Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Brad vs Lion
Matches counted2,235
Win rate50.2%
Widest swing6.5 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Octagon | 56.7% | +6.5 | 0.035 | 231 | |
| Single wall | 54.9% | +4.7 | 0.209 | 164 | |
| Half fence | 53.1% | +3.0 | 0.349 | 224 | |
| Open | 52.8% | +2.6 | 0.398 | 235 | |
| Low fence | 50.7% | +0.6 | 0.866 | 211 | |
| Breakable full fence | 49.6% | -0.6 | 0.849 | 226 | |
| Fence and open | 48.3% | -1.9 | 0.679 | 118 | |
| Full fence | 47.4% | -2.7 | 0.401 | 213 | |
| Rectangle | 44.9% | -5.2 | 0.105 | 216 | |
| Breakable half fence | 44.4% | -5.8 | 0.041 | 277 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.77 on 9 df, p 0.131
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 47.4%213 | -2.10.654 | -5.70.233 | +3.00.507 | -3.30.498 | -7.50.151 | -0.90.877 | -5.30.258 | +2.50.602 | -9.30.050 |
| Breakable full fence | +2.10.654 | 49.6%226 | -3.60.449 | +5.20.249 | -1.10.810 | -5.30.299 | +1.20.825 | -3.20.491 | +4.70.328 | -7.20.126 |
| Half fence | +5.70.233 | +3.60.449 | 53.1%224 | +8.70.052 | +2.40.615 | -1.80.732 | +4.80.397 | +0.30.939 | +8.20.085 | -3.60.442 |
| Breakable half fence | -3.00.507 | -5.20.249 | -8.70.052 | 44.4%277 | -6.30.167 | -10.50.033 | -3.90.476 | -8.40.059 | -0.50.911 | -12.30.006 |
| Low fence | +3.30.498 | +1.10.810 | -2.40.615 | +6.30.167 | 50.7%211 | -4.20.423 | +2.40.676 | -2.10.665 | +5.80.230 | -6.00.206 |
| Single wall | +7.50.151 | +5.30.299 | +1.80.732 | +10.50.033 | +4.20.423 | 54.9%164 | +6.60.276 | +2.10.677 | +10.00.054 | -1.80.718 |
| Fence and open | +0.90.877 | -1.20.825 | -4.80.397 | +3.90.476 | -2.40.676 | -6.60.276 | 48.3%118 | -4.50.429 | +3.40.552 | -8.40.136 |
| Open | +5.30.258 | +3.20.491 | -0.30.939 | +8.40.059 | +2.10.665 | -2.10.677 | +4.50.429 | 52.8%235 | +7.90.095 | -3.90.392 |
| Rectangle | -2.50.602 | -4.70.328 | -8.20.085 | +0.50.911 | -5.80.230 | -10.00.054 | -3.40.552 | -7.90.095 | 44.9%216 | -11.80.013 |
| Octagon | +9.30.050 | +7.20.126 | +3.60.442 | +12.30.006 | +6.00.206 | +1.80.718 | +8.40.136 | +3.90.392 | +11.80.013 | 56.7%231 |
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 counted1,700
Win rate49.8%
Widest swing9.8 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Octagon | 59.6% | +9.8 | 0.008 | 166 | |
| Single wall | 53.1% | +3.3 | 0.437 | 128 | |
| Low fence | 52.7% | +2.8 | 0.437 | 169 | |
| Open | 52.0% | +2.2 | 0.545 | 177 | |
| Half fence | 51.7% | +1.9 | 0.600 | 178 | |
| Breakable full fence | 47.3% | -2.5 | 0.493 | 167 | |
| Full fence | 46.9% | -2.9 | 0.433 | 160 | |
| Rectangle | 45.6% | -4.2 | 0.264 | 160 | |
| Breakable half fence | 44.9% | -4.9 | 0.122 | 216 | |
| Fence and open | 44.6% | -5.2 | 0.327 | 83 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 13.18 on 9 df, p 0.155
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 46.9%160 | -0.40.938 | -4.80.377 | +2.00.705 | -5.80.294 | -6.20.292 | +2.30.733 | -5.10.350 | +1.30.823 | -12.80.021 |
| Breakable full fence | +0.40.938 | 47.3%167 | -4.40.416 | +2.40.641 | -5.30.326 | -5.80.322 | +2.70.684 | -4.70.386 | +1.70.761 | -12.30.024 |
| Half fence | +4.80.377 | +4.40.416 | 51.7%178 | +6.80.180 | -1.00.855 | -1.40.804 | +7.10.285 | -0.30.956 | +6.10.266 | -8.00.138 |
