Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Goh vs Pai
Matches counted1,921
Win rate48.6%
Widest swing-5.9 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 | 52.9% | +4.3 | 0.371 | 104 | |
| Low fence | 52.5% | +3.8 | 0.275 | 183 | |
| Single wall | 51.8% | +3.1 | 0.350 | 199 | |
| Half fence | 51.1% | +2.5 | 0.484 | 176 | |
| Open | 48.8% | +0.1 | 0.964 | 203 | |
| Full fence | 47.9% | -0.8 | 0.829 | 188 | |
| Breakable half fence | 47.7% | -0.9 | 0.789 | 197 | |
| Rectangle | 47.3% | -1.3 | 0.707 | 186 | |
| Breakable full fence | 47.2% | -1.4 | 0.680 | 199 | |
| Octagon | 42.7% | -5.9 | 0.084 | 192 |
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 6.12 on 9 df, p 0.727
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 47.9%188 | +0.60.900 | -3.30.534 | +0.10.975 | -4.60.377 | -3.90.445 | -5.00.412 | -0.90.859 | +0.60.914 | +5.20.312 |
| Breakable full fence | -0.60.900 | 47.2%199 | -3.90.451 | -0.50.924 | -5.20.308 | -4.50.367 | -5.60.350 | -1.50.758 | -0.10.988 | +4.50.368 |
| Half fence | +3.30.534 | +3.90.451 | 51.1%176 | +3.40.509 | -1.30.802 | -0.60.904 | -1.70.777 | +2.40.646 | +3.80.467 | +8.40.106 |
| Breakable half fence | -0.10.975 | +0.50.924 | -3.40.509 | 47.7%197 | -4.70.355 | -4.00.421 | -5.20.394 | -1.10.833 | +0.40.937 | +5.00.321 |
| Low fence | +4.60.377 | +5.20.308 | +1.30.802 | +4.70.355 | 52.5%183 | +0.70.891 | -0.40.945 | +3.70.469 | +5.20.323 | +9.80.059 |
| Single wall | +3.90.445 | +4.50.367 | +0.60.904 | +4.00.421 | -0.70.891 | 51.8%199 | -1.10.852 | +3.00.549 | +4.50.383 | +9.10.073 |
| Fence and open | +5.00.412 | +5.60.350 | +1.70.777 | +5.20.394 | +0.40.945 | +1.10.852 | 52.9%104 | +4.10.495 | +5.60.363 | +10.20.094 |
| Open | +0.90.859 | +1.50.758 | -2.40.646 | +1.10.833 | -3.70.469 | -3.00.549 | -4.10.495 | 48.8%203 | +1.50.774 | +6.10.227 |
| Rectangle | -0.60.914 | +0.10.988 | -3.80.467 | -0.40.937 | -5.20.323 | -4.50.383 | -5.60.363 | -1.50.774 | 47.3%186 | +4.60.368 |
| Octagon | -5.20.312 | -4.50.368 | -8.40.106 | -5.00.321 | -9.80.059 | -9.10.073 | -10.20.094 | -6.10.227 | -4.60.368 | 42.7%192 |
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 counted1,826
Win rate48.6%
Widest swing-5.2 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 | 53.1% | +4.5 | 0.361 | 98 | |
| Single wall | 52.0% | +3.4 | 0.304 | 198 | |
| Low fence | 51.8% | +3.2 | 0.382 | 170 | |
| Half fence | 49.7% | +1.1 | 0.760 | 167 | |
| Full fence | 49.1% | +0.5 | 0.881 | 171 | |
| Open | 48.5% | -0.1 | 0.975 | 196 | |
| Breakable full fence | 48.4% | -0.2 | 0.953 | 184 | |
| Breakable half fence | 47.4% | -1.2 | 0.734 | 194 | |
| Rectangle | 45.5% | -3.1 | 0.383 | 176 | |
| Octagon | 43.3% | -5.2 | 0.138 | 180 |
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 5.29 on 9 df, p 0.808
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.1%171 | +0.80.887 | -0.60.915 | +1.70.746 | -2.60.626 | -2.90.579 | -3.90.534 | +0.70.901 | +3.70.494 | +5.80.277 |
| Breakable full fence | -0.80.887 | 48.4%184 | -1.30.803 | +0.90.854 | -3.40.523 | -3.60.476 | -4.70.453 | -0.10.984 | +2.90.580 | +5.00.335 |
| Half fence | +0.60.915 | +1.30.803 | 49.7%167 | +2.30.666 | -2.10.705 | -2.30.659 | -3.40.597 | +1.20.815 | +4.20.431 | +6.40.235 |
