Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Jeffry vs Lei Fei
Matches counted1,750
Win rate55.1%
Widest swing8.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jeffry's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable full fence | 63.3% | +8.2 | 0.031 | 158 | |
| Fence and open | 62.6% | +7.5 | 0.140 | 91 | |
| Octagon | 58.6% | +3.5 | 0.347 | 162 | |
| Rectangle | 57.6% | +2.5 | 0.477 | 184 | |
| Open | 57.1% | +2.0 | 0.567 | 182 | |
| Breakable half fence | 55.2% | +0.0 | 0.993 | 203 | |
| Half fence | 54.8% | -0.3 | 0.936 | 155 | |
| Low fence | 51.2% | -3.9 | 0.233 | 203 | |
| Full fence | 50.3% | -4.8 | 0.203 | 157 | |
| Single wall | 48.3% | -6.8 | 0.054 | 176 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jeffry's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.88 on 9 df, p 0.127
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.3%157 | -13.00.020 | -4.50.424 | -4.80.360 | -0.90.864 | +2.00.712 | -12.30.060 | -6.80.209 | -7.30.178 | -8.30.136 |
| Breakable full fence | +13.00.020 | 63.3%158 | +8.40.128 | +8.10.120 | +12.10.022 | +15.00.006 | +0.70.918 | +6.20.248 | +5.70.284 | +4.70.394 |
| Half fence | +4.50.424 | -8.40.128 | 54.8%155 | -0.30.950 | +3.60.498 | +6.50.235 | -7.80.232 | -2.30.671 | -2.80.608 | -3.80.494 |
| Breakable half fence | +4.80.360 | -8.10.120 | +0.30.950 | 55.2%203 | +3.90.426 | +6.90.181 | -7.50.231 | -2.00.697 | -2.40.629 | -3.50.506 |
| Low fence | +0.90.864 | -12.10.022 | -3.60.498 | -3.90.426 | 51.2%203 | +2.90.569 | -11.40.069 | -5.90.245 | -6.40.209 | -7.40.158 |
| Single wall | -2.00.712 | -15.00.006 | -6.50.235 | -6.90.181 | -2.90.569 | 48.3%176 | -14.30.026 | -8.80.094 | -9.30.077 | -10.30.057 |
| Fence and open | +12.30.060 | -0.70.918 | +7.80.232 | +7.50.231 | +11.40.069 | +14.30.026 | 62.6%91 | +5.50.384 | +5.00.425 | +4.00.533 |
| Open | +6.80.209 | -6.20.248 | +2.30.671 | +2.00.697 | +5.90.245 | +8.80.094 | -5.50.384 | 57.1%182 | -0.50.928 | -1.50.779 |
| Rectangle | +7.30.178 | -5.70.284 | +2.80.608 | +2.40.629 | +6.40.209 | +9.30.077 | -5.00.425 | +0.50.928 | 57.6%184 | -1.00.846 |
| Octagon | +8.30.136 | -4.70.394 | +3.80.494 | +3.50.506 | +7.40.158 | +10.30.057 | -4.00.533 | +1.50.779 | +1.00.846 | 58.6%162 |
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 counted1,575
Win rate55.8%
Widest swing8.7 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jeffry's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable full fence | 64.5% | +8.7 | 0.023 | 152 | |
| Fence and open | 63.5% | +7.8 | 0.137 | 85 | |
| Open | 59.1% | +3.4 | 0.366 | 159 | |
| Octagon | 58.5% | +2.8 | 0.480 | 147 | |
| Breakable half fence | 57.6% | +1.9 | 0.593 | 177 | |
| Rectangle | 57.1% | +1.4 | 0.700 | 168 | |
| Half fence | 55.2% | -0.5 | 0.899 | 134 | |
| Full fence | 50.7% | -5.0 | 0.205 | 142 | |
| Single wall | 50.6% | -5.1 | 0.169 | 160 | |
| Low fence | 50.0% | -5.8 | 0.101 | 178 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jeffry's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.75 on 9 df, p 0.132
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.7%142 | -13.80.017 | -4.50.452 | -6.90.217 | +0.70.900 | +0.10.989 | -12.80.060 | -8.40.143 | -6.40.257 | -7.80.183 |
| Breakable full fence | +13.80.017 | 64.5%152 | +9.20.111 | +6.80.205 | +14.50.008 | +13.80.013 | +0.90.884 | +5.30.331 | +7.30.180 | +6.00.289 |
