Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Eileen vs Taka
Matches counted246
Win rate53.2%
Widest swing-28.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Eileen's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 68.0% | +14.8 | 0.119 | 25 · | |
| Breakable half fence | 65.4% | +12.1 | 0.190 | 26 · | |
| Breakable full fence | 58.3% | +5.1 | 0.599 | 24 · | |
| Full fence | 51.9% | -1.4 | 0.877 | 27 · | |
| Octagon | 50.0% | -3.2 | 0.682 | 34 · | |
| Open | 50.0% | -3.2 | 0.737 | 24 · | |
| Rectangle | 50.0% | -3.2 | 0.737 | 24 · | |
| Single wall | 47.4% | -5.9 | 0.593 | 19 · | |
| Low fence | 41.2% | -12.1 | 0.301 | 17 · | |
| Fence and open | 25.0% | -28.2 | 0.044 | 12 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Eileen's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 9.39 on 9 df, p 0.402
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 51.9%27 | -6.50.642 | -16.10.236 | -13.50.318 | +10.70.490 | +4.50.765 | +26.90.119 | +1.90.895 | +1.90.895 | +1.90.886 |
| Breakable full fence | +6.50.642 | 58.3%24 | -9.70.483 | -7.00.608 | +17.10.279 | +11.00.474 | +33.30.059 | +8.30.562 | +8.30.562 | +8.30.531 |
| Half fence | +16.10.236 | +9.70.483 | 68.0%25 | +2.60.843 | +26.80.085 | +20.60.168 | +43.00.014 | +18.00.200 | +18.00.200 | +18.00.167 |
| Breakable half fence | +13.50.318 | +7.00.608 | -2.60.843 | 65.4%26 | +24.20.118 | +18.00.227 | +40.40.020 | +15.40.271 | +15.40.271 | +15.40.233 |
| Low fence | -10.70.490 | -17.10.279 | -26.80.085 | -24.20.118 | 41.2%17 | -6.20.709 | +16.20.367 | -8.80.577 | -8.80.577 | -8.80.552 |
| Single wall | -4.50.765 | -11.00.474 | -20.60.168 | -18.00.227 | +6.20.709 | 47.4%19 | +22.40.213 | -2.60.864 | -2.60.864 | -2.60.854 |
| Fence and open | -26.90.119 | -33.30.059 | -43.00.014 | -40.40.020 | -16.20.367 | -22.40.213 | 25.0%12 | -25.00.151 | -25.00.151 | -25.00.133 |
| Open | -1.90.895 | -8.30.562 | -18.00.200 | -15.40.271 | +8.80.577 | +2.60.864 | +25.00.151 | 50.0%24 | +0.01.000 | +0.01.000 |
| Rectangle | -1.90.895 | -8.30.562 | -18.00.200 | -15.40.271 | +8.80.577 | +2.60.864 | +25.00.151 | -0.01.000 | 50.0%24 | +0.01.000 |
| Octagon | -1.90.886 | -8.30.531 | -18.00.167 | -15.40.233 | +8.80.552 | +2.60.854 | +25.00.133 | -0.01.000 | -0.01.000 | 50.0%34 |
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 counted156
Win rate51.3%
Widest swing-51.3 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Eileen's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 78.6% | +27.3 | 0.032 | 14 · | |
| Breakable full fence | 64.7% | +13.4 | 0.241 | 17 · | |
| Breakable half fence | 61.5% | +10.3 | 0.440 | 13 · | |
| Octagon | 52.2% | +0.9 | 0.926 | 23 · | |
| Open | 47.1% | -4.2 | 0.712 | 17 · | |
| Low fence | 45.5% | -5.8 | 0.688 | 11 · | |
| Single wall | 45.5% | -5.8 | 0.688 | 11 · | |
| Full fence | 42.9% | -8.4 | 0.406 | 21 · | |
| Rectangle | 35.7% | -15.6 | 0.222 | 14 · | |
| Fence and open | 0.0% | -51.3 | 0.038 | 4 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Eileen's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 12.53 on 9 df, p 0.185
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 42.9%21 | -21.90.180 | -35.70.036 | -18.70.290 | -2.60.888 | -2.60.888 | +42.90.102 | -4.20.796 | +7.20.673 | -9.30.537 |
| Breakable full fence | +21.90.180 | 64.7%17 | -13.90.397 | +3.20.858 | +19.30.315 | +19.30.315 | +64.70.020 | +17.60.300 | +29.00.108 | +12.50.428 |
