Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Dural
Matches counted437
Win rate50.3%
Widest swing20.2 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 |
|---|---|---|---|---|---|
| Fence and open | 70.6% | +20.2 | 0.089 | 17 · | |
| Single wall | 56.5% | +6.2 | 0.376 | 46 · | |
| Low fence | 55.6% | +5.2 | 0.514 | 36 · | |
| Open | 54.7% | +4.4 | 0.497 | 53 · | |
| Octagon | 53.5% | +3.1 | 0.664 | 43 · | |
| Breakable full fence | 52.3% | +1.9 | 0.787 | 44 · | |
| Rectangle | 47.2% | -3.1 | 0.696 | 36 · | |
| Breakable half fence | 46.3% | -4.0 | 0.590 | 41 · | |
| Half fence | 42.5% | -7.8 | 0.258 | 47 · | |
| Full fence | 39.6% | -10.7 | 0.096 | 53 · |
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 8.50 on 9 df, p 0.485
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 39.6%53 | -12.70.213 | -2.90.766 | -6.70.514 | -15.90.139 | -16.90.093 | -31.00.026 | -15.10.120 | -7.60.477 | -13.90.175 |
| Breakable full fence | +12.70.213 | 52.3%44 | +9.70.353 | +5.90.585 | -3.30.770 | -4.20.686 | -18.30.195 | -2.50.810 | +5.00.653 | -1.20.910 |
| Half fence | +2.90.766 | -9.70.353 | 42.5%47 | -3.80.721 | -13.00.240 | -14.00.178 | -28.00.048 | -12.20.225 | -4.70.671 | -10.90.300 |
| Breakable half fence | +6.70.514 | -5.90.585 | +3.80.721 | 46.3%41 | -9.20.420 | -10.20.343 | -24.20.092 | -8.40.420 | -0.90.938 | -7.20.513 |
| Low fence | +15.90.139 | +3.30.770 | +13.00.240 | +9.20.420 | 55.6%36 | -1.00.930 | -15.00.296 | +0.80.938 | +8.30.479 | +2.10.854 |
| Single wall | +16.90.093 | +4.20.686 | +14.00.178 | +10.20.343 | +1.00.930 | 56.5%46 | -14.10.311 | +1.80.857 | +9.30.403 | +3.00.774 |
| Fence and open | +31.00.026 | +18.30.195 | +28.00.048 | +24.20.092 | +15.00.296 | +14.10.311 | 70.6%17 | +15.90.248 | +23.40.111 | +17.10.226 |
| Open | +15.10.120 | +2.50.810 | +12.20.225 | +8.40.420 | -0.80.938 | -1.80.857 | -15.90.248 | 54.7%53 | +7.50.487 | +1.20.904 |
| Rectangle | +7.60.477 | -5.00.653 | +4.70.671 | +0.90.938 | -8.30.479 | -9.30.403 | -23.40.111 | -7.50.487 | 47.2%36 | -6.30.579 |
| Octagon | +13.90.175 | +1.20.910 | +10.90.300 | +7.20.513 | -2.10.854 | -3.00.774 | -17.10.226 | -1.20.904 | +6.30.579 | 53.5%43 |
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 counted418
Win rate50.0%
Widest swing20.6 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 |
|---|---|---|---|---|---|
| Fence and open | 70.6% | +20.6 | 0.083 | 17 · | |
| Single wall | 56.5% | +6.5 | 0.348 | 46 · | |
| Octagon | 56.1% | +6.1 | 0.411 | 41 · | |
| Open | 56.0% | +6.0 | 0.366 | 50 · | |
| Low fence | 54.3% | +4.3 | 0.596 | 35 · | |
| Breakable full fence | 52.3% | +2.3 | 0.750 | 44 · | |
| Rectangle | 47.2% | -2.8 | 0.727 | 36 · | |
| Breakable half fence | 43.6% | -6.4 | 0.400 | 39 · | |
