Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Dural vs Eileen
Matches counted143
Win rate52.5%
Widest swing-25.8 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Dural's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 66.7% | +14.2 | 0.244 | 15 · | |
| Breakable half fence | 64.3% | +11.8 | 0.350 | 14 · | |
| Full fence | 62.5% | +10.1 | 0.393 | 16 · | |
| Octagon | 61.5% | +9.1 | 0.491 | 13 · | |
| Single wall | 60.0% | +7.5 | 0.731 | 5 · | |
| Rectangle | 57.1% | +4.7 | 0.799 | 7 · | |
| Low fence | 52.0% | -0.5 | 0.961 | 25 · | |
| Fence and open | 42.9% | -9.6 | 0.449 | 14 · | |
| Open | 42.9% | -9.6 | 0.449 | 14 · | |
| Breakable full fence | 26.7% | -25.8 | 0.035 | 15 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Dural's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.29 on 9 df, p 0.506
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 62.5%16 | +35.80.045 | -4.20.809 | -1.80.919 | +10.50.509 | +2.50.920 | +19.60.282 | +19.60.282 | +5.40.809 | +1.00.958 |
| Breakable full fence | -35.80.045 | 26.7%15 | -40.00.028 | -37.60.042 | -25.30.117 | -33.30.176 | -16.20.359 | -16.20.359 | -30.50.166 | -34.90.063 |
| Half fence | +4.20.809 | +40.00.028 | 66.7%15 | +2.40.893 | +14.70.364 | +6.70.787 | +23.80.198 | +23.80.198 | +9.50.665 | +5.10.778 |
| Breakable half fence | +1.80.919 | +37.60.042 | -2.40.893 | 64.3%14 | +12.30.458 | +4.30.865 | +21.40.256 | +21.40.256 | +7.20.751 | +2.80.883 |
| Low fence | -10.50.509 | +25.30.117 | -14.70.364 | -12.30.458 | 52.0%25 | -8.00.743 | +9.10.584 | +9.10.584 | -5.10.810 | -9.50.575 |
| Single wall | -2.50.920 | +33.30.176 | -6.70.787 | -4.30.865 | +8.00.743 | 60.0%5 | +17.10.510 | +17.10.510 | +2.90.921 | -1.50.952 |
| Fence and open | -19.60.282 | +16.20.359 | -23.80.198 | -21.40.256 | -9.10.584 | -17.10.510 | 42.9%14 | +0.01.000 | -14.30.537 | -18.70.332 |
| Open | -19.60.282 | +16.20.359 | -23.80.198 | -21.40.256 | -9.10.584 | -17.10.510 | -0.01.000 | 42.9%14 | -14.30.537 | -18.70.332 |
| Rectangle | -5.40.809 | +30.50.166 | -9.50.665 | -7.20.751 | +5.10.810 | -2.90.921 | +14.30.537 | +14.30.537 | 57.1%7 | -4.40.848 |
| Octagon | -1.00.958 | +34.90.063 | -5.10.778 | -2.80.883 | +9.50.575 | +1.50.952 | +18.70.332 | +18.70.332 | +4.40.848 | 61.5%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.
