Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Akira vs Lei Fei
Matches counted3,473
Win rate47.3%
Widest swing3.1 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Akira's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Single wall | 50.4% | +3.1 | 0.223 | 347 | |
| Open | 50.3% | +3.0 | 0.246 | 346 | |
| Full fence | 50.1% | +2.8 | 0.275 | 339 | |
| Breakable full fence | 48.3% | +1.0 | 0.717 | 323 | |
| Half fence | 46.6% | -0.7 | 0.765 | 418 | |
| Breakable half fence | 46.3% | -1.0 | 0.716 | 300 | |
| Octagon | 46.1% | -1.2 | 0.623 | 360 | |
| Rectangle | 45.2% | -2.1 | 0.400 | 356 | |
| Fence and open | 45.0% | -2.3 | 0.502 | 202 | |
| Low fence | 44.5% | -2.8 | 0.305 | 301 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Akira's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 6.16 on 9 df, p 0.723
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.1%339 | +1.90.634 | +3.50.338 | +3.80.336 | +5.60.155 | -0.30.941 | +5.10.251 | -0.10.970 | +4.90.194 | +4.00.286 |
| Breakable full fence | -1.90.634 | 48.3%323 | +1.60.656 | +2.00.624 | +3.80.344 | -2.10.581 | +3.20.468 | -2.00.607 | +3.10.423 | +2.20.568 |
| Half fence | -3.50.338 | -1.60.656 | 46.6%418 | +0.30.933 | +2.10.571 | -3.80.297 | +1.60.708 | -3.60.316 | +1.40.692 | +0.50.880 |
| Breakable half fence | -3.80.336 | -2.00.624 | -0.30.933 | 46.3%300 | +1.80.655 | -4.10.298 | +1.30.777 | -4.00.316 | +1.10.776 | +0.20.955 |
| Low fence | -5.60.155 | -3.80.344 | -2.10.571 | -1.80.655 | 44.5%301 | -5.90.133 | -0.50.906 | -5.80.143 | -0.70.856 | -1.60.682 |
| Single wall | +0.30.941 | +2.10.581 | +3.80.297 | +4.10.298 | +5.90.133 | 50.4%347 | +5.40.224 | +0.10.970 | +5.20.167 | +4.30.250 |
| Fence and open | -5.10.251 | -3.20.468 | -1.60.708 | -1.30.777 | +0.50.906 | -5.40.224 | 45.0%202 | -5.20.236 | -0.20.968 | -1.10.808 |
| Open | +0.10.970 | +2.00.607 | +3.60.316 | +4.00.316 | +5.80.143 | -0.10.970 | +5.20.236 | 50.3%346 | +5.10.179 | +4.20.267 |
| Rectangle | -4.90.194 | -3.10.423 | -1.40.692 | -1.10.776 | +0.70.856 | -5.20.167 | +0.20.968 | -5.10.179 | 45.2%356 | -0.90.812 |
| Octagon | -4.00.286 | -2.20.568 | -0.50.880 | -0.20.955 | +1.60.682 | -4.30.250 | +1.10.808 | -4.20.267 | +0.90.812 | 46.1%360 |
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 counted2,952
Win rate47.2%
Widest swing-7.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Akira's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Full fence | 51.6% | +4.4 | 0.118 | 287 | |
| Open | 50.3% | +3.1 | 0.250 | 300 | |
| Single wall | 50.0% | +2.8 | 0.309 | 294 | |
| Breakable full fence | 48.9% | +1.7 | 0.547 | 276 | |
| Octagon | 47.4% | +0.2 | 0.942 | 306 | |
| Half fence | 45.7% | -1.5 | 0.547 | 361 | |
| Breakable half fence | 45.5% | -1.7 | 0.554 | 264 | |
| Rectangle | 45.2% | -2.0 | 0.469 | 290 | |
| Low fence | 43.9% | -3.3 | 0.258 | 260 | |
| Fence and open | 40.0% | -7.2 | 0.061 | 160 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Akira's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 10.26 on 9 df, p 0.330
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 51.6%287 | +2.70.529 | +5.90.138 | +6.10.151 | +7.70.071 | +1.60.705 | +11.60.019 | +1.20.765 | +6.40.124 | +4.20.309 |
| Breakable full fence | -2.70.529 | 48.9%276 | +3.20.422 | +3.50.421 | +5.10.240 | -1.10.795 | +8.90.072 | -1.40.733 | +3.70.373 | +1.50.713 |
