10.90.419 | 62.5%24 | +4.60.759 | +10.80.431 | +6.70.595 | -5.20.685 |
| Fence and open | -3.90.782 | +4.60.790 | -1.10.938 | +4.60.754 | +6.30.665 | -4.60.759 | 57.9%19 | +6.20.675 | +2.10.879 | -9.80.481 |
| Open | -10.00.422 | -1.60.919 | -7.20.552 | -1.60.902 | +0.10.993 | -10.80.431 | -6.20.675 | 51.7%29 | -4.10.733 | -16.00.206 |
| Rectangle | -6.00.599 | +2.50.868 | -3.20.773 | +2.50.834 | +4.20.721 | -6.70.595 | -2.10.879 | +4.10.733 | 55.8%43 | -11.90.300 |
| Octagon | +6.00.615 | +14.40.343 | +8.80.451 | +14.40.249 | +16.10.196 | +5.20.685 | +9.80.481 | +16.00.206 | +11.90.300 | 67.7%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.