| Half fence | +4.50.452 | -9.20.111 | 55.2%134 | -2.40.672 | +5.20.360 | +4.60.432 | -8.30.224 | -3.90.502 | -1.90.738 | -3.30.579 |
| Breakable half fence | +6.90.217 | -6.80.205 | +2.40.672 | 57.6%177 | +7.60.150 | +7.00.198 | -5.90.362 | -1.50.782 | +0.50.928 | -0.90.874 |
| Low fence | -0.70.900 | -14.50.008 | -5.20.360 | -7.60.150 | 50.0%178 | -0.60.909 | -13.50.039 | -9.10.093 | -7.10.183 | -8.50.126 |
| Single wall | -0.10.989 | -13.80.013 | -4.60.432 | -7.00.198 | +0.60.909 | 50.6%160 | -12.90.053 | -8.50.127 | -6.50.236 | -7.90.166 |
| Fence and open | +12.80.060 | -0.90.884 | +8.30.224 | +5.90.362 | +13.50.039 | +12.90.053 | 63.5%85 | +4.40.502 | +6.40.329 | +5.00.451 |
| Open | +8.40.143 | -5.30.331 | +3.90.502 | +1.50.782 | +9.10.093 | +8.50.127 | -4.40.502 | 59.1%159 | +2.00.717 | +0.60.913 |
| Rectangle | +6.40.257 | -7.30.180 | +1.90.738 | -0.50.928 | +7.10.183 | +6.50.236 | -6.40.329 | -2.00.717 | 57.1%168 | -1.40.807 |
| Octagon | +7.80.183 | -6.00.289 | +3.30.579 | +0.90.874 | +8.50.126 | +7.90.166 | -5.00.451 | -0.60.913 | +1.40.807 | 58.5%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 counted175
Win rate49.7%
Widest swing-24.7 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Jeffry's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Rectangle | 62.5% | +12.8 | 0.283 | 16 · | |
| Low fence | 60.0% | +10.3 | 0.267 | 25 · | |
| Octagon | 60.0% | +10.3 | 0.405 | 15 · | |
| Half fence | 52.4% | +2.7 | 0.794 | 21 · | |
| Fence and open | 50.0% | +0.3 | 0.989 | 6 · | |
| Full fence | 46.7% | -3.0 | 0.805 | 15 · | |
| Open | 43.5% | -6.2 | 0.521 | 23 · | |
| Breakable half fence | 38.5% | -11.2 | 0.214 | 26 · | |
| Breakable full fence | 33.3% | -16.4 | 0.414 | 6 · | |
| Single wall | 25.0% | -24.7 | 0.038 | 16 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Jeffry's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.88 on 9 df, p 0.448
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 46.7%15 | +13.30.577 | -5.70.735 | +8.20.607 | -13.30.412 | +21.70.208 | -3.30.890 | +3.20.847 | -15.80.376 | -13.30.464 |
| Breakable full fence | -13.30.577 | 33.3%6 | -19.10.410 | -5.10.815 | -26.70.239 | +8.30.696 | -16.70.558 | -10.20.653 | -29.20.221 | -26.70.269 |
| Half fence | +5.70.735 | +19.10.410 | 52.4%21 | +13.90.340 | -7.60.604 | +27.40.093 | +2.40.918 | +8.90.555 | -10.10.538 | -7.60.650 |
| Breakable half fence | -8.20.607 | +5.10.815 | -13.90.340 | 38.5%26 | -21.50.124 | +13.50.369 | -11.50.604 | -5.00.721 | -24.00.130 | -21.50.183 |
| Low fence | +13.30.412 | +26.70.239 | +7.60.604 | +21.50.124 | 60.0%25 | +35.00.028 | +10.00.656 | +16.50.252 | -2.50.873 | +0.01.000 |
| Single wall | -21.70.208 | -8.30.696 | -27.40.093 | -13.50.369 | -35.00.028 | 25.0%16 | -25.00.262 | -18.50.237 | -37.50.033 | -35.00.048 |
| Fence and open | +3.30.890 | +16.70.558 | -2.40.918 | +11.50.604 | -10.00.656 | +25.00.262 | 50.0%6 | +6.50.775 | -12.50.595 | -10.00.676 |
| Open | -3.20.847 | +10.20.653 | -8.90.555 | +5.00.721 | -16.50.252 | +18.50.237 | -6.50.775 | 43.5%23 | -19.00.242 | -16.50.319 |
| Rectangle | +15.80.376 | +29.20.221 | +10.10.538 | +24.00.130 | +2.50.873 | +37.50.033 | +12.50.595 | +19.00.242 | 62.5%16 | +2.50.886 |
| Octagon | +13.30.464 | +26.70.269 | +7.60.650 | +21.50.183 | -0.01.000 | +35.00.048 | +10.00.676 | +16.50.319 | -2.50.886 | 60.0%15 |
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.