Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Eileen vs Sarah
Matches counted725
Win rate55.0%
Widest swing5.3 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 | 60.3% | +5.3 | 0.397 | 58 · | |
| Octagon | 58.3% | +3.3 | 0.553 | 72 · | |
| Single wall | 58.0% | +2.9 | 0.557 | 88 | |
| Low fence | 57.8% | +2.8 | 0.640 | 64 · | |
| Full fence | 57.5% | +2.4 | 0.731 | 47 · | |
| Fence and open | 56.8% | +1.8 | 0.806 | 44 · | |
| Breakable half fence | 54.9% | -0.1 | 0.985 | 71 · | |
| Breakable full fence | 54.3% | -0.7 | 0.891 | 81 | |
| Open | 50.8% | -4.3 | 0.459 | 67 · | |
| Rectangle | 50.0% | -5.0 | 0.305 | 90 |
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 3.01 on 9 df, p 0.964
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.5%47 | +3.10.732 | -2.90.764 | +2.50.787 | -0.40.969 | -0.50.955 | +0.60.952 | +6.70.480 | +7.50.407 | -0.90.924 |
| Breakable full fence | -3.10.732 | 54.3%81 | -6.00.480 | -0.60.940 | -3.50.674 | -3.60.634 | -2.50.789 | +3.60.665 | +4.30.572 | -4.00.618 |
| Half fence | +2.90.764 | +6.00.480 | 60.3%58 | +5.40.536 | +2.50.776 | +2.40.774 | +3.50.720 | +9.60.282 | +10.30.218 | +2.00.817 |
| Breakable half fence | -2.50.787 | +0.60.940 | -5.40.536 | 54.9%71 | -2.90.736 | -3.00.702 | -1.90.843 | +4.20.623 | +4.90.534 | -3.40.681 |
| Low fence | +0.40.969 | +3.50.674 | -2.50.776 | +2.90.736 | 57.8%64 | -0.10.986 | +1.00.918 | +7.10.417 | +7.80.338 | -0.50.951 |
| Single wall | +0.50.955 | +3.60.634 | -2.40.774 | +3.00.702 | +0.10.986 | 58.0%88 | +1.10.901 | +7.20.372 | +8.00.287 | -0.40.961 |
| Fence and open | -0.60.952 | +2.50.789 | -3.50.720 | +1.90.843 | -1.00.918 | -1.10.901 | 56.8%44 | +6.10.531 | +6.80.458 | -1.50.873 |
| Open | -6.70.480 | -3.60.665 | -9.60.282 | -4.20.623 | -7.10.417 | -7.20.372 | -6.10.531 | 50.8%67 | +0.80.926 | -7.60.369 |
| Rectangle | -7.50.407 | -4.30.572 | -10.30.218 | -4.90.534 | -7.80.338 | -8.00.287 | -6.80.458 | -0.80.926 | 50.0%90 | -8.30.291 |
| Octagon | +0.90.924 | +4.00.618 | -2.00.817 | +3.40.681 | +0.50.951 | +0.40.961 | +1.50.873 | +7.60.369 | +8.30.291 | 58.3%72 |
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 counted530
Win rate52.1%
Widest swing-7.6 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 |
|---|---|---|---|---|---|
| Single wall | 57.8% | +5.7 | 0.304 | 71 · | |
| Half fence | 57.1% | +5.1 | 0.493 | 42 · | |
| Low fence | 54.9% | +2.8 | 0.671 | 51 · | |
| Fence and open | 54.5% | +2.5 | 0.769 | 33 · | |
| Breakable half fence | 54.4% | +2.3 | 0.712 | 57 · | |
| Breakable full fence | 51.9% | -0.2 | 0.982 | 52 · | |
| Octagon | 51.3% | -0.8 | 0.918 | 39 · | |
| Full fence | 50.0% | -2.1 | 0.808 | 32 · | |
| Rectangle | 47.6% | -4.5 | 0.451 | 63 · | |
| Open | 44.4% | -7.6 | 0.236 | 54 · |
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 3.52 on 9 df, p 0.940
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%32 | -1.90.864 | -7.10.541 | -4.40.691 | -4.90.663 | -7.80.464 | -4.50.714 | +5.60.618 | +2.40.826 | -1.30.914 |
| Breakable full fence | +1.90.864 | 51.9%52 | -5.20.614 | -2.50.797 | -3.00.762 | -5.80.521 | -2.60.813 | +7.50.441 | +4.30.646 | +0.60.952 |
| Half fence | +7.10.541 | +5.20.614 | 57.1%42 | +2.80.785 | +2.20.829 | -0.60.950 | +2.60.822 | +12.70.217 | +9.50.339 | +5.90.597 |
