Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Goh vs Lei Fei
Matches counted1,675
Win rate47.2%
Widest swing5.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Fence and open | 52.3% | +5.2 | 0.397 | 65 · | |
| Rectangle | 51.7% | +4.5 | 0.248 | 149 | |
| Full fence | 51.1% | +4.0 | 0.266 | 174 | |
| Open | 50.5% | +3.4 | 0.324 | 190 | |
| Breakable half fence | 48.2% | +1.1 | 0.792 | 141 | |
| Half fence | 46.6% | -0.6 | 0.872 | 176 | |
| Octagon | 45.6% | -1.6 | 0.687 | 147 | |
| Low fence | 45.2% | -2.0 | 0.579 | 177 | |
| Single wall | 43.6% | -3.5 | 0.339 | 165 | |
| Breakable full fence | 42.8% | -4.4 | 0.194 | 194 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 6.67 on 9 df, p 0.672
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 51.1%174 | +8.40.108 | +4.60.394 | +2.90.606 | +6.00.265 | +7.50.166 | -1.20.873 | +0.60.905 | -0.50.925 | +5.60.320 |
| Breakable full fence | -8.40.108 | 42.8%194 | -3.80.462 | -5.50.323 | -2.40.640 | -0.90.871 | -9.50.182 | -7.80.128 | -8.90.102 | -2.80.607 |
| Half fence | -4.60.394 | +3.80.462 | 46.6%176 | -1.60.772 | +1.40.793 | +3.00.584 | -5.70.431 | -3.90.452 | -5.10.361 | +1.00.856 |
| Breakable half fence | -2.90.606 | +5.50.323 | +1.60.772 | 48.2%141 | +3.00.591 | +4.60.422 | -4.10.586 | -2.30.679 | -3.50.557 | +2.60.653 |
| Low fence | -6.00.265 | +2.40.640 | -1.40.793 | -3.00.591 | 45.2%177 | +1.60.772 | -7.10.326 | -5.30.307 | -6.50.243 | -0.40.945 |
| Single wall | -7.50.166 | +0.90.871 | -3.00.584 | -4.60.422 | -1.60.772 | 43.6%165 | -8.70.235 | -6.90.195 | -8.00.154 | -1.90.730 |
| Fence and open | +1.20.873 | +9.50.182 | +5.70.431 | +4.10.586 | +7.10.326 | +8.70.235 | 52.3%65 | +1.80.804 | +0.60.932 | +6.70.366 |
| Open | -0.60.905 | +7.80.128 | +3.90.452 | +2.30.679 | +5.30.307 | +6.90.195 | -1.80.804 | 50.5%190 | -1.10.833 | +5.00.367 |
| Rectangle | +0.50.925 | +8.90.102 | +5.10.361 | +3.50.557 | +6.50.243 | +8.00.154 | -0.60.932 | +1.10.833 | 51.7%149 | +6.10.294 |
| Octagon | -5.60.320 | +2.80.607 | -1.00.856 | -2.60.653 | +0.40.945 | +1.90.730 | -6.70.366 | -5.00.367 | -6.10.294 | 45.6%147 |
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 counted1,550
Win rate48.1%
Widest swing6.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Fence and open | 54.2% | +6.2 | 0.333 | 59 · | |
| Full fence | 52.1% | +4.1 | 0.270 | 165 | |
| Open | 51.7% | +3.7 | 0.306 | 172 | |
| Rectangle | 50.7% | +2.7 | 0.512 | 138 | |
| Half fence | 48.5% | +0.4 | 0.918 | 161 | |
| Breakable half fence | 47.7% | -0.4 | 0.923 | 128 | |
| Low fence | 47.0% | -1.1 | 0.763 | 164 | |
| Octagon | 46.5% | -1.6 | 0.712 | 129 | |
| Breakable full fence | 44.6% | -3.4 | 0.332 | 177 | |
| Single wall | 43.8% | -4.2 | 0.254 | 162 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 5.50 on 9 df, p 0.789
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.1%165 | +7.50.166 | +3.70.507 | +4.50.448 | +5.20.348 | +8.30.133 | -2.10.780 | +0.40.945 | +1.40.809 | +5.60.340 |
| Breakable full fence | -7.50.166 | 44.6%177 | -3.80.483 | -3.00.601 | -2.30.668 | +0.80.881 | -9.60.201 | -7.10.184 | -6.10.283 | -1.90.744 |
| Half fence | -3.70.507 | +3.80.483 | 48.5%161 | +0.80.894 | +1.50.787 | +4.60.405 | -5.80.447 | -3.30.548 | -2.30.695 | +1.90.743 |