Matches counted125
Win rate52.0%
Widest swing-25.3 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Dural's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Full fence | 66.7% | +14.7 | 0.285 | 12 · | |
| Half fence | 66.7% | +14.7 | 0.226 | 15 · | |
| Rectangle | 66.7% | +14.7 | 0.461 | 6 · | |
| Octagon | 61.5% | +9.5 | 0.467 | 13 · | |
| Single wall | 60.0% | +8.0 | 0.715 | 5 · | |
| Breakable half fence | 55.6% | +3.6 | 0.825 | 9 · | |
| Low fence | 47.6% | -4.4 | 0.660 | 21 · | |
| Fence and open | 46.1% | -5.8 | 0.656 | 13 · | |
| Open | 46.1% | -5.8 | 0.656 | 13 · | |
| Breakable full fence | 26.7% | -25.3 | 0.036 | 15 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Dural's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 7.86 on 9 df, p 0.548
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 66.7%12 | +40.00.038 | +0.01.000 | +11.10.604 | +19.10.290 | +6.70.793 | +20.50.302 | +20.50.302 | +0.01.000 | +5.10.790 |
| Breakable full fence | -40.00.038 | 26.7%15 | -40.00.028 | -28.90.157 | -20.90.204 | -33.30.176 | -19.50.283 | -19.50.283 | -40.00.088 | -34.90.063 |
| Half fence | -0.01.000 | +40.00.028 | 66.7%15 | +11.10.586 | +19.10.257 | +6.70.787 | +20.50.274 | +20.50.274 | +0.01.000 | +5.10.778 |
| Breakable half fence | -11.10.604 | +28.90.157 | -11.10.586 | 55.6%9 | +7.90.690 | -4.40.872 | +9.40.665 | +9.40.665 | -11.10.667 | -6.00.779 |
| Low fence | -19.10.290 | +20.90.204 | -19.10.257 | -7.90.690 | 47.6%21 | -12.40.619 | +1.50.934 | +1.50.934 | -19.10.410 | -13.90.429 |
| Single wall | -6.70.793 | +33.30.176 | -6.70.787 | +4.40.872 | +12.40.619 | 60.0%5 | +13.80.599 | +13.80.599 | -6.70.819 | -1.50.952 |
| Fence and open | -20.50.302 | +19.50.283 | -20.50.274 | -9.40.665 | -1.50.934 | -13.80.599 | 46.1%13 | +0.01.000 | -20.50.405 | -15.40.431 |
| Open | -20.50.302 | +19.50.283 | -20.50.274 | -9.40.665 | -1.50.934 | -13.80.599 | -0.01.000 | 46.1%13 | -20.50.405 | -15.40.431 |
| Rectangle | -0.01.000 | +40.00.088 | -0.01.000 | +11.10.667 | +19.10.410 | +6.70.819 | +20.50.405 | +20.50.405 | 66.7%6 | +5.10.829 |
| Octagon | -5.10.790 | +34.90.063 | -5.10.778 | +6.00.779 | +13.90.429 | +1.50.952 | +15.40.431 | +15.40.431 | -5.10.829 | 61.5%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.
Matches counted18
Win rate55.6%
Widest swing-55.6 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Dural's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable half fence | 80.0% | +24.4 | 0.196 | 5 · | |
| Low fence | 75.0% | +19.4 | 0.375 | 4 · | |
| Full fence | 50.0% | -5.6 | 0.800 | 4 · | |
| Fence and open | 0.0% | -55.6 | 0.250 | 1 · | |
| Open | 0.0% | -55.6 | 0.250 | 1 · | |
| Rectangle | 0.0% | -55.6 | 0.250 | 1 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Dural's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 5.64 on 5 df, p 0.343
| Stage | Full | Break. half | Low | Mixed | Open | Rect |
|---|---|---|---|---|---|---|
| Full fence | 50.0%4 | -30.00.343 | -25.00.465 | +50.00.361 | +50.00.361 | +50.00.361 |
| Breakable half fence | +30.00.343 | 80.0%5 | +5.00.858 | +80.00.121 | +80.00.121 | +80.00.121 |
| Low fence | +25.00.465 | -5.00.858 | 75.0%4 | +75.00.171 | +75.00.171 | +75.00.171 |
| Fence and open | -50.00.361 | -80.00.121 | -75.00.171 | 0.0%1 | +0.01.000 | +0.01.000 |
| Open | -50.00.361 | -80.00.121 | -75.00.171 | -0.01.000 | 0.0%1 | +0.01.000 |
| Rectangle | -50.00.361 | -80.00.121 | -75.00.171 | -0.01.000 | -0.01.000 | 0.0%1 |
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.