Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Eileen vs Pai
Matches counted549
Win rate55.4%
Widest swing-10.3 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 | 64.8% | +9.4 | 0.142 | 54 · | |
| Open | 63.3% | +8.0 | 0.189 | 60 · | |
| Low fence | 62.3% | +6.9 | 0.288 | 53 · | |
| Full fence | 57.1% | +1.8 | 0.794 | 49 · | |
| Rectangle | 53.5% | -1.9 | 0.796 | 43 · | |
| Octagon | 51.9% | -3.5 | 0.599 | 52 · | |
| Breakable full fence | 49.1% | -6.2 | 0.316 | 57 · | |
| Breakable half fence | 46.1% | -9.2 | 0.111 | 65 · | |
| Fence and open | 46.1% | -9.2 | 0.333 | 26 · | |
| Single wall | 45.1% | -10.3 | 0.121 | 51 · |
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 10.81 on 9 df, p 0.289
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.1%49 | +8.00.410 | -7.70.425 | +11.00.245 | -5.10.598 | +12.00.228 | +11.00.364 | -6.20.511 | +3.60.725 | +5.20.599 |
| Breakable full fence | -8.00.410 | 49.1%57 | -15.70.095 | +3.00.743 | -13.10.166 | +4.00.676 | +3.00.802 | -14.20.121 | -4.40.665 | -2.80.770 |
| Half fence | +7.70.425 | +15.70.095 | 64.8%54 | +18.70.042 | +2.50.784 | +19.70.042 | +18.70.112 | +1.50.869 | +11.30.258 | +12.90.178 |
| Breakable half fence | -11.00.245 | -3.00.743 | -18.70.042 | 46.1%65 | -16.10.081 | +1.10.910 | +0.01.000 | -17.20.054 | -7.30.455 | -5.80.535 |
| Low fence | +5.10.598 | +13.10.166 | -2.50.784 | +16.10.081 | 62.3%53 | +17.20.079 | +16.10.174 | -1.10.907 | +8.80.386 | +10.30.284 |
| Single wall | -12.00.228 | -4.00.676 | -19.70.042 | -1.10.910 | -17.20.079 | 45.1%51 | -1.10.930 | -18.20.054 | -8.40.418 | -6.80.488 |
| Fence and open | -11.00.364 | -3.00.802 | -18.70.112 | -0.01.000 | -16.10.174 | +1.10.930 | 46.1%26 | -17.20.138 | -7.30.555 | -5.80.631 |
| Open | +6.20.511 | +14.20.121 | -1.50.869 | +17.20.054 | +1.10.907 | +18.20.054 | +17.20.138 | 63.3%60 | +9.80.316 | +11.40.222 |
| Rectangle | -3.60.725 | +4.40.665 | -11.30.258 | +7.30.455 | -8.80.386 | +8.40.418 | +7.30.555 | -9.80.316 | 53.5%43 | +1.60.879 |
| Octagon | -5.20.599 | +2.80.770 | -12.90.178 | +5.80.535 | -10.30.284 | +6.80.488 | +5.80.631 | -11.40.222 | -1.60.879 | 51.9%52 |
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 counted381
Win rate53.3%
Widest swing-13.9 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 |
|---|---|---|---|---|---|
| Open | 65.8% | +12.5 | 0.103 | 38 · | |
| Half fence | 62.9% | +9.6 | 0.233 | 35 · | |
| Full fence | 61.3% | +8.0 | 0.351 | 31 · | |
| Low fence | 57.5% | +4.2 | 0.572 | 40 · | |
| Single wall | 51.4% | -1.9 | 0.804 | 37 · | |
| Octagon | 48.6% | -4.6 | 0.552 | 37 · | |
| Breakable half fence | 46.7% | -6.6 | 0.344 | 45 · | |
| Rectangle | 43.8% | -9.5 | 0.259 | 32 · | |
| Fence and open | 43.5% | -9.8 | 0.331 | 23 · | |
| Breakable full fence | 39.4% | -13.9 | 0.094 | 33 · |
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 10.42 on 9 df, p 0.318
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 61.3%31 | +21.90.080 | -1.60.896 | +14.60.210 | +3.80.747 | +9.90.411 | +17.80.194 | -4.50.699 | +17.50.163 | +12.60.297 |
| Breakable full fence | -21.90.080 | 39.4%33 | -23.50.053 | -7.30.522 | -18.10.124 | -12.00.316 | -4.10.760 | -26.40.026 | -4.40.722 | -9.30.436 |