| Breakable half fence | -1.70.746 | -0.90.854 | -2.30.666 | 47.4%194 | -4.30.408 | -4.60.363 | -5.60.363 | -1.10.836 | +2.00.705 | +4.10.427 |
| Low fence | +2.60.626 | +3.40.523 | +2.10.705 | +4.30.408 | 51.8%170 | -0.30.961 | -1.30.838 | +3.30.529 | +6.30.240 | +8.40.114 |
| Single wall | +2.90.579 | +3.60.476 | +2.30.659 | +4.60.363 | +0.30.961 | 52.0%198 | -1.00.866 | +3.50.481 | +6.60.205 | +8.70.091 |
| Fence and open | +3.90.534 | +4.70.453 | +3.40.597 | +5.60.363 | +1.30.838 | +1.00.866 | 53.1%98 | +4.60.458 | +7.60.227 | +9.70.120 |
| Open | -0.70.901 | +0.10.984 | -1.20.815 | +1.10.836 | -3.30.529 | -3.50.481 | -4.60.458 | 48.5%196 | +3.00.561 | +5.10.318 |
| Rectangle | -3.70.494 | -2.90.580 | -4.20.431 | -2.00.705 | -6.30.240 | -6.60.205 | -7.60.227 | -3.00.561 | 45.5%176 | +2.10.687 |
| Octagon | -5.80.277 | -5.00.335 | -6.40.235 | -4.10.427 | -8.40.114 | -8.70.091 | -9.70.120 | -5.10.318 | -2.10.687 | 43.3%180 |
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 counted95
Win rate49.5%
Widest swing-49.5 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 | 80.0% | +30.5 | 0.041 | 10 · | |
| Half fence | 77.8% | +28.3 | 0.074 | 9 · | |
| Breakable half fence | 66.7% | +17.2 | 0.545 | 3 · | |
| Low fence | 61.5% | +12.1 | 0.349 | 13 · | |
| Open | 57.1% | +7.7 | 0.673 | 7 · | |
| Fence and open | 50.0% | +0.5 | 0.979 | 6 · | |
| Full fence | 35.3% | -14.2 | 0.197 | 17 · | |
| Breakable full fence | 33.3% | -16.1 | 0.173 | 15 · | |
| Octagon | 33.3% | -16.1 | 0.232 | 12 · | |
| Single wall | 0.0% | -49.5 | 0.320 | 1 · |
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 13.01 on 9 df, p 0.162
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 35.3%17 | +2.00.907 | -42.50.039 | -31.40.306 | -26.20.153 | +35.30.467 | -14.70.526 | -21.90.324 | -44.70.025 | +2.00.913 |
| Breakable full fence | -2.00.907 | 33.3%15 | -44.50.035 | -33.30.280 | -28.20.136 | +33.30.486 | -16.70.477 | -23.80.290 | -46.70.022 | +0.01.000 |
| Half fence | +42.50.039 | +44.50.035 | 77.8%9 | +11.10.700 | +16.20.421 | +77.80.107 | +27.80.264 | +20.60.377 | -2.20.906 | +44.50.044 |
| Breakable half fence | +31.40.306 | +33.30.280 | -11.10.700 | 66.7%3 | +5.10.869 | +66.70.248 | +16.70.635 | +9.50.778 | -13.30.631 | +33.30.292 |
| Low fence | +26.20.153 | +28.20.136 | -16.20.421 | -5.10.869 | 61.5%13 | +61.50.231 | +11.50.636 | +4.40.848 | -18.50.340 | +28.20.158 |
| Single wall | -35.30.467 | -33.30.486 | -77.80.107 | -66.70.248 | -61.50.231 | 0.0%1 | -50.00.350 | -57.10.285 | -80.00.087 | -33.30.488 |
| Fence and open | +14.70.526 | +16.70.477 | -27.80.264 | -16.70.635 | -11.50.636 | +50.00.350 | 50.0%6 | -7.10.797 | -30.00.210 | +16.70.494 |
| Open | +21.90.324 | +23.80.290 | -20.60.377 | -9.50.778 | -4.40.848 | +57.10.285 | +7.10.797 | 57.1%7 | -22.90.309 | +23.80.311 |
| Rectangle | +44.70.025 | +46.70.022 | +2.20.906 | +13.30.631 | +18.50.340 | +80.00.087 | +30.00.210 | +22.90.309 | 80.0%10 | +46.70.029 |
| Octagon | -2.00.913 | -0.01.000 | -44.50.044 | -33.30.292 | -28.20.158 | +33.30.488 | -16.70.494 | -23.80.311 | -46.70.029 | 33.3%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.