Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Lau vs Lei Fei
Matches counted1,958
Win rate52.3%
Widest swing-6.5 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Lau's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Rectangle | 57.6% | +5.3 | 0.130 | 184 | |
| Open | 57.1% | +4.8 | 0.183 | 177 | |
| Breakable full fence | 55.3% | +3.0 | 0.361 | 208 | |
| Octagon | 54.9% | +2.6 | 0.442 | 193 | |
| Low fence | 54.2% | +1.9 | 0.579 | 190 | |
| Fence and open | 53.3% | +1.0 | 0.836 | 107 | |
| Full fence | 50.0% | -2.3 | 0.476 | 214 | |
| Breakable half fence | 47.7% | -4.6 | 0.203 | 174 | |
| Half fence | 46.9% | -5.4 | 0.113 | 192 | |
| Single wall | 45.8% | -6.5 | 0.051 | 203 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Lau's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 12.89 on 9 df, p 0.168
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%214 | -5.30.277 | +3.10.529 | +2.30.652 | -4.20.398 | +4.20.392 | -3.30.580 | -7.10.164 | -7.60.129 | -4.90.321 |
| Breakable full fence | +5.30.277 | 55.3%208 | +8.40.093 | +7.60.139 | +1.10.829 | +9.50.055 | +2.00.733 | -1.80.727 | -2.30.644 | +0.40.941 |
| Half fence | -3.10.529 | -8.40.093 | 46.9%192 | -0.80.874 | -7.30.152 | +1.10.832 | -6.40.289 | -10.20.050 | -10.70.037 | -8.00.114 |
| Breakable half fence | -2.30.652 | -7.60.139 | +0.80.874 | 47.7%174 | -6.50.215 | +1.90.714 | -5.60.365 | -9.40.079 | -9.90.061 | -7.20.167 |
| Low fence | +4.20.398 | -1.10.829 | +7.30.152 | +6.50.215 | 54.2%190 | +8.40.096 | +0.90.876 | -2.90.583 | -3.40.508 | -0.70.889 |
| Single wall | -4.20.392 | -9.50.055 | -1.10.832 | -1.90.714 | -8.40.096 | 45.8%203 | -7.50.212 | -11.20.029 | -11.80.020 | -9.10.070 |
| Fence and open | +3.30.580 | -2.00.733 | +6.40.289 | +5.60.365 | -0.90.876 | +7.50.212 | 53.3%107 | -3.80.533 | -4.30.472 | -1.60.783 |
| Open | +7.10.164 | +1.80.727 | +10.20.050 | +9.40.079 | +2.90.583 | +11.20.029 | +3.80.533 | 57.1%177 | -0.60.916 | +2.10.679 |
| Rectangle | +7.60.129 | +2.30.644 | +10.70.037 | +9.90.061 | +3.40.508 | +11.80.020 | +4.30.472 | +0.60.916 | 57.6%184 | +2.70.599 |
| Octagon | +4.90.321 | -0.40.941 | +8.00.114 | +7.20.167 | +0.70.889 | +9.10.070 | +1.60.783 | -2.10.679 | -2.70.599 | 54.9%193 |
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 counted1,485
Win rate52.7%
Widest swing-7.0 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Lau's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Rectangle | 59.7% | +7.0 | 0.105 | 124 | |
| Breakable full fence | 58.0% | +5.3 | 0.152 | 162 | |
| Fence and open | 57.1% | +4.4 | 0.404 | 84 | |
| Open | 56.3% | +3.6 | 0.384 | 135 | |
| Low fence | 54.0% | +1.2 | 0.760 | 139 | |
| Full fence | 52.8% | +0.0 | 0.993 | 163 | |
| Octagon | 52.4% | -0.3 | 0.929 | 147 | |
| Breakable half fence | 47.0% | -5.8 | 0.165 | 132 | |
| Half fence | 46.0% | -6.7 | 0.082 | 150 | |
| Single wall | 45.7% | -7.0 | 0.057 | 162 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Lau's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.36 on 9 df, p 0.147
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.8%163 | -5.30.340 | +6.80.232 | +5.80.323 | -1.20.835 | +7.10.202 | -4.40.513 | -3.50.542 | -6.90.243 | +0.40.947 |
| Breakable full fence | +5.30.340 | 58.0%162 | +12.00.034 | +11.10.059 | +4.10.478 | +12.30.026 | +0.90.894 | +1.70.764 | -1.70.778 | +5.60.319 |
