Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Jean vs Lei Fei
Matches counted1,641
Win rate47.7%
Widest swing7.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jean's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable half fence | 54.9% | +7.2 | 0.072 | 142 | |
| Single wall | 50.9% | +3.1 | 0.379 | 175 | |
| Full fence | 48.4% | +0.7 | 0.841 | 188 | |
| Half fence | 48.0% | +0.3 | 0.936 | 152 | |
| Breakable full fence | 48.0% | +0.2 | 0.947 | 171 | |
| Fence and open | 47.3% | -0.4 | 0.941 | 74 · | |
| Open | 46.8% | -1.0 | 0.790 | 169 | |
| Rectangle | 46.6% | -1.1 | 0.742 | 191 | |
| Octagon | 46.4% | -1.3 | 0.733 | 153 | |
| Low fence | 45.0% | -2.7 | 0.481 | 149 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jean's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 4.26 on 9 df, p 0.893
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.4%188 | +0.50.932 | +0.40.945 | -6.50.240 | +3.40.530 | -2.50.640 | +1.10.872 | +1.60.754 | +1.80.725 | +2.00.713 |
| Breakable full fence | -0.50.932 | 48.0%171 | -0.10.990 | -7.00.219 | +3.00.593 | -2.90.589 | +0.70.925 | +1.20.824 | +1.40.796 | +1.50.781 |
| Half fence | -0.40.945 | +0.10.990 | 48.0%152 | -6.90.237 | +3.10.595 | -2.80.610 | +0.70.918 | +1.30.819 | +1.40.792 | +1.60.777 |
| Breakable half fence | +6.50.240 | +7.00.219 | +6.90.237 | 54.9%142 | +10.00.089 | +4.10.470 | +7.60.287 | +8.20.150 | +8.30.133 | +8.50.143 |
| Low fence | -3.40.530 | -3.00.593 | -3.10.595 | -10.00.089 | 45.0%149 | -5.90.290 | -2.30.742 | -1.80.751 | -1.60.765 | -1.40.802 |
| Single wall | +2.50.640 | +2.90.589 | +2.80.610 | -4.10.470 | +5.90.290 | 50.9%175 | +3.60.608 | +4.10.446 | +4.30.415 | +4.50.421 |
| Fence and open | -1.10.872 | -0.70.925 | -0.70.918 | -7.60.287 | +2.30.742 | -3.60.608 | 47.3%74 | +0.60.937 | +0.70.918 | +0.90.900 |
| Open | -1.60.754 | -1.20.824 | -1.30.819 | -8.20.150 | +1.80.751 | -4.10.446 | -0.60.937 | 46.8%169 | +0.10.977 | +0.30.951 |
| Rectangle | -1.80.725 | -1.40.796 | -1.40.792 | -8.30.133 | +1.60.765 | -4.30.415 | -0.70.918 | -0.10.977 | 46.6%191 | +0.20.972 |
| Octagon | -2.00.713 | -1.50.781 | -1.60.777 | -8.50.143 | +1.40.802 | -4.50.421 | -0.90.900 | -0.30.951 | -0.20.972 | 46.4%153 |
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 counted1,498
Win rate48.8%
Widest swing7.9 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jean's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable half fence | 56.7% | +7.9 | 0.063 | 127 | |
| Single wall | 51.2% | +2.4 | 0.511 | 162 | |
| Breakable full fence | 50.3% | +1.5 | 0.691 | 151 | |
| Fence and open | 49.2% | +0.4 | 0.943 | 65 · | |
| Octagon | 48.9% | +0.1 | 0.973 | 141 | |
| Half fence | 48.9% | +0.1 | 0.986 | 133 | |
| Full fence | 48.1% | -0.7 | 0.834 | 181 | |
| Open | 47.8% | -1.0 | 0.790 | 159 | |
| Low fence | 47.3% | -1.5 | 0.719 | 129 | |
| Rectangle | 46.6% | -2.2 | 0.533 | 176 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jean's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 4.08 on 9 df, p 0.906
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.1%181 | -2.30.681 | -0.80.888 | -8.60.136 | +0.80.892 | -3.20.558 | -1.20.872 | +0.30.961 | +1.50.780 | -0.90.877 |
| Breakable full fence | +2.30.681 | 50.3%151 | +1.50.806 | -6.40.290 | +3.00.611 | -0.90.873 | +1.10.882 | +2.50.656 | +3.70.500 | +1.40.812 |
