Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Sarah
Matches counted2,078
Win rate50.6%
Widest swing6.8 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 | 57.4% | +6.8 | 0.035 | 216 | |
| Fence and open | 55.0% | +4.4 | 0.370 | 100 | |
| Breakable full fence | 52.5% | +1.8 | 0.603 | 183 | |
| Low fence | 50.9% | +0.3 | 0.924 | 214 | |
| Rectangle | 49.5% | -1.1 | 0.736 | 210 | |
| Single wall | 49.3% | -1.3 | 0.682 | 213 | |
| Breakable half fence | 49.1% | -1.5 | 0.631 | 232 | |
| Half fence | 49.0% | -1.6 | 0.629 | 204 | |
| Full fence | 48.3% | -2.3 | 0.520 | 180 | |
| Open | 47.8% | -2.9 | 0.364 | 222 |
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 6.78 on 9 df, p 0.660
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.3%180 | -4.10.432 | -0.70.893 | -0.80.871 | -2.60.607 | -1.00.849 | -6.70.285 | +0.60.907 | -1.20.815 | -9.10.072 |
| Breakable full fence | +4.10.432 | 52.5%183 | +3.40.499 | +3.30.502 | +1.50.762 | +3.20.530 | -2.50.682 | +4.70.345 | +2.90.562 | -5.00.322 |
| Half fence | +0.70.893 | -3.40.499 | 49.0%204 | -0.10.980 | -1.90.695 | -0.30.955 | -6.00.327 | +1.30.793 | -0.50.918 | -8.40.085 |
| Breakable half fence | +0.80.871 | -3.30.502 | +0.10.980 | 49.1%232 | -1.80.705 | -0.20.973 | -5.90.327 | +1.40.767 | -0.40.935 | -8.30.080 |
| Low fence | +2.60.607 | -1.50.762 | +1.90.695 | +1.80.705 | 50.9%214 | +1.60.735 | -4.10.502 | +3.20.506 | +1.40.771 | -6.50.178 |
| Single wall | +1.00.849 | -3.20.530 | +0.30.955 | +0.20.973 | -1.60.735 | 49.3%213 | -5.70.347 | +1.60.747 | -0.20.963 | -8.10.092 |
| Fence and open | +6.70.285 | +2.50.682 | +6.00.327 | +5.90.327 | +4.10.502 | +5.70.347 | 55.0%100 | +7.20.228 | +5.50.367 | -2.40.688 |
| Open | -0.60.907 | -4.70.345 | -1.30.793 | -1.40.767 | -3.20.506 | -1.60.747 | -7.20.228 | 47.8%222 | -1.80.712 | -9.70.043 |
| Rectangle | +1.20.815 | -2.90.562 | +0.50.918 | +0.40.935 | -1.40.771 | +0.20.963 | -5.50.367 | +1.80.712 | 49.5%210 | -7.90.103 |
| Octagon | +9.10.072 | +5.00.322 | +8.40.085 | +8.30.080 | +6.50.178 | +8.10.092 | +2.40.688 | +9.70.043 | +7.90.103 | 57.4%216 |
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 counted1,884
Win rate50.4%
Widest swing6.8 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 | 57.2% | +6.8 | 0.042 | 201 | |
| Fence and open | 57.0% | +6.6 | 0.194 | 93 | |
| Breakable full fence | 52.1% | +1.7 | 0.645 | 163 | |
| Half fence | 50.3% | -0.1 | 0.967 | 181 | |
| Low fence | 50.3% | -0.2 | 0.961 | 193 | |
| Full fence | 49.4% | -1.0 | 0.781 | 160 | |
| Rectangle | 49.2% | -1.2 | 0.725 | 197 | |
| Breakable half fence | 47.9% | -2.5 | 0.430 | 211 | |
| Single wall | 47.6% | -2.8 | 0.415 | 187 | |
| Open | 46.3% | -4.1 | 0.214 | 203 |
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 8.21 on 9 df, p 0.513
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.4%160 | -2.80.618 | -0.90.868 | +1.50.774 | -0.90.869 | +1.80.741 | -7.60.242 | +3.10.561 | +0.10.980 | -7.80.138 |
| Breakable full fence | +2.80.618 | 52.1%163 | +1.90.729 | +4.30.412 | +1.90.723 | +4.60.395 | -4.80.455 | +5.80.266 | +2.90.583 | -5.10.334 |
| Half fence | +0.90.868 | -1.90.729 | 50.3%181 | +2.40.634 | +0.00.997 | +2.70.607 | -6.70.292 | +4.00.437 | +1.00.840 | -6.90.174 |