| Half fence | +35.70.036 | +13.90.397 | 78.6%14 | +17.00.333 | +33.10.087 | +33.10.087 | +78.60.004 | +31.50.073 | +42.90.022 | +26.40.108 |
| Breakable half fence | +18.70.290 | -3.20.858 | -17.00.333 | 61.5%13 | +16.10.431 | +16.10.431 | +61.50.031 | +14.50.431 | +25.80.180 | +9.40.587 |
| Low fence | +2.60.888 | -19.30.315 | -33.10.087 | -16.10.431 | 45.5%11 | +0.01.000 | +45.50.099 | -1.60.934 | +9.70.622 | -6.70.714 |
| Single wall | +2.60.888 | -19.30.315 | -33.10.087 | -16.10.431 | -0.01.000 | 45.5%11 | +45.50.099 | -1.60.934 | +9.70.622 | -6.70.714 |
| Fence and open | -42.90.102 | -64.70.020 | -78.60.004 | -61.50.031 | -45.50.099 | -45.50.099 | 0.0%4 | -47.10.081 | -35.70.160 | -52.20.053 |
| Open | +4.20.796 | -17.60.300 | -31.50.073 | -14.50.431 | +1.60.934 | +1.60.934 | +47.10.081 | 47.1%17 | +11.30.524 | -5.10.749 |
| Rectangle | -7.20.673 | -29.00.108 | -42.90.022 | -25.80.180 | -9.70.622 | -9.70.622 | +35.70.160 | -11.30.524 | 35.7%14 | -16.50.330 |
| Octagon | +9.30.537 | -12.50.428 | -26.40.108 | -9.40.587 | +6.70.714 | +6.70.714 | +52.20.053 | +5.10.749 | +16.50.330 | 52.2%23 |
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 counted90
Win rate56.7%
Widest swing26.7 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Eileen's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Full fence | 83.3% | +26.7 | 0.172 | 6 · | |
| Rectangle | 70.0% | +13.3 | 0.367 | 10 · | |
| Breakable half fence | 69.2% | +12.6 | 0.323 | 13 · | |
| Open | 57.1% | +0.5 | 0.979 | 7 · | |
| Half fence | 54.5% | -2.1 | 0.880 | 11 · | |
| Single wall | 50.0% | -6.7 | 0.690 | 8 · | |
| Octagon | 45.5% | -11.2 | 0.423 | 11 · | |
| Breakable full fence | 42.9% | -13.8 | 0.443 | 7 · | |
| Fence and open | 37.5% | -19.2 | 0.252 | 8 · | |
| Low fence | 33.3% | -23.3 | 0.233 | 6 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Eileen'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 0.641
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 83.3%6 | +40.50.135 | +28.80.235 | +14.10.516 | +50.00.079 | +33.30.198 | +45.80.086 | +26.20.308 | +13.30.551 | +37.90.129 |
| Breakable full fence | -40.50.135 | 42.9%7 | -11.70.629 | -26.40.251 | +9.50.725 | -7.10.782 | +5.40.833 | -14.30.593 | -27.10.263 | -2.60.914 |
| Half fence | -28.80.235 | +11.70.629 | 54.5%11 | -14.70.459 | +21.20.402 | +4.50.845 | +17.10.463 | -2.60.914 | -15.40.466 | +9.10.670 |
| Breakable half fence | -14.10.516 | +26.40.251 | +14.70.459 | 69.2%13 | +35.90.141 | +19.20.378 | +31.70.154 | +12.10.589 | -0.80.968 | +23.80.239 |
| Low fence | -50.00.079 | -9.50.725 | -21.20.402 | -35.90.141 | 33.3%6 | -16.70.533 | -4.20.872 | -23.80.391 | -36.70.152 | -12.10.627 |
| Single wall | -33.30.198 | +7.10.782 | -4.50.845 | -19.20.378 | +16.70.533 | 50.0%8 | +12.50.614 | -7.10.782 | -20.00.387 | +4.50.845 |
| Fence and open | -45.80.086 | -5.40.833 | -17.10.463 | -31.70.154 | +4.20.872 | -12.50.614 | 37.5%8 | -19.60.447 | -32.50.168 | -8.00.729 |
| Open | -26.20.308 | +14.30.593 | +2.60.914 | -12.10.589 | +23.80.391 | +7.10.782 | +19.60.447 | 57.1%7 | -12.90.585 | +11.70.629 |
| Rectangle | -13.30.551 | +27.10.263 | +15.40.466 | +0.80.968 | +36.70.152 | +20.00.387 | +32.50.168 | +12.90.585 | 70.0%10 | +24.60.256 |
| Octagon | -37.90.129 | +2.60.914 | -9.10.670 | -23.80.239 | +12.10.627 | -4.50.845 | +8.00.729 | -11.70.629 | -24.60.256 | 45.5%11 |
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.