Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Dural vs Sarah
Matches counted407
Win rate48.4%
Widest swing-29.6 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 |
|---|---|---|---|---|---|
| Full fence | 60.0% | +11.6 | 0.151 | 35 · | |
| Rectangle | 58.5% | +10.1 | 0.171 | 41 · | |
| Breakable full fence | 50.0% | +1.6 | 0.841 | 36 · | |
| Open | 50.0% | +1.6 | 0.789 | 60 · | |
| Low fence | 48.8% | +0.4 | 0.952 | 43 · | |
| Half fence | 48.6% | +0.2 | 0.975 | 37 · | |
| Octagon | 44.7% | -3.7 | 0.635 | 38 · | |
| Breakable half fence | 42.9% | -5.5 | 0.448 | 42 · | |
| Single wall | 38.5% | -9.9 | 0.191 | 39 · | |
| Fence and open | 18.8% | -29.6 | 0.015 | 16 · |
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 11.53 on 9 df, p 0.241
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 60.0%35 | +10.00.397 | +11.30.334 | +17.10.134 | +11.20.325 | +21.50.064 | +41.20.006 | +10.00.346 | +1.50.897 | +15.30.192 |
| Breakable full fence | -10.00.397 | 50.0%36 | +1.40.908 | +7.10.528 | +1.20.918 | +11.50.315 | +31.20.034 | +0.01.000 | -8.50.453 | +5.30.650 |
| Half fence | -11.30.334 | -1.40.908 | 48.6%37 | +5.80.606 | -0.20.987 | +10.20.370 | +29.90.041 | -1.40.897 | -9.90.382 | +3.90.734 |
| Breakable half fence | -17.10.134 | -7.10.528 | -5.80.606 | 42.9%42 | -6.00.580 | +4.40.687 | +24.10.088 | -7.10.477 | -15.70.153 | -1.90.866 |
| Low fence | -11.20.325 | -1.20.918 | +0.20.987 | +6.00.580 | 48.8%43 | +10.40.344 | +30.10.036 | -1.20.907 | -9.70.373 | +4.10.712 |
| Single wall | -21.50.064 | -11.50.315 | -10.20.370 | -4.40.687 | -10.40.344 | 38.5%39 | +19.70.157 | -11.50.260 | -20.10.073 | -6.30.576 |
| Fence and open | -41.20.006 | -31.20.034 | -29.90.041 | -24.10.088 | -30.10.036 | -19.70.157 | 18.8%16 | -31.20.025 | -39.80.007 | -26.00.071 |
| Open | -10.00.346 | -0.01.000 | +1.40.897 | +7.10.477 | +1.20.907 | +11.50.260 | +31.20.025 | 50.0%60 | -8.50.398 | +5.30.611 |
| Rectangle | -1.50.897 | +8.50.453 | +9.90.382 | +15.70.153 | +9.70.373 | +20.10.073 | +39.80.007 | +8.50.398 | 58.5%41 | +13.80.220 |
| Octagon | -15.30.192 | -5.30.650 | -3.90.734 | +1.90.866 | -4.10.712 | +6.30.576 | +26.00.071 | -5.30.611 | -13.80.220 | 44.7%38 |
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 counted368
Win rate47.5%
Widest swing-26.1 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 | 61.3% | +13.7 | 0.110 | 31 · | |
| Rectangle | 57.9% | +10.3 | 0.178 | 38 · | |
| Low fence | 48.7% | +1.2 | 0.878 | 39 · | |
| Octagon | 48.6% | +1.0 | 0.899 | 35 · | |
| Open | 47.3% | -0.3 | 0.964 | 55 · | |
| Breakable full fence | 47.1% | -0.5 | 0.952 | 34 · | |
| Breakable half fence | 42.9% | -4.7 | 0.559 | 35 · | |
| Half fence | 42.4% | -5.1 | 0.536 | 33 · | |
| Single wall | 40.5% | -7.0 | 0.368 | 37 · | |
| Fence and open | 21.4% | -26.1 | 0.046 | 14 · |
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 9.23 on 9 df, p 0.416
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 61.3%31 | +14.20.250 | +18.90.131 | +18.40.135 | +12.60.294 | +20.80.088 | +39.90.013 | +14.00.211 | +3.40.775 | +12.70.300 |
| Breakable full fence | -14.20.250 | 47.1%34 | +4.60.703 | +4.20.726 | -1.70.887 | +6.50.580 | +25.60.099 | -0.20.984 | -10.80.358 | -1.50.900 |