| Breakable half fence | -1.50.774 | -4.30.412 | -2.40.634 | 47.9%211 | -2.40.631 | +0.30.956 | -9.10.143 | +1.60.750 | -1.40.782 | -9.30.058 |
| Low fence | +0.90.869 | -1.90.723 | -0.00.997 | +2.40.631 | 50.3%193 | +2.70.603 | -6.70.286 | +4.00.431 | +1.00.840 | -7.00.166 |
| Single wall | -1.80.741 | -4.60.395 | -2.70.607 | -0.30.956 | -2.70.603 | 47.6%187 | -9.40.139 | +1.30.799 | -1.60.747 | -9.60.058 |
| Fence and open | +7.60.242 | +4.80.455 | +6.70.292 | +9.10.143 | +6.70.286 | +9.40.139 | 57.0%93 | +10.70.088 | +7.80.218 | -0.20.971 |
| Open | -3.10.561 | -5.80.266 | -4.00.437 | -1.60.750 | -4.00.431 | -1.30.799 | -10.70.088 | 46.3%203 | -2.90.557 | -10.90.028 |
| Rectangle | -0.10.980 | -2.90.583 | -1.00.840 | +1.40.782 | -1.00.840 | +1.60.747 | -7.80.218 | +2.90.557 | 49.2%197 | -8.00.111 |
| Octagon | +7.80.138 | +5.10.334 | +6.90.174 | +9.30.058 | +7.00.166 | +9.60.058 | +0.20.971 | +10.90.028 | +8.00.111 | 57.2%201 |
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 counted194
Win rate52.6%
Widest swing-24.0 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 |
|---|---|---|---|---|---|
| Open | 63.2% | +10.6 | 0.331 | 19 · | |
| Breakable half fence | 61.9% | +9.3 | 0.365 | 21 · | |
| Single wall | 61.5% | +9.0 | 0.325 | 26 · | |
| Octagon | 60.0% | +7.4 | 0.549 | 15 · | |
| Low fence | 57.1% | +4.6 | 0.657 | 21 · | |
| Breakable full fence | 55.0% | +2.4 | 0.819 | 20 · | |
| Rectangle | 53.9% | +1.3 | 0.924 | 13 · | |
| Full fence | 40.0% | -12.6 | 0.234 | 20 · | |
| Half fence | 39.1% | -13.4 | 0.169 | 23 · | |
| Fence and open | 28.6% | -24.0 | 0.195 | 7 · |
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 7.49 on 9 df, p 0.586
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 40.0%20 | -15.00.342 | +0.90.954 | -21.90.161 | -17.10.272 | -21.50.147 | +11.40.590 | -23.20.148 | -13.80.435 | -20.00.241 |
| Breakable full fence | +15.00.342 | 55.0%20 | +15.90.298 | -6.90.654 | -2.10.890 | -6.50.655 | +26.40.228 | -8.20.605 | +1.10.948 | -5.00.767 |
| Half fence | -0.90.954 | -15.90.298 | 39.1%23 | -22.80.131 | -18.00.232 | -22.40.117 | +10.60.612 | -24.00.121 | -14.70.393 | -20.90.208 |
| Breakable half fence | +21.90.161 | +6.90.654 | +22.80.131 | 61.9%21 | +4.80.753 | +0.40.980 | +33.30.126 | -1.30.935 | +8.10.643 | +1.90.908 |
| Low fence | +17.10.272 | +2.10.890 | +18.00.232 | -4.80.753 | 57.1%21 | -4.40.760 | +28.60.190 | -6.00.698 | +3.30.851 | -2.90.864 |
| Single wall | +21.50.147 | +6.50.655 | +22.40.117 | -0.40.980 | +4.40.760 | 61.5%26 | +33.00.120 | -1.60.912 | +7.70.645 | +1.50.923 |
| Fence and open | -11.40.590 | -26.40.228 | -10.60.612 | -33.30.126 | -28.60.190 | -33.00.120 | 28.6%7 | -34.60.117 | -25.30.279 | -31.40.170 |
| Open | +23.20.148 | +8.20.605 | +24.00.121 | +1.30.935 | +6.00.698 | +1.60.912 | +34.60.117 | 63.2%19 | +9.30.598 | +3.20.851 |
| Rectangle | +13.80.435 | -1.10.948 | +14.70.393 | -8.10.643 | -3.30.851 | -7.70.645 | +25.30.279 | -9.30.598 | 53.9%13 | -6.20.743 |
| Octagon | +20.00.241 | +5.00.767 | +20.90.208 | -1.90.908 | +2.90.864 | -1.50.923 | +31.40.170 | -3.20.851 | +6.20.743 | 60.0%15 |
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.