Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Akira vs Sarah
Matches counted4,213
Win rate49.5%
Widest swing5.5 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 |
|---|---|---|---|---|---|
| Rectangle | 55.1% | +5.5 | 0.018 | 414 | |
| Breakable full fence | 51.0% | +1.5 | 0.511 | 443 | |
| Half fence | 50.6% | +1.1 | 0.648 | 411 | |
| Single wall | 50.4% | +0.8 | 0.724 | 405 | |
| Breakable half fence | 49.6% | +0.1 | 0.965 | 417 | |
| Full fence | 49.1% | -0.4 | 0.865 | 405 | |
| Low fence | 48.5% | -1.1 | 0.644 | 427 | |
| Fence and open | 48.3% | -1.2 | 0.721 | 211 | |
| Open | 47.6% | -2.0 | 0.401 | 412 | |
| Octagon | 46.2% | -3.3 | 0.130 | 463 |
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 8.75 on 9 df, p 0.461
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.1%405 | -1.90.584 | -1.50.674 | -0.50.885 | +0.70.849 | -1.20.725 | +0.80.851 | +1.60.655 | -5.90.089 | +2.90.391 |
| Breakable full fence | +1.90.584 | 51.0%443 | +0.40.905 | +1.40.687 | +2.50.454 | +0.70.851 | +2.70.522 | +3.50.314 | -4.00.234 | +4.80.149 |
| Half fence | +1.50.674 | -0.40.905 | 50.6%411 | +1.00.781 | +2.10.537 | +0.20.946 | +2.30.592 | +3.00.384 | -4.50.199 | +4.40.195 |
| Breakable half fence | +0.50.885 | -1.40.687 | -1.00.781 | 49.6%417 | +1.20.736 | -0.70.834 | +1.30.758 | +2.10.552 | -5.40.117 | +3.40.311 |
| Low fence | -0.70.849 | -2.50.454 | -2.10.537 | -1.20.736 | 48.5%427 | -1.90.585 | +0.10.974 | +0.90.793 | -6.60.056 | +2.30.500 |
| Single wall | +1.20.725 | -0.70.851 | -0.20.946 | +0.70.834 | +1.90.585 | 50.4%405 | +2.00.633 | +2.80.424 | -4.70.178 | +4.20.222 |
| Fence and open | -0.80.851 | -2.70.522 | -2.30.592 | -1.30.758 | -0.10.974 | -2.00.633 | 48.3%211 | +0.80.856 | -6.70.111 | +2.10.609 |
| Open | -1.60.655 | -3.50.314 | -3.00.384 | -2.10.552 | -0.90.793 | -2.80.424 | -0.80.856 | 47.6%412 | -7.50.031 | +1.40.689 |
| Rectangle | +5.90.089 | +4.00.234 | +4.50.199 | +5.40.117 | +6.60.056 | +4.70.178 | +6.70.111 | +7.50.031 | 55.1%414 | +8.80.009 |
| Octagon | -2.90.391 | -4.80.149 | -4.40.195 | -3.40.311 | -2.30.500 | -4.20.222 | -2.10.609 | -1.40.689 | -8.80.009 | 46.2%463 |
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 counted3,684
Win rate48.9%
Widest swing4.9 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 |
|---|---|---|---|---|---|
| Rectangle | 53.9% | +4.9 | 0.053 | 351 | |
| Breakable full fence | 51.1% | +2.2 | 0.355 | 399 | |
| Fence and open | 50.3% | +1.3 | 0.710 | 187 | |
| Half fence | 49.9% | +0.9 | 0.721 | 345 | |
| Single wall | 49.7% | +0.8 | 0.755 | 360 | |
| Full fence | 49.0% | +0.1 | 0.970 | 363 | |
| Breakable half fence | 48.9% | -0.0 | 0.987 | 364 | |
| Open | 47.2% | -1.8 | 0.474 | 373 | |
| Low fence | 46.6% | -2.3 | 0.350 | 373 | |
| Octagon | 45.4% | -3.6 | 0.130 | 399 |
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 7.74 on 9 df, p 0.561
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.0%363 | -2.10.564 | -0.80.827 | +0.10.971 | +2.40.517 | -0.70.854 | -1.20.784 | +1.90.615 | -4.80.199 | +3.70.310 |
| Breakable full fence | +2.10.564 | 51.1%399 | +1.30.729 | +2.20.539 | +4.50.213 | +1.40.699 | +0.90.846 | +4.00.274 | -2.70.457 | +5.80.103 |