| Half fence | -5.90.138 | -3.20.422 | 45.7%361 | +0.30.950 | +1.90.646 | -4.30.274 | +5.70.226 | -4.60.236 | +0.50.892 | -1.70.665 |
| Breakable half fence | -6.10.151 | -3.50.421 | -0.30.950 | 45.5%264 | +1.60.711 | -4.50.283 | +5.50.272 | -4.90.247 | +0.30.947 | -1.90.645 |
| Low fence | -7.70.071 | -5.10.240 | -1.90.646 | -1.60.711 | 43.9%260 | -6.20.148 | +3.90.439 | -6.50.125 | -1.30.755 | -3.50.400 |
| Single wall | -1.60.705 | +1.10.795 | +4.30.274 | +4.50.283 | +6.20.148 | 50.0%294 | +10.00.041 | -0.30.935 | +4.80.243 | +2.60.522 |
| Fence and open | -11.60.019 | -8.90.072 | -5.70.226 | -5.50.272 | -3.90.439 | -10.00.041 | 40.0%160 | -10.30.034 | -5.20.289 | -7.40.128 |
| Open | -1.20.765 | +1.40.733 | +4.60.236 | +4.90.247 | +6.50.125 | +0.30.935 | +10.30.034 | 50.3%300 | +5.20.210 | +2.90.468 |
| Rectangle | -6.40.124 | -3.70.373 | -0.50.892 | -0.30.947 | +1.30.755 | -4.80.243 | +5.20.289 | -5.20.210 | 45.2%290 | -2.20.588 |
| Octagon | -4.20.309 | -1.50.713 | +1.70.665 | +1.90.645 | +3.50.400 | -2.60.522 | +7.40.128 | -2.90.468 | +2.20.588 | 47.4%306 |
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 counted521
Win rate48.2%
Widest swing16.1 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Akira's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Fence and open | 64.3% | +16.1 | 0.029 | 42 · | |
| Single wall | 52.8% | +4.7 | 0.474 | 53 · | |
| Breakable half fence | 52.8% | +4.6 | 0.567 | 36 · | |
| Half fence | 52.6% | +4.5 | 0.476 | 57 · | |
| Open | 50.0% | +1.8 | 0.795 | 46 · | |
| Low fence | 48.8% | +0.6 | 0.936 | 41 · | |
| Rectangle | 45.5% | -2.7 | 0.636 | 66 · | |
| Breakable full fence | 44.7% | -3.5 | 0.615 | 47 · | |
| Full fence | 42.3% | -5.9 | 0.372 | 52 · | |
| Octagon | 38.9% | -9.3 | 0.149 | 54 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Akira's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 8.58 on 9 df, p 0.477
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 42.3%52 | -2.40.812 | -10.30.281 | -10.50.333 | -6.50.533 | -10.50.280 | -22.00.034 | -7.70.446 | -3.10.732 | +3.40.720 |
| Breakable full fence | +2.40.812 | 44.7%47 | -8.00.420 | -8.10.464 | -4.10.701 | -8.20.416 | -19.60.064 | -5.30.607 | -0.80.935 | +5.80.556 |
| Half fence | +10.30.281 | +8.00.420 | 52.6%57 | -0.10.989 | +3.90.707 | -0.20.983 | -11.70.246 | +2.60.791 | +7.20.427 | +13.70.146 |
| Breakable half fence | +10.50.333 | +8.10.464 | +0.10.989 | 52.8%36 | +4.00.726 | -0.10.996 | -11.50.303 | +2.80.803 | +7.30.479 | +13.90.194 |
| Low fence | +6.50.533 | +4.10.701 | -3.90.707 | -4.00.726 | 48.8%41 | -4.00.697 | -15.50.154 | -1.20.910 | +3.30.737 | +9.90.335 |
| Single wall | +10.50.280 | +8.20.416 | +0.20.983 | +0.10.996 | +4.00.697 | 52.8%53 | -11.50.261 | +2.80.779 | +7.40.424 | +13.90.148 |
| Fence and open | +22.00.034 | +19.60.064 | +11.70.246 | +11.50.303 | +15.50.154 | +11.50.261 | 64.3%42 | +14.30.177 | +18.80.056 | +25.40.014 |
| Open | +7.70.446 | +5.30.607 | -2.60.791 | -2.80.803 | +1.20.910 | -2.80.779 | -14.30.177 | 50.0%46 | +4.50.636 | +11.10.265 |
| Rectangle | +3.10.732 | +0.80.935 | -7.20.427 | -7.30.479 | -3.30.737 | -7.40.424 | -18.80.056 | -4.50.636 | 45.5%66 | +6.60.469 |
| Octagon | -3.40.720 | -5.80.556 | -13.70.146 | -13.90.194 | -9.90.335 | -13.90.148 | -25.40.014 | -11.10.265 | -6.60.469 | 38.9%54 |
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.