| Half fence | 38.5% | -11.5 | 0.130 | 39 · | |
| Full fence | 38.0% | -12.0 | 0.071 | 50 · |
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 11.05 on 9 df, p 0.272
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 38.0%50 | -14.30.165 | -0.50.965 | -5.60.594 | -16.30.137 | -18.50.069 | -32.60.020 | -18.00.071 | -9.20.392 | -18.10.085 |
| Breakable full fence | +14.30.165 | 52.3%44 | +13.80.208 | +8.70.429 | -2.00.859 | -4.20.686 | -18.30.195 | -3.70.717 | +5.00.653 | -3.80.724 |
| Half fence | +0.50.965 | -13.80.208 | 38.5%39 | -5.10.645 | -15.80.173 | -18.10.097 | -32.10.027 | -17.50.100 | -8.80.443 | -17.60.114 |
| Breakable half fence | +5.60.594 | -8.70.429 | +5.10.645 | 43.6%39 | -10.70.358 | -12.90.235 | -27.00.063 | -12.40.245 | -3.60.752 | -12.50.263 |
| Low fence | +16.30.137 | +2.00.859 | +15.80.173 | +10.70.358 | 54.3%35 | -2.20.841 | -16.30.261 | -1.70.876 | +7.10.552 | -1.80.874 |
| Single wall | +18.50.069 | +4.20.686 | +18.10.097 | +12.90.235 | +2.20.841 | 56.5%46 | -14.10.311 | +0.50.959 | +9.30.403 | +0.40.968 |
| Fence and open | +32.60.020 | +18.30.195 | +32.10.027 | +27.00.063 | +16.30.261 | +14.10.311 | 70.6%17 | +14.60.289 | +23.40.111 | +14.50.304 |
| Open | +18.00.071 | +3.70.717 | +17.50.100 | +12.40.245 | +1.70.876 | -0.50.959 | -14.60.289 | 56.0%50 | +8.80.421 | -0.10.993 |
| Rectangle | +9.20.392 | -5.00.653 | +8.80.443 | +3.60.752 | -7.10.552 | -9.30.403 | -23.40.111 | -8.80.421 | 47.2%36 | -8.90.437 |
| Octagon | +18.10.085 | +3.80.724 | +17.60.114 | +12.50.263 | +1.80.874 | -0.40.968 | -14.50.304 | +0.10.993 | +8.90.437 | 56.1%41 |
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 counted19
Win rate57.9%
Widest swing-57.9 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 | 100.0% | +42.1 | 0.202 | 2 · | |
| Low fence | 100.0% | +42.1 | 0.381 | 1 · | |
| Full fence | 66.7% | +8.8 | 0.737 | 3 · | |
| Half fence | 62.5% | +4.6 | 0.729 | 8 · | |
| Open | 33.3% | -24.6 | 0.348 | 3 · | |
| Octagon | 0.0% | -57.9 | 0.080 | 2 · |
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 5.84 on 5 df, p 0.322
| Stage | Full | Half | Break. half | Low | Open | Oct |
|---|---|---|---|---|---|---|
| Full fence | 66.7%3 | +4.20.898 | -33.30.361 | -33.30.505 | +33.30.414 | +66.70.136 |
| Half fence | -4.20.898 | 62.5%8 | -37.50.301 | -37.50.453 | +29.20.387 | +62.50.114 |
| Breakable half fence | +33.30.361 | +37.50.301 | 100.0%2 | +0.01.000 | +66.70.136 | +100.00.046 |
| Low fence | +33.30.505 | +37.50.453 | -0.01.000 | 100.0%1 | +66.70.248 | +100.00.083 |
| Open | -33.30.414 | -29.20.387 | -66.70.136 | -66.70.248 | 33.3%3 | +33.30.361 |
| Octagon | -66.70.136 | -62.50.114 | -100.00.046 | -100.00.083 | -33.30.361 | 0.0%2 |
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.