Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Eileen vs Lei Fei
Matches counted468
Win rate47.2%
Widest swing8.6 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 | 55.8% | +8.6 | 0.190 | 52 · | |
| Rectangle | 55.6% | +8.3 | 0.297 | 36 · | |
| Fence and open | 50.0% | +2.8 | 0.821 | 16 · | |
| Breakable half fence | 47.4% | +0.1 | 0.985 | 38 · | |
| Open | 45.9% | -1.3 | 0.825 | 61 · | |
| Full fence | 45.2% | -2.0 | 0.707 | 73 · | |
| Breakable full fence | 43.4% | -3.8 | 0.554 | 53 · | |
| Octagon | 42.5% | -4.7 | 0.499 | 47 · | |
| Single wall | 41.7% | -5.5 | 0.487 | 36 · | |
| Low fence | 40.6% | -6.6 | 0.439 | 32 · |
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 4.41 on 9 df, p 0.883
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 45.2%73 | +1.80.840 | -10.60.244 | -2.20.828 | +4.60.663 | +3.50.726 | -4.80.727 | -0.70.936 | -10.30.309 | +2.70.775 |
| Breakable full fence | -1.80.840 | 43.4%53 | -12.40.205 | -4.00.707 | +2.80.802 | +1.70.871 | -6.60.642 | -2.50.788 | -12.20.260 | +0.80.932 |
| Half fence | +10.60.244 | +12.40.205 | 55.8%52 | +8.40.431 | +15.20.178 | +14.10.193 | +5.80.685 | +9.90.296 | +0.20.984 | +13.20.189 |
| Breakable half fence | +2.20.828 | +4.00.707 | -8.40.431 | 47.4%38 | +6.80.572 | +5.70.622 | -2.60.860 | +1.50.887 | -8.20.481 | +4.80.657 |
| Low fence | -4.60.663 | -2.80.802 | -15.20.178 | -6.80.572 | 40.6%32 | -1.10.931 | -9.40.537 | -5.30.626 | -14.90.219 | -1.90.865 |
| Single wall | -3.50.726 | -1.70.871 | -14.10.193 | -5.70.622 | +1.10.931 | 41.7%36 | -8.30.577 | -4.20.685 | -13.90.238 | -0.90.935 |
| Fence and open | +4.80.727 | +6.60.642 | -5.80.685 | +2.60.860 | +9.40.537 | +8.30.577 | 50.0%16 | +4.10.770 | -5.60.711 | +7.50.605 |
| Open | +0.70.936 | +2.50.788 | -9.90.296 | -1.50.887 | +5.30.626 | +4.20.685 | -4.10.770 | 45.9%61 | -9.70.358 | +3.40.728 |
| Rectangle | +10.30.309 | +12.20.260 | -0.20.984 | +8.20.481 | +14.90.219 | +13.90.238 | +5.60.711 | +9.70.358 | 55.6%36 | +13.00.240 |
| Octagon | -2.70.775 | -0.80.932 | -13.20.189 | -4.80.657 | +1.90.865 | +0.90.935 | -7.50.605 | -3.40.728 | -13.00.240 | 42.5%47 |
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 counted295
Win rate51.2%
Widest swing-15.9 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 |
|---|---|---|---|---|---|
| Fence and open | 62.5% | +11.3 | 0.516 | 8 · | |
| Half fence | 59.4% | +8.2 | 0.326 | 32 · | |
| Octagon | 57.1% | +6.0 | 0.507 | 28 · | |
| Full fence | 55.8% | +4.6 | 0.466 | 52 · | |
| Open | 53.3% | +2.1 | 0.804 | 30 · | |
| Breakable half fence | 48.3% | -2.9 | 0.741 | 29 · | |
| Rectangle | 47.8% | -3.4 | 0.737 | 23 · | |
| Breakable full fence | 47.4% | -3.8 | 0.614 | 38 · | |
| Single wall | 38.1% | -13.1 | 0.213 | 21 · | |
| Low fence | 35.3% | -15.9 | 0.177 | 17 · |
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 5.74 on 9 df, p 0.766
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 55.8%52 | +8.40.431 | -3.60.746 | +7.50.517 | +20.50.143 | +17.70.172 | -6.70.721 | +2.40.831 | +7.90.525 | -1.40.906 |
| Breakable full fence | -8.40.431 | 47.4%38 | -12.00.316 | -0.90.941 | +12.10.404 | +9.30.492 | -15.10.437 | -6.00.625 | -0.50.972 | -9.80.432 |
| Half fence | +3.60.746 | +12.00.316 | 59.4%32 | +11.10.385 | +24.10.108 | +21.30.130 | -3.10.872 | +6.00.632 | +11.60.396 | +2.20.861 |
