Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Jeffry
Matches counted2,348
Win rate51.7%
Widest swing8.3 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 |
|---|---|---|---|---|---|
| Breakable half fence | 60.0% | +8.3 | 0.008 | 230 | |
| Fence and open | 54.2% | +2.5 | 0.579 | 120 | |
| Octagon | 53.9% | +2.2 | 0.485 | 232 | |
| Full fence | 52.5% | +0.8 | 0.772 | 278 | |
| Low fence | 52.2% | +0.5 | 0.868 | 224 | |
| Single wall | 52.2% | +0.5 | 0.868 | 224 | |
| Open | 51.0% | -0.7 | 0.807 | 255 | |
| Half fence | 50.0% | -1.7 | 0.601 | 214 | |
| Breakable full fence | 46.7% | -5.0 | 0.126 | 210 | |
| Rectangle | 44.7% | -7.0 | 0.020 | 244 |
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 14.46 on 9 df, p 0.107
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.5%278 | +5.80.201 | +2.50.580 | -7.50.091 | +0.30.949 | +0.30.949 | -1.60.762 | +1.50.723 | +7.80.074 | -1.40.759 |
| Breakable full fence | -5.80.201 | 46.7%210 | -3.30.492 | -13.30.005 | -5.60.247 | -5.60.247 | -7.50.190 | -4.30.354 | +2.00.671 | -7.20.130 |
| Half fence | -2.50.580 | +3.30.492 | 50.0%214 | -10.00.034 | -2.20.640 | -2.20.640 | -4.20.465 | -1.00.832 | +5.30.254 | -3.90.413 |
| Breakable half fence | +7.50.091 | +13.30.005 | +10.00.034 | 60.0%230 | +7.80.095 | +7.80.095 | +5.80.294 | +9.00.046 | +15.3<0.001 | +6.10.184 |
| Low fence | -0.30.949 | +5.60.247 | +2.20.640 | -7.80.095 | 52.2%224 | +0.01.000 | -1.90.732 | +1.20.784 | +7.60.102 | -1.60.725 |
| Single wall | -0.30.949 | +5.60.247 | +2.20.640 | -7.80.095 | -0.01.000 | 52.2%224 | -1.90.732 | +1.20.784 | +7.60.102 | -1.60.725 |
| Fence and open | +1.60.762 | +7.50.190 | +4.20.465 | -5.80.294 | +1.90.732 | +1.90.732 | 54.2%120 | +3.20.565 | +9.50.088 | +0.30.959 |
| Open | -1.50.723 | +4.30.354 | +1.00.832 | -9.00.046 | -1.20.784 | -1.20.784 | -3.20.565 | 51.0%255 | +6.30.159 | -2.90.522 |
| Rectangle | -7.80.074 | -2.00.671 | -5.30.254 | -15.3<0.001 | -7.60.102 | -7.60.102 | -9.50.088 | -6.30.159 | 44.7%244 | -9.20.045 |
| Octagon | +1.40.759 | +7.20.130 | +3.90.413 | -6.10.184 | +1.60.725 | +1.60.725 | -0.30.959 | +2.90.522 | +9.20.045 | 53.9%232 |
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 counted2,074
Win rate52.0%
Widest swing8.6 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 |
|---|---|---|---|---|---|
| Breakable half fence | 60.6% | +8.6 | 0.010 | 203 | |
| Fence and open | 55.5% | +3.4 | 0.459 | 110 | |
| Low fence | 54.2% | +2.1 | 0.533 | 192 | |
| Half fence | 52.7% | +0.6 | 0.855 | 188 | |
| Single wall | 52.1% | +0.1 | 0.971 | 186 | |
| Full fence | 52.0% | -0.1 | 0.985 | 254 | |
| Octagon | 51.2% | -0.8 | 0.804 | 207 | |
| Open | 50.4% | -1.6 | 0.611 | 228 | |
| Breakable full fence | 46.6% | -5.4 | 0.115 | 193 | |
| Rectangle | 45.8% | -6.3 | 0.054 | 212 |
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 12.73 on 9 df, p 0.175
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.0%254 | +5.30.264 | -0.70.886 | -8.60.065 | -2.20.645 | -0.20.970 | -3.50.541 | +1.50.737 | +6.20.182 | +0.80.871 |
| Breakable full fence | -5.30.264 | 46.6%193 | -6.00.239 | -14.00.005 | -7.50.139 | -5.50.283 | -8.80.140 | -3.80.436 | +0.90.860 | -4.60.360 |
