Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Kage vs Lei Fei
Matches counted1,990
Win rate51.8%
Widest swing8.1 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Kage's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable full fence | 59.9% | +8.1 | 0.032 | 162 | |
| Low fence | 55.6% | +3.8 | 0.279 | 189 | |
| Half fence | 55.0% | +3.2 | 0.352 | 189 | |
| Full fence | 54.7% | +2.9 | 0.362 | 225 | |
| Fence and open | 52.7% | +0.9 | 0.843 | 110 | |
| Octagon | 51.3% | -0.5 | 0.873 | 197 | |
| Breakable half fence | 49.5% | -2.3 | 0.476 | 220 | |
| Open | 49.0% | -2.8 | 0.397 | 210 | |
| Single wall | 46.8% | -5.0 | 0.150 | 190 | |
| Rectangle | 46.1% | -5.7 | 0.096 | 193 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Kage's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 12.34 on 9 df, p 0.195
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 54.7%225 | -5.20.307 | -0.40.942 | +5.10.280 | -0.90.856 | +7.80.112 | +1.90.738 | +5.60.241 | +8.60.081 | +3.40.485 |
| Breakable full fence | +5.20.307 | 59.9%162 | +4.80.360 | +10.30.045 | +4.30.414 | +13.00.015 | +7.20.242 | +10.80.038 | +13.80.010 | +8.60.103 |
| Half fence | +0.40.942 | -4.80.360 | 55.0%189 | +5.50.269 | -0.50.918 | +8.20.111 | +2.30.700 | +6.00.233 | +8.90.082 | +3.80.460 |
| Breakable half fence | -5.10.280 | -10.30.045 | -5.50.269 | 49.5%220 | -6.00.225 | +2.70.585 | -3.20.586 | +0.50.918 | +3.40.486 | -1.70.725 |
| Low fence | +0.90.856 | -4.30.414 | +0.50.918 | +6.00.225 | 55.6%189 | +8.70.090 | +2.80.636 | +6.50.194 | +9.40.065 | +4.30.399 |
| Single wall | -7.80.112 | -13.00.015 | -8.20.111 | -2.70.585 | -8.70.090 | 46.8%190 | -5.90.326 | -2.20.659 | +0.70.886 | -4.40.384 |
| Fence and open | -1.90.738 | -7.20.242 | -2.30.700 | +3.20.586 | -2.80.636 | +5.90.326 | 52.7%110 | +3.70.532 | +6.60.268 | +1.50.806 |
| Open | -5.60.241 | -10.80.038 | -6.00.233 | -0.50.918 | -6.50.194 | +2.20.659 | -3.70.532 | 49.0%210 | +2.90.556 | -2.20.654 |
| Rectangle | -8.60.081 | -13.80.010 | -8.90.082 | -3.40.486 | -9.40.065 | -0.70.886 | -6.60.268 | -2.90.556 | 46.1%193 | -5.20.309 |
| Octagon | -3.40.485 | -8.60.103 | -3.80.460 | +1.70.725 | -4.30.399 | +4.40.384 | -1.50.806 | +2.20.654 | +5.20.309 | 51.3%197 |
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 counted1,566
Win rate53.1%
Widest swing8.0 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Kage's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Breakable full fence | 61.1% | +8.0 | 0.059 | 126 | |
| Full fence | 57.1% | +4.1 | 0.231 | 189 | |
| Low fence | 56.6% | +3.5 | 0.361 | 152 | |
| Half fence | 54.7% | +1.6 | 0.690 | 139 | |
| Octagon | 53.5% | +0.4 | 0.908 | 157 | |
| Fence and open | 52.4% | -0.6 | 0.907 | 82 | |
| Single wall | 52.2% | -0.9 | 0.826 | 138 | |
| Open | 49.3% | -3.7 | 0.333 | 152 | |
| Breakable half fence | 49.2% | -3.9 | 0.264 | 181 | |
| Rectangle | 47.8% | -5.3 | 0.161 | 157 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Kage's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 9.19 on 9 df, p 0.420
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.1%189 | -4.00.483 | +2.50.656 | +8.00.124 | +0.60.917 | +5.00.372 | +4.70.474 | +7.80.151 | +9.40.082 | +3.60.498 |
| Breakable full fence | +4.00.483 | 61.1%126 | +6.40.290 | +11.90.039 | +4.50.445 | +8.90.144 | +8.70.216 | +11.80.050 | +13.30.025 | +7.60.199 |