| Half fence | +1.60.896 | +23.50.053 | 62.9%35 | +16.20.150 | +5.40.637 | +11.50.324 | +19.40.147 | -2.90.794 | +19.10.117 | +14.20.225 |
| Breakable half fence | -14.60.210 | +7.30.522 | -16.20.150 | 46.7%45 | -10.80.318 | -4.70.673 | +3.20.803 | -19.10.081 | +2.90.800 | -2.00.858 |
| Low fence | -3.80.747 | +18.10.124 | -5.40.637 | +10.80.318 | 57.5%40 | +6.20.588 | +14.00.283 | -8.30.452 | +13.80.246 | +8.80.437 |
| Single wall | -9.90.411 | +12.00.316 | -11.50.324 | +4.70.673 | -6.20.588 | 51.4%37 | +7.90.553 | -14.40.204 | +7.60.528 | +2.70.816 |
| Fence and open | -17.80.194 | +4.10.760 | -19.40.147 | -3.20.803 | -14.00.283 | -7.90.553 | 43.5%23 | -22.30.088 | -0.30.984 | -5.20.696 |
| Open | +4.50.699 | +26.40.026 | +2.90.794 | +19.10.081 | +8.30.452 | +14.40.204 | +22.30.088 | 65.8%38 | +22.00.064 | +17.10.133 |
| Rectangle | -17.50.163 | +4.40.722 | -19.10.117 | -2.90.800 | -13.80.246 | -7.60.528 | +0.30.984 | -22.00.064 | 43.8%32 | -4.90.684 |
| Octagon | -12.60.297 | +9.30.436 | -14.20.225 | +2.00.858 | -8.80.437 | -2.70.816 | +5.20.696 | -17.10.133 | +4.90.684 | 48.6%37 |
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 counted168
Win rate60.1%
Widest swing-31.6 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 |
|---|---|---|---|---|---|
| Rectangle | 81.8% | +21.7 | 0.128 | 11 · | |
| Low fence | 76.9% | +16.8 | 0.198 | 13 · | |
| Half fence | 68.4% | +8.3 | 0.433 | 19 · | |
| Fence and open | 66.7% | +6.5 | 0.815 | 3 · | |
| Breakable full fence | 62.5% | +2.4 | 0.797 | 24 · | |
| Octagon | 60.0% | -0.1 | 0.992 | 15 · | |
| Open | 59.1% | -1.0 | 0.916 | 22 · | |
| Full fence | 50.0% | -10.1 | 0.353 | 18 · | |
| Breakable half fence | 45.0% | -15.1 | 0.141 | 20 · | |
| Single wall | 28.6% | -31.6 | 0.012 | 14 · |
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 12.51 on 9 df, p 0.186
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%18 | -12.50.418 | -18.40.254 | +5.00.758 | -26.90.129 | +21.40.221 | -16.70.593 | -9.10.565 | -31.80.087 | -10.00.566 |
| Breakable full fence | +12.50.418 | 62.5%24 | -5.90.686 | +17.50.246 | -14.40.371 | +33.90.044 | -4.20.888 | +3.40.813 | -19.30.253 | +2.50.876 |
| Half fence | +18.40.254 | +5.90.686 | 68.4%19 | +23.40.140 | -8.50.599 | +39.90.024 | +1.80.952 | +9.30.536 | -13.40.424 | +8.40.610 |
| Breakable half fence | -5.00.758 | -17.50.246 | -23.40.140 | 45.0%20 | -31.90.070 | +16.40.332 | -21.70.484 | -14.10.361 | -36.80.047 | -15.00.380 |
| Low fence | +26.90.129 | +14.40.371 | +8.50.599 | +31.90.070 | 76.9%13 | +48.40.012 | +10.20.712 | +17.80.283 | -4.90.769 | +16.90.339 |
| Single wall | -21.40.221 | -33.90.044 | -39.90.024 | -16.40.332 | -48.40.012 | 28.6%14 | -38.10.210 | -30.50.074 | -53.20.008 | -31.40.089 |
| Fence and open | +16.70.593 | +4.20.888 | -1.80.952 | +21.70.484 | -10.20.712 | +38.10.210 | 66.7%3 | +7.60.802 | -15.20.571 | +6.70.829 |
| Open | +9.10.565 | -3.40.813 | -9.30.536 | +14.10.361 | -17.80.283 | +30.50.074 | -7.60.802 | 59.1%22 | -22.70.192 | -0.90.956 |
| Rectangle | +31.80.087 | +19.30.253 | +13.40.424 | +36.80.047 | +4.90.769 | +53.20.008 | +15.20.571 | +22.70.192 | 81.8%11 | +21.80.234 |
| Octagon | +10.00.566 | -2.50.876 | -8.40.610 | +15.00.380 | -16.90.339 | +31.40.089 | -6.70.829 | +0.90.956 | -21.80.234 | 60.0%15 |
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.