| Breakable half fence | -7.50.517 | +0.90.941 | -11.10.385 | 48.3%29 | +13.00.391 | +10.20.474 | -14.20.476 | -5.00.698 | +0.50.974 | -8.90.503 |
| Low fence | -20.50.143 | -12.10.404 | -24.10.108 | -13.00.391 | 35.3%17 | -2.80.859 | -27.20.201 | -18.00.234 | -12.50.428 | -21.90.155 |
| Single wall | -17.70.172 | -9.30.492 | -21.30.130 | -10.20.474 | +2.80.859 | 38.1%21 | -24.40.238 | -15.20.283 | -9.70.515 | -19.00.187 |
| Fence and open | +6.70.721 | +15.10.437 | +3.10.872 | +14.20.476 | +27.20.201 | +24.40.238 | 62.5%8 | +9.20.643 | +14.70.474 | +5.40.786 |
| Open | -2.40.831 | +6.00.625 | -6.00.632 | +5.00.698 | +18.00.234 | +15.20.283 | -9.20.643 | 53.3%30 | +5.50.691 | -3.80.771 |
| Rectangle | -7.90.525 | +0.50.972 | -11.60.396 | -0.50.974 | +12.50.428 | +9.70.515 | -14.70.474 | -5.50.691 | 47.8%23 | -9.30.507 |
| Octagon | +1.40.906 | +9.80.432 | -2.20.861 | +8.90.503 | +21.90.155 | +19.00.187 | -5.40.786 | +3.80.771 | +9.30.507 | 57.1%28 |
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 counted173
Win rate40.5%
Widest swing28.8 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 |
|---|---|---|---|---|---|
| Rectangle | 69.2% | +28.8 | 0.028 | 13 · | |
| Half fence | 50.0% | +9.5 | 0.355 | 20 · | |
| Low fence | 46.7% | +6.2 | 0.608 | 15 · | |
| Single wall | 46.7% | +6.2 | 0.608 | 15 · | |
| Breakable half fence | 44.4% | +4.0 | 0.803 | 9 · | |
| Open | 38.7% | -1.8 | 0.826 | 31 · | |
| Fence and open | 37.5% | -3.0 | 0.861 | 8 · | |
| Breakable full fence | 33.3% | -7.1 | 0.556 | 15 · | |
| Octagon | 21.1% | -19.4 | 0.068 | 19 · | |
| Full fence | 19.1% | -21.4 | 0.033 | 21 · |
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 13.14 on 9 df, p 0.156
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 19.1%21 | -14.30.329 | -30.90.037 | -25.40.149 | -27.60.076 | -27.60.076 | -18.40.299 | -19.70.132 | -50.20.003 | -2.00.874 |
| Breakable full fence | +14.30.329 | 33.3%15 | -16.70.324 | -11.10.586 | -13.30.456 | -13.30.456 | -4.20.842 | -5.40.723 | -35.90.058 | +12.30.420 |
| Half fence | +30.90.037 | +16.70.324 | 50.0%20 | +5.60.782 | +3.30.845 | +3.30.845 | +12.50.549 | +11.30.427 | -19.20.275 | +28.90.060 |
| Breakable half fence | +25.40.149 | +11.10.586 | -5.60.782 | 44.4%9 | -2.20.916 | -2.20.916 | +6.90.772 | +5.70.757 | -24.80.245 | +23.40.201 |
| Low fence | +27.60.076 | +13.30.456 | -3.30.845 | +2.20.916 | 46.7%15 | +0.01.000 | +9.20.673 | +8.00.607 | -22.60.229 | +25.60.113 |
| Single wall | +27.60.076 | +13.30.456 | -3.30.845 | +2.20.916 | -0.01.000 | 46.7%15 | +9.20.673 | +8.00.607 | -22.60.229 | +25.60.113 |
| Fence and open | +18.40.299 | +4.20.842 | -12.50.549 | -6.90.772 | -9.20.673 | -9.20.673 | 37.5%8 | -1.20.950 | -31.70.154 | +16.40.373 |
| Open | +19.70.132 | +5.40.723 | -11.30.427 | -5.70.757 | -8.00.607 | -8.00.607 | +1.20.950 | 38.7%31 | -30.50.064 | +17.70.194 |
| Rectangle | +50.20.003 | +35.90.058 | +19.20.275 | +24.80.245 | +22.60.229 | +22.60.229 | +31.70.154 | +30.50.064 | 69.2%13 | +48.20.006 |
| Octagon | +2.00.874 | -12.30.420 | -28.90.060 | -23.40.201 | -25.60.113 | -25.60.113 | -16.40.373 | -17.70.194 | -48.20.006 | 21.1%19 |
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.