| Breakable half fence | -2.00.705 | -2.40.641 | -6.80.180 | 44.9%216 | -7.80.131 | -8.20.140 | +0.30.959 | -7.10.163 | -0.70.890 | -14.70.004 |
| Low fence | +5.80.294 | +5.30.326 | +1.00.855 | +7.80.131 | 52.7%169 | -0.50.937 | +8.10.228 | +0.70.898 | +7.00.202 | -7.00.198 |
| Single wall | +6.20.292 | +5.80.322 | +1.40.804 | +8.20.140 | +0.50.937 | 53.1%128 | +8.50.225 | +1.10.843 | +7.50.206 | -6.50.264 |
| Fence and open | -2.30.733 | -2.70.684 | -7.10.285 | -0.30.959 | -8.10.228 | -8.50.225 | 44.6%83 | -7.40.266 | -1.00.876 | -15.10.024 |
| Open | +5.10.350 | +4.70.386 | +0.30.956 | +7.10.163 | -0.70.898 | -1.10.843 | +7.40.266 | 52.0%177 | +6.40.244 | -7.70.153 |
| Rectangle | -1.30.823 | -1.70.761 | -6.10.266 | +0.70.890 | -7.00.202 | -7.50.206 | +1.00.876 | -6.40.244 | 45.6%160 | -14.00.011 |
| Octagon | +12.80.021 | +12.30.024 | +8.00.138 | +14.70.004 | +7.00.198 | +6.50.264 | +15.10.024 | +7.70.153 | +14.00.011 | 59.6%166 |
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 counted535
Win rate51.2%
Widest swing9.9 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Brad's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Single wall | 61.1% | +9.9 | 0.219 | 36 · | |
| Half fence | 58.7% | +7.5 | 0.288 | 46 · | |
| Fence and open | 57.1% | +5.9 | 0.468 | 35 · | |
| Breakable full fence | 55.9% | +4.7 | 0.442 | 59 · | |
| Open | 55.2% | +4.0 | 0.523 | 58 · | |
| Octagon | 49.2% | -2.0 | 0.733 | 65 · | |
| Full fence | 49.1% | -2.1 | 0.741 | 53 · | |
| Low fence | 42.9% | -8.3 | 0.259 | 42 · | |
| Rectangle | 42.9% | -8.3 | 0.186 | 56 · | |
| Breakable half fence | 42.6% | -8.6 | 0.154 | 61 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Brad's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.54 on 9 df, p 0.481
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.1%53 | -6.90.467 | -9.60.338 | +6.40.492 | +6.20.547 | -12.10.263 | -8.10.457 | -6.10.519 | +6.20.516 | -0.20.985 |
| Breakable full fence | +6.90.467 | 55.9%59 | -2.80.776 | +13.30.145 | +13.10.195 | -5.20.620 | -1.20.909 | +0.80.934 | +13.10.161 | +6.70.456 |
| Half fence | +9.60.338 | +2.80.776 | 58.7%46 | +16.10.100 | +15.80.138 | -2.40.825 | +1.60.888 | +3.50.719 | +15.80.111 | +9.50.325 |
| Breakable half fence | -6.40.492 | -13.30.145 | -16.10.100 | 42.6%61 | -0.20.981 | -18.50.079 | -14.50.170 | -12.60.171 | -0.20.980 | -6.60.457 |
| Low fence | -6.20.547 | -13.10.195 | -15.80.138 | +0.20.981 | 42.9%42 | -18.20.108 | -14.30.212 | -12.30.224 | +0.01.000 | -6.40.519 |
| Single wall | +12.10.263 | +5.20.620 | +2.40.825 | +18.50.079 | +18.20.108 | 61.1%36 | +4.00.734 | +5.90.571 | +18.20.087 | +11.90.252 |
| Fence and open | +8.10.457 | +1.20.909 | -1.60.888 | +14.50.170 | +14.30.212 | -4.00.734 | 57.1%35 | +2.00.853 | +14.30.185 | +7.90.450 |
| Open | +6.10.519 | -0.80.934 | -3.50.719 | +12.60.171 | +12.30.224 | -5.90.571 | -2.00.853 | 55.2%58 | +12.30.189 | +5.90.510 |
| Rectangle | -6.20.516 | -13.10.161 | -15.80.111 | +0.20.980 | -0.01.000 | -18.20.087 | -14.30.185 | -12.30.189 | 42.9%56 | -6.40.483 |
| Octagon | +0.20.985 | -6.70.456 | -9.50.325 | +6.60.457 | +6.40.519 | -11.90.252 | -7.90.450 | -5.90.510 | +6.40.483 | 49.2%65 |
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.