Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Dural vs Goh
Matches counted467
Win rate48.2%
Widest swing9.0 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 | 57.1% | +9.0 | 0.184 | 49 · | |
| Breakable full fence | 51.9% | +3.7 | 0.567 | 52 · | |
| Open | 51.7% | +3.5 | 0.693 | 29 · | |
| Rectangle | 51.0% | +2.8 | 0.674 | 49 · | |
| Fence and open | 50.0% | +1.8 | 0.875 | 18 · | |
| Single wall | 45.8% | -2.4 | 0.731 | 48 · | |
| Octagon | 44.6% | -3.6 | 0.535 | 65 · | |
| Low fence | 43.9% | -4.3 | 0.486 | 57 · | |
| Breakable half fence | 42.2% | -6.0 | 0.400 | 45 · | |
| Full fence | 41.7% | -6.5 | 0.416 | 36 · |
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 4.28 on 9 df, p 0.892
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 41.7%36 | -10.20.344 | -15.50.158 | -0.60.960 | -2.20.835 | -4.20.703 | -8.30.561 | -10.10.419 | -9.30.393 | -3.00.775 |
| Breakable full fence | +10.20.344 | 51.9%52 | -5.20.599 | +9.70.340 | +8.10.400 | +6.10.543 | +1.90.888 | +0.20.986 | +0.90.928 | +7.30.432 |
| Half fence | +15.50.158 | +5.20.599 | 57.1%49 | +14.90.148 | +13.30.173 | +11.30.265 | +7.10.602 | +5.40.642 | +6.10.543 | +12.50.185 |
| Breakable half fence | +0.60.960 | -9.70.340 | -14.90.148 | 42.2%45 | -1.60.868 | -3.60.726 | -7.80.575 | -9.50.423 | -8.80.393 | -2.40.803 |
| Low fence | +2.20.835 | -8.10.400 | -13.30.173 | +1.60.868 | 43.9%57 | -2.00.839 | -6.10.648 | -7.90.489 | -7.20.462 | -0.80.933 |
| Single wall | +4.20.703 | -6.10.543 | -11.30.265 | +3.60.726 | +2.00.839 | 45.8%48 | -4.20.763 | -5.90.616 | -5.20.609 | +1.20.898 |
| Fence and open | +8.30.561 | -1.90.888 | -7.10.602 | +7.80.575 | +6.10.648 | +4.20.763 | 50.0%18 | -1.70.908 | -1.00.941 | +5.40.685 |
| Open | +10.10.419 | -0.20.986 | -5.40.642 | +9.50.423 | +7.90.489 | +5.90.616 | +1.70.908 | 51.7%29 | +0.70.952 | +7.10.523 |
| Rectangle | +9.30.393 | -0.90.928 | -6.10.543 | +8.80.393 | +7.20.462 | +5.20.609 | +1.00.941 | -0.70.952 | 51.0%49 | +6.40.498 |
| Octagon | +3.00.775 | -7.30.432 | -12.50.185 | +2.40.803 | +0.80.933 | -1.20.898 | -5.40.685 | -7.10.523 | -6.40.498 | 44.6%65 |
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 counted439
Win rate48.5%
Widest swing10.6 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 |
|---|---|---|---|---|---|
| Half fence | 59.1% | +10.6 | 0.139 | 44 · | |
| Rectangle | 53.2% | +4.7 | 0.498 | 47 · | |
| Open | 51.7% | +3.2 | 0.721 | 29 · | |
| Breakable full fence | 51.0% | +2.5 | 0.710 | 49 · | |
| Fence and open | 50.0% | +1.5 | 0.898 | 18 · | |
| Single wall | 45.6% | -2.9 | 0.681 | 46 · | |
| Breakable half fence | 45.2% | -3.3 | 0.655 | 42 · | |
| Full fence | 43.8% | -4.8 | 0.575 | 32 · | |
| Octagon | 43.5% | -5.0 | 0.398 | 62 · | |
| Low fence | 40.4% | -8.1 | 0.211 | 52 · |
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 5.21 on 9 df, p 0.816
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 43.8%32 | -7.30.522 | -15.30.186 | -1.50.898 | +3.40.761 | -1.90.868 | -6.20.670 | -8.00.533 | -9.40.410 | +0.20.985 |