| Half fence | -2.50.656 | -6.40.290 | 54.7%139 | +5.50.329 | -1.90.744 | +2.50.676 | +2.20.747 | +5.30.363 | +6.90.236 | +1.20.840 |
| Breakable half fence | -8.00.124 | -11.90.039 | -5.50.329 | 49.2%181 | -7.40.178 | -3.00.595 | -3.30.623 | -0.20.975 | +1.40.797 | -4.30.427 |
| Low fence | -0.60.917 | -4.50.445 | +1.90.744 | +7.40.178 | 56.6%152 | +4.40.452 | +4.10.544 | +7.20.206 | +8.80.121 | +3.10.587 |
| Single wall | -5.00.372 | -8.90.144 | -2.50.676 | +3.00.595 | -4.40.452 | 52.2%138 | -0.30.970 | +2.80.630 | +4.40.450 | -1.30.819 |
| Fence and open | -4.70.474 | -8.70.216 | -2.20.747 | +3.30.623 | -4.10.544 | +0.30.970 | 52.4%82 | +3.10.651 | +4.70.493 | -1.10.876 |
| Open | -7.80.151 | -11.80.050 | -5.30.363 | +0.20.975 | -7.20.206 | -2.80.630 | -3.10.651 | 49.3%152 | +1.60.782 | -4.20.464 |
| Rectangle | -9.40.082 | -13.30.025 | -6.90.236 | -1.40.797 | -8.80.121 | -4.40.450 | -4.70.493 | -1.60.782 | 47.8%157 | -5.70.310 |
| Octagon | -3.60.498 | -7.60.199 | -1.20.840 | +4.30.427 | -3.10.587 | +1.30.819 | +1.10.876 | +4.20.464 | +5.70.310 | 53.5%157 |
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 counted424
Win rate47.2%
Widest swing-14.5 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Kage's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Half fence | 56.0% | +8.8 | 0.183 | 50 · | |
| Breakable full fence | 55.6% | +8.4 | 0.292 | 36 · | |
| Fence and open | 53.6% | +6.4 | 0.483 | 28 · | |
| Low fence | 51.4% | +4.2 | 0.594 | 37 · | |
| Breakable half fence | 51.3% | +4.1 | 0.589 | 39 · | |
| Open | 48.3% | +1.1 | 0.856 | 58 · | |
| Octagon | 42.5% | -4.7 | 0.534 | 40 · | |
| Full fence | 41.7% | -5.5 | 0.489 | 36 · | |
| Rectangle | 38.9% | -8.3 | 0.298 | 36 · | |
| Single wall | 32.7% | -14.5 | 0.026 | 52 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Kage's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 9.74 on 9 df, p 0.372
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 41.7%36 | -13.90.238 | -14.30.190 | -9.60.404 | -9.70.407 | +9.00.390 | -11.90.344 | -6.60.532 | +2.80.810 | -0.80.941 |
| Breakable full fence | +13.90.238 | 55.6%36 | -0.40.967 | +4.30.711 | +4.20.719 | +22.90.033 | +2.00.874 | +7.30.492 | +16.70.157 | +13.10.256 |
| Half fence | +14.30.190 | +0.40.967 | 56.0%50 | +4.70.658 | +4.70.667 | +23.30.018 | +2.40.836 | +7.70.423 | +17.10.117 | +13.50.203 |
| Breakable half fence | +9.60.404 | -4.30.711 | -4.70.658 | 51.3%39 | -0.10.995 | +18.60.074 | -2.30.853 | +3.00.772 | +12.40.281 | +8.80.434 |
| Low fence | +9.70.407 | -4.20.719 | -4.70.667 | +0.10.995 | 51.4%37 | +18.70.077 | -2.20.859 | +3.10.770 | +12.50.285 | +8.80.437 |
| Single wall | -9.00.390 | -22.90.033 | -23.30.018 | -18.60.074 | -18.70.077 | 32.7%52 | -20.90.069 | -15.60.097 | -6.20.550 | -9.80.334 |
| Fence and open | +11.90.344 | -2.00.874 | -2.40.836 | +2.30.853 | +2.20.859 | +20.90.069 | 53.6%28 | +5.30.645 | +14.70.242 | +11.10.368 |
| Open | +6.60.532 | -7.30.492 | -7.70.423 | -3.00.772 | -3.10.770 | +15.60.097 | -5.30.645 | 48.3%58 | +9.40.374 | +5.80.573 |
| Rectangle | -2.80.810 | -16.70.157 | -17.10.117 | -12.40.281 | -12.50.285 | +6.20.550 | -14.70.242 | -9.40.374 | 38.9%36 | -3.60.749 |
| Octagon | +0.80.941 | -13.10.256 | -13.50.203 | -8.80.434 | -8.80.437 | +9.80.334 | -11.10.368 | -5.80.573 | +3.60.749 | 42.5%40 |
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.