| Half fence | +0.80.827 | -1.30.729 | 49.9%345 | +1.00.800 | +3.20.390 | +0.10.972 | -0.40.928 | +2.70.474 | -4.00.292 | +4.50.221 |
| Breakable half fence | -0.10.971 | -2.20.539 | -1.00.800 | 48.9%364 | +2.20.541 | -0.80.825 | -1.40.761 | +1.70.641 | -5.00.186 | +3.50.328 |
| Low fence | -2.40.517 | -4.50.213 | -3.20.390 | -2.20.541 | 46.6%373 | -3.10.405 | -3.60.419 | -0.50.883 | -7.20.053 | +1.30.720 |
| Single wall | +0.70.854 | -1.40.699 | -0.10.972 | +0.80.825 | +3.10.405 | 49.7%360 | -0.60.904 | +2.50.492 | -4.10.271 | +4.40.230 |
| Fence and open | +1.20.784 | -0.90.846 | +0.40.928 | +1.40.761 | +3.60.419 | +0.60.904 | 50.3%187 | +3.10.491 | -3.60.429 | +4.90.268 |
| Open | -1.90.615 | -4.00.274 | -2.70.474 | -1.70.641 | +0.50.883 | -2.50.492 | -3.10.491 | 47.2%373 | -6.70.073 | +1.80.612 |
| Rectangle | +4.80.199 | +2.70.457 | +4.00.292 | +5.00.186 | +7.20.053 | +4.10.271 | +3.60.429 | +6.70.073 | 53.9%351 | +8.50.020 |
| Octagon | -3.70.310 | -5.80.103 | -4.50.221 | -3.50.328 | -1.30.720 | -4.40.230 | -4.90.268 | -1.80.612 | -8.50.020 | 45.4%399 |
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 counted529
Win rate53.7%
Widest swing-20.4 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 |
|---|---|---|---|---|---|
| Rectangle | 61.9% | +8.2 | 0.163 | 63 · | |
| Low fence | 61.1% | +7.4 | 0.248 | 54 · | |
| Single wall | 55.6% | +1.9 | 0.793 | 45 · | |
| Breakable half fence | 54.7% | +1.0 | 0.874 | 53 · | |
| Half fence | 54.5% | +0.9 | 0.881 | 66 · | |
| Octagon | 51.6% | -2.1 | 0.716 | 64 · | |
| Open | 51.3% | -2.4 | 0.754 | 39 · | |
| Breakable full fence | 50.0% | -3.7 | 0.609 | 44 · | |
| Full fence | 50.0% | -3.7 | 0.618 | 42 · | |
| Fence and open | 33.3% | -20.4 | 0.041 | 24 · |
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 7.69 on 9 df, p 0.566
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 50.0%42 | +0.01.000 | -4.50.645 | -4.70.647 | -11.10.276 | -5.60.604 | +16.70.189 | -1.30.908 | -11.90.227 | -1.60.875 |
| Breakable full fence | -0.01.000 | 50.0%44 | -4.50.640 | -4.70.643 | -11.10.270 | -5.60.600 | +16.70.186 | -1.30.907 | -11.90.221 | -1.60.873 |
| Half fence | +4.50.645 | +4.50.640 | 54.5%66 | -0.20.985 | -6.60.469 | -1.00.916 | +21.20.075 | +3.30.746 | -7.30.397 | +3.00.733 |
| Breakable half fence | +4.70.647 | +4.70.643 | +0.20.985 | 54.7%53 | -6.40.503 | -0.80.934 | +21.40.082 | +3.40.744 | -7.20.434 | +3.20.734 |
| Low fence | +11.10.276 | +11.10.270 | +6.60.469 | +6.40.503 | 61.1%54 | +5.50.576 | +27.80.023 | +9.80.345 | -0.80.930 | +9.60.298 |
| Single wall | +5.60.604 | +5.60.600 | +1.00.916 | +0.80.934 | -5.50.576 | 55.6%45 | +22.20.078 | +4.30.695 | -6.30.508 | +4.00.681 |
| Fence and open | -16.70.189 | -16.70.186 | -21.20.075 | -21.40.082 | -27.80.023 | -22.20.078 | 33.3%24 | -17.90.164 | -28.60.017 | -18.20.127 |
| Open | +1.30.908 | +1.30.907 | -3.30.746 | -3.40.744 | -9.80.345 | -4.30.695 | +17.90.164 | 51.3%39 | -10.60.291 | -0.30.978 |
| Rectangle | +11.90.227 | +11.90.221 | +7.30.397 | +7.20.434 | +0.80.930 | +6.30.508 | +28.60.017 | +10.60.291 | 61.9%63 | +10.30.240 |
| Octagon | +1.60.875 | +1.60.873 | -3.00.733 | -3.20.734 | -9.60.298 | -4.00.681 | +18.20.127 | +0.30.978 | -10.30.240 | 51.6%64 |
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.