| Breakable half fence | -4.50.448 | +3.00.601 | -0.80.894 | 47.7%128 | +0.70.905 | +3.80.516 | -6.60.403 | -4.10.484 | -3.10.617 | +1.10.854 |
| Low fence | -5.20.348 | +2.30.668 | -1.50.787 | -0.70.905 | 47.0%164 | +3.10.571 | -7.30.337 | -4.80.380 | -3.80.513 | +0.40.940 |
| Single wall | -8.30.133 | -0.80.881 | -4.60.405 | -3.80.516 | -3.10.571 | 43.8%162 | -10.40.170 | -7.90.148 | -6.90.233 | -2.70.647 |
| Fence and open | +2.10.780 | +9.60.201 | +5.80.447 | +6.60.403 | +7.30.337 | +10.40.170 | 54.2%59 | +2.50.741 | +3.50.651 | +7.70.325 |
| Open | -0.40.945 | +7.10.184 | +3.30.548 | +4.10.484 | +4.80.380 | +7.90.148 | -2.50.741 | 51.7%172 | +1.00.858 | +5.20.369 |
| Rectangle | -1.40.809 | +6.10.283 | +2.30.695 | +3.10.617 | +3.80.513 | +6.90.233 | -3.50.651 | -1.00.858 | 50.7%138 | +4.20.491 |
| Octagon | -5.60.340 | +1.90.744 | -1.90.743 | -1.10.854 | -0.40.940 | +2.70.647 | -7.70.325 | -5.20.369 | -4.20.491 | 46.5%129 |
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 counted125
Win rate36.0%
Widest swing27.6 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Rectangle | 63.6% | +27.6 | 0.046 | 11 · | |
| Breakable half fence | 53.9% | +17.9 | 0.157 | 13 · | |
| Octagon | 38.9% | +2.9 | 0.783 | 18 · | |
| Open | 38.9% | +2.9 | 0.783 | 18 · | |
| Full fence | 33.3% | -2.7 | 0.863 | 9 · | |
| Fence and open | 33.3% | -2.7 | 0.889 | 6 · | |
| Single wall | 33.3% | -2.7 | 0.922 | 3 · | |
| Half fence | 26.7% | -9.3 | 0.422 | 15 · | |
| Breakable full fence | 23.5% | -12.5 | 0.249 | 17 · | |
| Low fence | 23.1% | -12.9 | 0.305 | 13 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.21 on 9 df, p 0.513
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 33.3%9 | +9.80.592 | +6.70.728 | -20.50.342 | +10.20.595 | +0.01.000 | +0.01.000 | -5.60.778 | -30.30.178 | -5.60.778 |
| Breakable full fence | -9.80.592 | 23.5%17 | -3.10.838 | -30.30.088 | +0.50.977 | -9.80.718 | -9.80.638 | -15.40.328 | -40.10.034 | -15.40.328 |
| Half fence | -6.70.728 | +3.10.838 | 26.7%15 | -27.20.142 | +3.60.827 | -6.70.814 | -6.70.760 | -12.20.458 | -37.00.059 | -12.20.458 |
| Breakable half fence | +20.50.342 | +30.30.088 | +27.20.142 | 53.9%13 | +30.80.107 | +20.50.522 | +20.50.405 | +15.00.409 | -9.80.628 | +15.00.409 |
| Low fence | -10.20.595 | -0.50.977 | -3.60.827 | -30.80.107 | 23.1%13 | -10.20.712 | -10.20.637 | -15.80.353 | -40.60.045 | -15.80.353 |
| Single wall | -0.01.000 | +9.80.718 | +6.70.814 | -20.50.522 | +10.20.712 | 33.3%3 | +0.01.000 | -5.60.854 | -30.30.347 | -5.60.854 |
| Fence and open | -0.01.000 | +9.80.638 | +6.70.760 | -20.50.405 | +10.20.637 | -0.01.000 | 33.3%6 | -5.60.808 | -30.30.232 | -5.60.808 |
| Open | +5.60.778 | +15.40.328 | +12.20.458 | -15.00.409 | +15.80.353 | +5.60.854 | +5.60.808 | 38.9%18 | -24.80.196 | +0.01.000 |
| Rectangle | +30.30.178 | +40.10.034 | +37.00.059 | +9.80.628 | +40.60.045 | +30.30.347 | +30.30.232 | +24.80.196 | 63.6%11 | +24.80.196 |
| Octagon | +5.60.778 | +15.40.328 | +12.20.458 | -15.00.409 | +15.80.353 | +5.60.854 | +5.60.808 | -0.01.000 | -24.80.196 | 38.9%18 |
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.