| Breakable full fence | +7.30.522 | 51.0%49 | -8.10.435 | +5.80.582 | +10.60.283 | +5.40.601 | +1.00.941 | -0.70.952 | -2.20.831 | +7.50.433 |
| Half fence | +15.30.186 | +8.10.435 | 59.1%44 | +13.80.199 | +18.70.068 | +13.40.202 | +9.10.512 | +7.40.535 | +5.90.571 | +15.50.115 |
| Breakable half fence | +1.50.898 | -5.80.582 | -13.80.199 | 45.2%42 | +4.90.636 | -0.40.969 | -4.80.735 | -6.50.591 | -8.00.454 | +1.70.865 |
| Low fence | -3.40.761 | -10.60.283 | -18.70.068 | -4.90.636 | 40.4%52 | -5.30.599 | -9.60.477 | -11.30.325 | -12.80.202 | -3.20.733 |
| Single wall | +1.90.868 | -5.40.601 | -13.40.202 | +0.40.969 | +5.30.599 | 45.6%46 | -4.30.754 | -6.10.608 | -7.50.467 | +2.10.828 |
| Fence and open | +6.20.670 | -1.00.941 | -9.10.512 | +4.80.735 | +9.60.477 | +4.30.754 | 50.0%18 | -1.70.908 | -3.20.818 | +6.50.628 |
| Open | +8.00.533 | +0.70.952 | -7.40.535 | +6.50.591 | +11.30.325 | +6.10.608 | +1.70.908 | 51.7%29 | -1.50.901 | +8.20.466 |
| Rectangle | +9.40.410 | +2.20.831 | -5.90.571 | +8.00.454 | +12.80.202 | +7.50.467 | +3.20.818 | +1.50.901 | 53.2%47 | +9.60.318 |
| Octagon | -0.20.985 | -7.50.433 | -15.50.115 | -1.70.865 | +3.20.733 | -2.10.828 | -6.50.628 | -8.20.466 | -9.60.318 | 43.5%62 |
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 counted28
Win rate42.9%
Widest swing-42.9 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 |
|---|---|---|---|---|---|
| Low fence | 80.0% | +37.1 | 0.064 | 5 · | |
| Breakable full fence | 66.7% | +23.8 | 0.378 | 3 · | |
| Octagon | 66.7% | +23.8 | 0.378 | 3 · | |
| Single wall | 50.0% | +7.1 | 0.832 | 2 · | |
| Half fence | 40.0% | -2.9 | 0.887 | 5 · | |
| Full fence | 25.0% | -17.9 | 0.436 | 4 · | |
| Breakable half fence | 0.0% | -42.9 | 0.112 | 3 · | |
| Rectangle | 0.0% | -42.9 | 0.204 | 2 · |
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 8.44 on 7 df, p 0.296
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Rect | Oct |
|---|---|---|---|---|---|---|---|---|
| Full fence | 25.0%4 | -41.70.270 | -15.00.635 | +25.00.350 | -55.00.099 | -25.00.540 | +25.00.439 | -41.70.270 |
| Breakable full fence | +41.70.270 | 66.7%3 | +26.70.465 | +66.70.083 | -13.30.673 | +16.70.709 | +66.70.136 | +0.01.000 |
| Half fence | +15.00.635 | -26.70.465 | 40.0%5 | +40.00.206 | -40.00.197 | -10.00.809 | +40.00.290 | -26.70.465 |
| Breakable half fence | -25.00.350 | -66.70.083 | -40.00.206 | 0.0%3 | -80.00.028 | -50.00.171 | +0.01.000 | -66.70.083 |
| Low fence | +55.00.099 | +13.30.673 | +40.00.197 | +80.00.028 | 80.0%5 | +30.00.427 | +80.00.053 | +13.30.673 |
| Single wall | +25.00.540 | -16.70.709 | +10.00.809 | +50.00.171 | -30.00.427 | 50.0%2 | +50.00.248 | -16.70.709 |
| Rectangle | -25.00.439 | -66.70.136 | -40.00.290 | -0.01.000 | -80.00.053 | -50.00.248 | 0.0%2 | -66.70.136 |
| Octagon | +41.70.270 | -0.01.000 | +26.70.465 | +66.70.083 | -13.30.673 | +16.70.709 | +66.70.136 | 66.7%3 |
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.