| Half fence | +0.80.888 | -1.50.806 | 48.9%133 | -7.80.207 | +1.60.797 | -2.40.686 | -0.40.962 | +1.10.855 | +2.30.691 | -0.10.992 |
| Breakable half fence | +8.60.136 | +6.40.290 | +7.80.207 | 56.7%127 | +9.40.132 | +5.50.356 | +7.50.326 | +8.90.135 | +10.10.083 | +7.80.204 |
| Low fence | -0.80.892 | -3.00.611 | -1.60.797 | -9.40.132 | 47.3%129 | -3.90.503 | -1.90.798 | -0.50.931 | +0.70.904 | -1.60.786 |
| Single wall | +3.20.558 | +0.90.873 | +2.40.686 | -5.50.356 | +3.90.503 | 51.2%162 | +2.00.785 | +3.40.538 | +4.60.394 | +2.30.690 |
| Fence and open | +1.20.872 | -1.10.882 | +0.40.962 | -7.50.326 | +1.90.798 | -2.00.785 | 49.2%65 | +1.40.846 | +2.60.716 | +0.30.969 |
| Open | -0.30.961 | -2.50.656 | -1.10.855 | -8.90.135 | +0.50.931 | -3.40.538 | -1.40.846 | 47.8%159 | +1.20.825 | -1.10.844 |
| Rectangle | -1.50.780 | -3.70.500 | -2.30.691 | -10.10.083 | -0.70.904 | -4.60.394 | -2.60.716 | -1.20.825 | 46.6%176 | -2.40.678 |
| Octagon | +0.90.877 | -1.40.812 | +0.10.992 | -7.80.204 | +1.60.786 | -2.30.690 | -0.30.969 | +1.10.844 | +2.40.678 | 48.9%141 |
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 counted143
Win rate36.4%
Widest swing20.8 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jean's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Full fence | 57.1% | +20.8 | 0.241 | 7 · | |
| Rectangle | 46.7% | +10.3 | 0.381 | 15 · | |
| Single wall | 46.1% | +9.8 | 0.442 | 13 · | |
| Half fence | 42.1% | +5.8 | 0.576 | 19 · | |
| Breakable half fence | 40.0% | +3.6 | 0.757 | 15 · | |
| Fence and open | 33.3% | -3.0 | 0.845 | 9 · | |
| Breakable full fence | 30.0% | -6.4 | 0.524 | 20 · | |
| Low fence | 30.0% | -6.4 | 0.524 | 20 · | |
| Open | 30.0% | -6.4 | 0.664 | 10 · | |
| Octagon | 16.7% | -19.7 | 0.138 | 12 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jean's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 5.81 on 9 df, p 0.759
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.1%7 | +27.10.201 | +15.00.495 | +17.10.452 | +27.10.201 | +11.00.639 | +23.80.341 | +27.10.263 | +10.50.647 | +40.50.067 |
| Breakable full fence | -27.10.201 | 30.0%20 | -12.10.431 | -10.00.537 | +0.01.000 | -16.10.346 | -3.30.858 | +0.01.000 | -16.70.313 | +13.30.399 |
| Half fence | -15.00.495 | +12.10.431 | 42.1%19 | +2.10.901 | +12.10.431 | -4.00.821 | +8.80.657 | +12.10.523 | -4.60.790 | +25.40.140 |
| Breakable half fence | -17.10.452 | +10.00.537 | -2.10.901 | 40.0%15 | +10.00.537 | -6.20.743 | +6.70.744 | +10.00.610 | -6.70.713 | +23.30.187 |
| Low fence | -27.10.201 | -0.01.000 | -12.10.431 | -10.00.537 | 30.0%20 | -16.10.346 | -3.30.858 | +0.01.000 | -16.70.313 | +13.30.399 |
| Single wall | -11.00.639 | +16.10.346 | +4.00.821 | +6.20.743 | +16.10.346 | 46.1%13 | +12.80.548 | +16.10.431 | -0.50.978 | +29.50.114 |
| Fence and open | -23.80.341 | +3.30.858 | -8.80.657 | -6.70.744 | +3.30.858 | -12.80.548 | 33.3%9 | +3.30.876 | -13.30.521 | +16.70.375 |
| Open | -27.10.263 | -0.01.000 | -12.10.523 | -10.00.610 | -0.01.000 | -16.10.431 | -3.30.876 | 30.0%10 | -16.70.405 | +13.30.457 |
| Rectangle | -10.50.647 | +16.70.313 | +4.60.790 | +6.70.713 | +16.70.313 | +0.50.978 | +13.30.521 | +16.70.405 | 46.7%15 | +30.00.100 |
| Octagon | -40.50.067 | -13.30.399 | -25.40.140 | -23.30.187 | -13.30.399 | -29.50.114 | -16.70.375 | -13.30.457 | -30.00.100 | 16.7%12 |
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.