Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Dural vs Jean
Matches counted425
Win rate52.9%
Widest swing11.2 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 | 64.1% | +11.2 | 0.143 | 39 · | |
| Breakable full fence | 61.4% | +8.4 | 0.237 | 44 · | |
| Octagon | 59.4% | +6.4 | 0.448 | 32 · | |
| Single wall | 53.2% | +0.2 | 0.971 | 47 · | |
| Full fence | 52.8% | -0.2 | 0.984 | 36 · | |
| Fence and open | 52.4% | -0.6 | 0.958 | 21 · | |
| Open | 50.0% | -2.9 | 0.650 | 52 · | |
| Low fence | 47.9% | -5.0 | 0.459 | 48 · | |
| Half fence | 44.9% | -8.0 | 0.230 | 49 · | |
| Rectangle | 44.2% | -8.8 | 0.225 | 43 · |
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 6.97 on 9 df, p 0.640
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.8%36 | -8.60.440 | +7.90.473 | -11.30.320 | +4.90.659 | -0.40.970 | +0.40.977 | +2.80.798 | +8.60.447 | -6.60.584 |
| Breakable full fence | +8.60.440 | 61.4%44 | +16.50.112 | -2.70.797 | +13.40.196 | +8.20.431 | +9.00.492 | +11.40.265 | +17.20.109 | +2.00.861 |
| Half fence | -7.90.473 | -16.50.112 | 44.9%49 | -19.20.073 | -3.00.766 | -8.30.416 | -7.50.565 | -5.10.608 | +0.70.945 | -14.50.203 |
| Breakable half fence | +11.30.320 | +2.70.797 | +19.20.073 | 64.1%39 | +16.20.131 | +10.90.307 | +11.70.377 | +14.10.180 | +19.90.071 | +4.70.683 |
| Low fence | -4.90.659 | -13.40.196 | +3.00.766 | -16.20.131 | 47.9%48 | -5.30.607 | -4.50.733 | -2.10.835 | +3.70.722 | -11.50.315 |
| Single wall | +0.40.970 | -8.20.431 | +8.30.416 | -10.90.307 | +5.30.607 | 53.2%47 | +0.80.951 | +3.20.751 | +9.00.393 | -6.20.587 |
| Fence and open | -0.40.977 | -9.00.492 | +7.50.565 | -11.70.377 | +4.50.733 | -0.80.951 | 52.4%21 | +2.40.854 | +8.20.537 | -7.00.615 |
| Open | -2.80.798 | -11.40.265 | +5.10.608 | -14.10.180 | +2.10.835 | -3.20.751 | -2.40.854 | 50.0%52 | +5.80.572 | -9.40.403 |
| Rectangle | -8.60.447 | -17.20.109 | -0.70.945 | -19.90.071 | -3.70.722 | -9.00.393 | -8.20.537 | -5.80.572 | 44.2%43 | -15.20.193 |
| Octagon | +6.60.584 | -2.00.861 | +14.50.203 | -4.70.683 | +11.50.315 | +6.20.587 | +7.00.615 | +9.40.403 | +15.20.193 | 59.4%32 |
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 counted413
Win rate52.5%
Widest swing8.6 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 | 61.1% | +8.6 | 0.281 | 36 · | |
| Breakable full fence | 61.0% | +8.4 | 0.255 | 41 · | |
| Octagon | 59.4% | +6.8 | 0.420 | 32 · | |
| Single wall | 53.2% | +0.7 | 0.925 | 47 · | |
| Full fence | 52.8% | +0.2 | 0.976 | 36 · | |
| Fence and open | 52.4% | -0.2 | 0.988 | 21 · | |
| Low fence | 48.9% | -3.6 | 0.603 | 45 · | |
| Open | 48.0% | -4.5 | 0.493 | 50 · | |
| Rectangle | 45.2% | -7.3 | 0.317 | 42 · | |
| Half fence | 44.9% | -7.6 | 0.254 | 49 · |
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 5.51 on 9 df, p 0.788
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.8%36 | -8.20.468 | +7.90.473 | -8.30.475 | +3.90.728 | -0.40.970 | +0.40.977 | +4.80.662 | +7.50.507 | -6.60.584 |
| Breakable full fence | +8.20.468 | 61.0%41 | +16.10.128 | -0.10.990 | +12.10.261 | +7.80.462 | +8.60.516 | +13.00.217 | +15.70.151 | +1.60.890 |
| Half fence | -7.90.473 | -16.10.128 | 44.9%49 | -16.20.139 | -4.00.698 | -8.30.416 | -7.50.565 | -3.10.757 | -0.30.974 | -14.50.203 |
| Breakable half fence | +8.30.475 | +0.10.990 | +16.20.139 | 61.1%36 | +12.20.273 | +7.90.471 | +8.70.520 | +13.10.229 | +15.90.162 | +1.70.884 |
| Low fence | -3.90.728 | -12.10.261 | +4.00.698 | -12.20.273 | 48.9%45 | -4.30.680 | -3.50.792 | +0.90.931 | +3.60.733 | -10.50.363 |
| Single wall | +0.40.970 | -7.80.462 | +8.30.416 | -7.90.471 | +4.30.680 | 53.2%47 | +0.80.951 | +5.20.609 | +8.00.454 | -6.20.587 |
| Fence and open | -0.40.977 | -8.60.516 | +7.50.565 | -8.70.520 | +3.50.792 | -0.80.951 | 52.4%21 | +4.40.736 | +7.10.593 | -7.00.615 |
| Open | -4.80.662 | -13.00.217 | +3.10.757 | -13.10.229 | -0.90.931 | -5.20.609 | -4.40.736 | 48.0%50 | +2.80.791 | -11.40.314 |
| Rectangle | -7.50.507 | -15.70.151 | +0.30.974 | -15.90.162 | -3.60.733 | -8.00.454 | -7.10.593 | -2.80.791 | 45.2%42 | -14.10.228 |
| Octagon | +6.60.584 | -1.60.890 | +14.50.203 | -1.70.884 | +10.50.363 | +6.20.587 | +7.00.615 | +11.40.314 | +14.10.228 | 59.4%32 |
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 counted12
Win rate66.7%
Widest swing-66.7 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 | 100.0% | +33.3 | 0.157 | 3 · | |
| Open | 100.0% | +33.3 | 0.273 | 2 · | |
| Breakable full fence | 66.7% | +0.0 | 1.000 | 3 · | |
| Low fence | 33.3% | -33.3 | 0.157 | 3 · | |
| Rectangle | 0.0% | -66.7 | 0.140 | 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 6.00 on 4 df, p 0.199
| Stage | Break. full | Break. half | Low | Open | Rect |
|---|---|---|---|---|---|
| Breakable full fence | 66.7%3 | -33.30.273 | +33.30.414 | -33.30.361 | +66.70.248 |
| Breakable half fence | +33.30.273 | 100.0%3 | +66.70.083 | +0.01.000 | +100.00.046 |
| Low fence | -33.30.414 | -66.70.083 | 33.3%3 | -66.70.136 | +33.30.505 |
| Open | +33.30.361 | -0.01.000 | +66.70.136 | 100.0%2 | +100.00.083 |
| Rectangle | -66.70.248 | -100.00.046 | -33.30.505 | -100.00.083 | 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.