| Breakable half fence | +4.40.691 | +2.50.797 | -2.80.785 | 54.4%57 | -0.50.957 | -3.40.703 | -0.20.988 | +9.90.295 | +6.80.459 | +3.10.765 |
| Low fence | +4.90.663 | +3.00.762 | -2.20.829 | +0.50.957 | 54.9%51 | -2.90.755 | +0.30.974 | +10.50.284 | +7.30.439 | +3.60.733 |
| Single wall | +7.80.464 | +5.80.521 | +0.60.950 | +3.40.703 | +2.90.755 | 57.8%71 | +3.20.759 | +13.30.140 | +10.10.241 | +6.50.514 |
| Fence and open | +4.50.714 | +2.60.813 | -2.60.822 | +0.20.988 | -0.30.974 | -3.20.759 | 54.5%33 | +10.10.360 | +6.90.519 | +3.30.782 |
| Open | -5.60.618 | -7.50.441 | -12.70.217 | -9.90.295 | -10.50.284 | -13.30.140 | -10.10.360 | 44.4%54 | -3.20.731 | -6.80.515 |
| Rectangle | -2.40.826 | -4.30.646 | -9.50.339 | -6.80.459 | -7.30.439 | -10.10.241 | -6.90.519 | +3.20.731 | 47.6%63 | -3.70.719 |
| Octagon | +1.30.914 | -0.60.952 | -5.90.597 | -3.10.765 | -3.60.733 | -6.50.514 | -3.30.782 | +6.80.515 | +3.70.719 | 51.3%39 |
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 counted195
Win rate63.1%
Widest swing13.8 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 |
|---|---|---|---|---|---|
| Open | 76.9% | +13.8 | 0.284 | 13 · | |
| Full fence | 73.3% | +10.2 | 0.392 | 15 · | |
| Low fence | 69.2% | +6.2 | 0.634 | 13 · | |
| Half fence | 68.8% | +5.7 | 0.624 | 16 · | |
| Octagon | 66.7% | +3.6 | 0.639 | 33 · | |
| Fence and open | 63.6% | +0.6 | 0.968 | 11 · | |
| Single wall | 58.8% | -4.3 | 0.704 | 17 · | |
| Breakable full fence | 58.6% | -4.5 | 0.590 | 29 · | |
| Breakable half fence | 57.1% | -5.9 | 0.633 | 14 · | |
| Rectangle | 55.6% | -7.5 | 0.383 | 27 · |
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 3.60 on 9 df, p 0.936
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 73.3%15 | +14.70.336 | +4.60.779 | +16.20.359 | +4.10.811 | +14.50.388 | +9.70.597 | -3.60.827 | +17.80.256 | +6.70.644 |
| Breakable full fence | -14.70.336 | 58.6%29 | -10.10.502 | +1.50.927 | -10.60.513 | -0.20.989 | -5.00.772 | -18.30.252 | +3.10.817 | -8.10.513 |
| Half fence | -4.60.779 | +10.10.502 | 68.8%16 | +11.60.510 | -0.50.978 | +9.90.554 | +5.10.782 | -8.20.624 | +13.20.392 | +2.10.884 |
| Breakable half fence | -16.20.359 | -1.50.927 | -11.60.510 | 57.1%14 | -12.10.516 | -1.70.925 | -6.50.742 | -19.80.276 | +1.60.923 | -9.50.534 |
| Low fence | -4.10.811 | +10.60.513 | +0.50.978 | +12.10.516 | 69.2%13 | +10.40.558 | +5.60.772 | -7.70.658 | +13.70.408 | +2.60.867 |
| Single wall | -14.50.388 | +0.20.989 | -9.90.554 | +1.70.925 | -10.40.558 | 58.8%17 | -4.80.799 | -18.10.297 | +3.30.831 | -7.80.584 |
| Fence and open | -9.70.597 | +5.00.772 | -5.10.782 | +6.50.742 | -5.60.772 | +4.80.799 | 63.6%11 | -13.30.476 | +8.10.647 | -3.00.854 |
| Open | +3.60.827 | +18.30.252 | +8.20.624 | +19.80.276 | +7.70.658 | +18.10.297 | +13.30.476 | 76.9%13 | +21.40.191 | +10.20.496 |
| Rectangle | -17.80.256 | -3.10.817 | -13.20.392 | -1.60.923 | -13.70.408 | -3.30.831 | -8.10.647 | -21.40.191 | 55.6%27 | -11.10.379 |
| Octagon | -6.70.644 | +8.10.513 | -2.10.884 | +9.50.534 | -2.60.867 | +7.80.584 | +3.00.854 | -10.20.496 | +11.10.379 | 66.7%33 |
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.