| Half fence | +0.70.886 | +6.00.239 | 52.7%188 | -7.90.114 | -1.50.768 | +0.50.921 | -2.80.641 | +2.20.652 | +6.90.168 | +1.40.773 |
| Breakable half fence | +8.60.065 | +14.00.005 | +7.90.114 | 60.6%203 | +6.40.197 | +8.40.093 | +5.10.378 | +10.20.034 | +14.80.002 | +9.40.056 |
| Low fence | +2.20.645 | +7.50.139 | +1.50.768 | -6.40.197 | 54.2%192 | +2.00.695 | -1.30.829 | +3.70.446 | +8.40.091 | +3.00.554 |
| Single wall | +0.20.970 | +5.50.283 | -0.50.921 | -8.40.093 | -2.00.695 | 52.1%186 | -3.30.582 | +1.70.729 | +6.40.203 | +0.90.852 |
| Fence and open | +3.50.541 | +8.80.140 | +2.80.641 | -5.10.378 | +1.30.829 | +3.30.582 | 55.5%110 | +5.00.387 | +9.70.099 | +4.20.471 |
| Open | -1.50.737 | +3.80.436 | -2.20.652 | -10.20.034 | -3.70.446 | -1.70.729 | -5.00.387 | 50.4%228 | +4.70.326 | -0.80.873 |
| Rectangle | -6.20.182 | -0.90.860 | -6.90.168 | -14.80.002 | -8.40.091 | -6.40.203 | -9.70.099 | -4.70.326 | 45.8%212 | -5.50.264 |
| Octagon | -0.80.871 | +4.60.360 | -1.40.773 | -9.40.056 | -3.00.554 | -0.90.852 | -4.20.471 | +0.80.873 | +5.50.264 | 51.2%207 |
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 counted274
Win rate49.3%
Widest swing26.7 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 | 76.0% | +26.7 | 0.005 | 25 · | |
| Full fence | 58.3% | +9.1 | 0.352 | 24 · | |
| Breakable half fence | 55.6% | +6.3 | 0.491 | 27 · | |
| Open | 55.6% | +6.3 | 0.491 | 27 · | |
| Single wall | 52.6% | +3.4 | 0.655 | 38 · | |
| Breakable full fence | 47.1% | -2.2 | 0.851 | 17 · | |
| Low fence | 40.6% | -8.7 | 0.298 | 32 · | |
| Fence and open | 40.0% | -9.3 | 0.550 | 10 · | |
| Rectangle | 37.5% | -11.8 | 0.156 | 32 · | |
| Half fence | 30.8% | -18.5 | 0.047 | 26 · |
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 15.61 on 9 df, p 0.075
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 58.3%24 | +11.30.476 | +27.60.050 | +2.80.842 | +17.70.189 | +5.70.660 | +18.30.329 | +2.80.842 | +20.80.122 | -17.70.187 |
| Breakable full fence | -11.30.476 | 47.1%17 | +16.30.280 | -8.50.583 | +6.40.665 | -5.60.702 | +7.10.722 | -8.50.583 | +9.60.517 | -28.90.055 |
| Half fence | -27.60.050 | -16.30.280 | 30.8%26 | -24.80.069 | -9.80.437 | -21.90.083 | -9.20.599 | -24.80.069 | -6.70.592 | -45.20.001 |
| Breakable half fence | -2.80.842 | +8.50.583 | +24.80.069 | 55.6%27 | +14.90.253 | +2.90.816 | +15.60.401 | +0.01.000 | +18.10.165 | -20.40.122 |
| Low fence | -17.70.189 | -6.40.665 | +9.80.437 | -14.90.253 | 40.6%32 | -12.00.316 | +0.60.972 | -14.90.253 | +3.10.798 | -35.40.008 |
| Single wall | -5.70.660 | +5.60.702 | +21.90.083 | -2.90.816 | +12.00.316 | 52.6%38 | +12.60.477 | -2.90.816 | +15.10.206 | -23.40.062 |
| Fence and open | -18.30.329 | -7.10.722 | +9.20.599 | -15.60.401 | -0.60.972 | -12.60.477 | 40.0%10 | -15.60.401 | +2.50.887 | -36.00.043 |
| Open | -2.80.842 | +8.50.583 | +24.80.069 | -0.01.000 | +14.90.253 | +2.90.816 | +15.60.401 | 55.6%27 | +18.10.165 | -20.40.122 |
| Rectangle | -20.80.122 | -9.60.517 | +6.70.592 | -18.10.165 | -3.10.798 | -15.10.206 | -2.50.887 | -18.10.165 | 37.5%32 | -38.50.004 |
| Octagon | +17.70.187 | +28.90.055 | +45.20.001 | +20.40.122 | +35.40.008 | +23.40.062 | +36.00.043 | +20.40.122 | +38.50.004 | 76.0%25 |
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.