| Half fence | -18.90.131 | -4.60.703 | 42.4%33 | -0.40.971 | -6.30.593 | +1.90.873 | +21.00.171 | -4.80.658 | -15.50.193 | -6.20.611 |
| Breakable half fence | -18.40.135 | -4.20.726 | +0.40.971 | 42.9%35 | -5.90.613 | +2.30.842 | +21.40.160 | -4.40.682 | -15.00.199 | -5.70.631 |
| Low fence | -12.60.294 | +1.70.887 | +6.30.593 | +5.90.613 | 48.7%39 | +8.20.474 | +27.30.075 | +1.40.890 | -9.20.420 | +0.10.990 |
| Single wall | -20.80.088 | -6.50.580 | -1.90.873 | -2.30.842 | -8.20.474 | 40.5%37 | +19.10.202 | -6.70.524 | -17.40.133 | -8.00.493 |
| Fence and open | -39.90.013 | -25.60.099 | -21.00.171 | -21.40.160 | -27.30.075 | -19.10.202 | 21.4%14 | -25.80.080 | -36.50.020 | -27.10.081 |
| Open | -14.00.211 | +0.20.984 | +4.80.658 | +4.40.682 | -1.40.890 | +6.70.524 | +25.80.080 | 47.3%55 | -10.60.314 | -1.30.904 |
| Rectangle | -3.40.775 | +10.80.358 | +15.50.193 | +15.00.199 | +9.20.420 | +17.40.133 | +36.50.020 | +10.60.314 | 57.9%38 | +9.30.425 |
| Octagon | -12.70.300 | +1.50.900 | +6.20.611 | +5.70.631 | -0.10.990 | +8.00.493 | +27.10.081 | +1.30.904 | -9.30.425 | 48.6%35 |
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 counted39
Win rate56.4%
Widest swing-56.4 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 full fence | 100.0% | +43.6 | 0.202 | 2 · | |
| Half fence | 100.0% | +43.6 | 0.063 | 4 · | |
| Open | 80.0% | +23.6 | 0.255 | 5 · | |
| Rectangle | 66.7% | +10.3 | 0.709 | 3 · | |
| Full fence | 50.0% | -6.4 | 0.785 | 4 · | |
| Low fence | 50.0% | -6.4 | 0.785 | 4 · | |
| Breakable half fence | 42.9% | -13.6 | 0.425 | 7 · | |
| Fence and open | 0.0% | -56.4 | 0.099 | 2 · | |
| Octagon | 0.0% | -56.4 | 0.040 | 3 · | |
| Single wall | 0.0% | -56.4 | 0.099 | 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 15.21 on 9 df, p 0.085
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%4 | -50.00.221 | -50.00.102 | +7.10.819 | +0.01.000 | +50.00.221 | +50.00.221 | -30.00.343 | -16.70.659 | +50.00.147 |
| Breakable full fence | +50.00.221 | 100.0%2 | +0.01.000 | +57.10.151 | +50.00.221 | +100.00.046 | +100.00.046 | +20.00.495 | +33.30.361 | +100.00.025 |
| Half fence | +50.00.102 | -0.01.000 | 100.0%4 | +57.10.058 | +50.00.102 | +100.00.014 | +100.00.014 | +20.00.343 | +33.30.212 | +100.00.008 |
| Breakable half fence | -7.10.819 | -57.10.151 | -57.10.058 | 42.9%7 | -7.10.819 | +42.90.257 | +42.90.257 | -37.10.198 | -23.80.490 | +42.90.175 |
| Low fence | -0.01.000 | -50.00.221 | -50.00.102 | +7.10.819 | 50.0%4 | +50.00.221 | +50.00.221 | -30.00.343 | -16.70.659 | +50.00.147 |
| Single wall | -50.00.221 | -100.00.046 | -100.00.014 | -42.90.257 | -50.00.221 | 0.0%2 | +0.01.000 | -80.00.053 | -66.70.136 | +0.01.000 |
| Fence and open | -50.00.221 | -100.00.046 | -100.00.014 | -42.90.257 | -50.00.221 | -0.01.000 | 0.0%2 | -80.00.053 | -66.70.136 | +0.01.000 |
| Open | +30.00.343 | -20.00.495 | -20.00.343 | +37.10.198 | +30.00.343 | +80.00.053 | +80.00.053 | 80.0%5 | +13.30.673 | +80.00.028 |
| Rectangle | +16.70.659 | -33.30.361 | -33.30.212 | +23.80.490 | +16.70.659 | +66.70.136 | +66.70.136 | -13.30.673 | 66.7%3 | +66.70.083 |
| Octagon | -50.00.147 | -100.00.025 | -100.00.008 | -42.90.175 | -50.00.147 | -0.01.000 | -0.01.000 | -80.00.028 | -66.70.083 | 0.0%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.