| Half fence | -6.80.232 | -12.00.034 | 46.0%150 | -1.00.871 | -8.00.176 | +0.30.955 | -11.10.102 | -10.30.083 | -13.70.024 | -6.40.271 |
| Breakable half fence | -5.80.323 | -11.10.059 | +1.00.871 | 47.0%132 | -7.00.250 | +1.30.825 | -10.20.145 | -9.30.127 | -12.70.042 | -5.40.367 |
| Low fence | +1.20.835 | -4.10.478 | +8.00.176 | +7.00.250 | 54.0%139 | +8.30.152 | -3.20.643 | -2.30.697 | -5.70.350 | +1.60.790 |
| Single wall | -7.10.202 | -12.30.026 | -0.30.955 | -1.30.825 | -8.30.152 | 45.7%162 | -11.50.088 | -10.60.068 | -14.00.019 | -6.70.239 |
| Fence and open | +4.40.513 | -0.90.894 | +11.10.102 | +10.20.145 | +3.20.643 | +11.50.088 | 57.1%84 | +0.80.902 | -2.50.716 | +4.80.485 |
| Open | +3.50.542 | -1.70.764 | +10.30.083 | +9.30.127 | +2.30.697 | +10.60.068 | -0.80.902 | 56.3%135 | -3.40.582 | +3.90.510 |
| Rectangle | +6.90.243 | +1.70.778 | +13.70.024 | +12.70.042 | +5.70.350 | +14.00.019 | +2.50.716 | +3.40.582 | 59.7%124 | +7.30.228 |
| Octagon | -0.40.947 | -5.60.319 | +6.40.271 | +5.40.367 | -1.60.790 | +6.70.239 | -4.80.485 | -3.90.510 | -7.30.228 | 52.4%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 counted473
Win rate51.0%
Widest swing12.1 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Lau's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Octagon | 63.0% | +12.1 | 0.084 | 46 · | |
| Open | 59.5% | +8.6 | 0.244 | 42 · | |
| Low fence | 54.9% | +4.0 | 0.550 | 51 · | |
| Rectangle | 53.3% | +2.4 | 0.693 | 60 · | |
| Breakable half fence | 50.0% | -0.9 | 0.897 | 42 · | |
| Half fence | 50.0% | -0.9 | 0.897 | 42 · | |
| Single wall | 46.3% | -4.6 | 0.537 | 41 · | |
| Breakable full fence | 45.6% | -5.3 | 0.449 | 46 · | |
| Full fence | 41.2% | -9.8 | 0.139 | 51 · | |
| Fence and open | 39.1% | -11.8 | 0.245 | 23 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Lau's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.51 on 9 df, p 0.483
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 41.2%51 | -4.50.657 | -8.80.395 | -8.80.395 | -13.70.165 | -5.20.619 | +2.00.868 | -18.30.078 | -12.20.201 | -21.90.031 |
| Breakable full fence | +4.50.657 | 45.6%46 | -4.30.683 | -4.30.683 | -9.20.363 | -0.70.949 | +6.50.606 | -13.90.193 | -7.70.433 | -17.40.094 |
| Half fence | +8.80.395 | +4.30.683 | 50.0%42 | +0.01.000 | -4.90.638 | +3.70.739 | +10.90.401 | -9.50.381 | -3.30.740 | -13.00.217 |
| Breakable half fence | +8.80.395 | +4.30.683 | -0.01.000 | 50.0%42 | -4.90.638 | +3.70.739 | +10.90.401 | -9.50.381 | -3.30.740 | -13.00.217 |
| Low fence | +13.70.165 | +9.20.363 | +4.90.638 | +4.90.638 | 54.9%51 | +8.60.414 | +15.80.209 | -4.60.654 | +1.60.869 | -8.10.416 |
| Single wall | +5.20.619 | +0.70.949 | -3.70.739 | -3.70.739 | -8.60.414 | 46.3%41 | +7.20.577 | -13.20.229 | -7.00.490 | -16.70.118 |
| Fence and open | -2.00.868 | -6.50.606 | -10.90.401 | -10.90.401 | -15.80.209 | -7.20.577 | 39.1%23 | -20.40.115 | -14.20.247 | -23.90.060 |
| Open | +18.30.078 | +13.90.193 | +9.50.381 | +9.50.381 | +4.60.654 | +13.20.229 | +20.40.115 | 59.5%42 | +6.20.535 | -3.50.735 |
| Rectangle | +12.20.201 | +7.70.433 | +3.30.740 | +3.30.740 | -1.60.869 | +7.00.490 | +14.20.247 | -6.20.535 | 53.3%60 | -9.70.316 |
| Octagon | +21.90.031 | +17.40.094 | +13.00.217 | +13.00.217 | +8.10.416 | +16.70.118 | +23.90.060 | +3.50.735 | +9.70.316 | 63.0%46 |
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.