Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Sarah vs Taka
Matches counted640
Win rate45.8%
Widest swing-10.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Fence and open | 54.2% | +8.4 | 0.401 | 24 · | |
| Full fence | 52.9% | +7.2 | 0.210 | 68 · | |
| Open | 51.4% | +5.7 | 0.315 | 70 · | |
| Octagon | 48.2% | +2.4 | 0.702 | 56 · | |
| Single wall | 44.6% | -1.2 | 0.842 | 65 · | |
| Rectangle | 44.3% | -1.5 | 0.790 | 70 · | |
| Breakable half fence | 43.1% | -2.7 | 0.644 | 65 · | |
| Half fence | 42.5% | -3.3 | 0.529 | 80 | |
| Breakable full fence | 40.5% | -5.2 | 0.336 | 74 · | |
| Low fence | 35.6% | -10.2 | 0.153 | 45 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 6.44 on 9 df, p 0.695
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 52.9%68 | +12.40.139 | +10.40.205 | +9.90.255 | +17.40.069 | +8.30.337 | -1.20.918 | +1.50.859 | +8.70.309 | +4.70.600 |
| Breakable full fence | -12.40.139 | 40.5%74 | -2.00.805 | -2.50.762 | +5.00.588 | -4.10.628 | -13.60.242 | -10.90.190 | -3.80.649 | -7.70.383 |
| Half fence | -10.40.205 | +2.00.805 | 42.5%80 | -0.60.944 | +6.90.447 | -2.10.798 | -11.70.314 | -8.90.274 | -1.80.826 | -5.70.510 |
| Breakable half fence | -9.90.255 | +2.50.762 | +0.60.944 | 43.1%65 | +7.50.429 | -1.50.860 | -11.10.352 | -8.30.332 | -1.20.887 | -5.10.571 |
| Low fence | -17.40.069 | -5.00.588 | -6.90.447 | -7.50.429 | 35.6%45 | -9.10.342 | -18.60.136 | -15.90.095 | -8.70.353 | -12.70.201 |
| Single wall | -8.30.337 | +4.10.628 | +2.10.798 | +1.50.860 | +9.10.342 | 44.6%65 | -9.60.423 | -6.80.429 | +0.30.969 | -3.60.692 |
| Fence and open | +1.20.918 | +13.60.242 | +11.70.314 | +11.10.352 | +18.60.136 | +9.60.423 | 54.2%24 | +2.70.817 | +9.90.403 | +6.00.626 |
| Open | -1.50.859 | +10.90.190 | +8.90.274 | +8.30.332 | +15.90.095 | +6.80.429 | -2.70.817 | 51.4%70 | +7.10.398 | +3.20.720 |
| Rectangle | -8.70.309 | +3.80.649 | +1.80.826 | +1.20.887 | +8.70.353 | -0.30.969 | -9.90.403 | -7.10.398 | 44.3%70 | -3.90.660 |
| Octagon | -4.70.600 | +7.70.383 | +5.70.510 | +5.10.571 | +12.70.201 | +3.60.692 | -6.00.626 | -3.20.720 | +3.90.660 | 48.2%56 |
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 counted548
Win rate47.1%
Widest swing12.9 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Open | 60.0% | +12.9 | 0.043 | 55 · | |
| Full fence | 57.9% | +10.8 | 0.084 | 57 · | |
| Fence and open | 57.1% | +10.1 | 0.346 | 21 · | |
| Octagon | 49.1% | +2.0 | 0.762 | 53 · | |
| Half fence | 45.7% | -1.4 | 0.806 | 70 · | |
| Single wall | 45.5% | -1.6 | 0.799 | 55 · | |
| Breakable half fence | 42.9% | -4.2 | 0.504 | 56 · | |
| Rectangle | 42.4% | -4.7 | 0.443 | 59 · | |
| Breakable full fence | 40.3% | -6.8 | 0.235 | 67 · | |
| Low fence | 34.2% | -12.9 | 0.099 | 38 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 12.10 on 9 df, p 0.208
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 57.9%57 | +17.60.051 | +12.20.172 | +15.00.110 | +23.70.024 | +12.40.188 | +0.80.952 | -2.10.821 | +15.50.095 | +8.80.353 |
| Breakable full fence | -17.60.051 | 40.3%67 | -5.40.522 | -2.60.774 | +6.10.537 | -5.20.567 | -16.80.175 | -19.70.030 | -2.10.813 | -8.80.337 |
| Half fence | -12.20.172 | +5.40.522 | 45.7%70 | +2.90.748 | +11.50.247 | +0.30.977 | -11.40.358 | -14.30.113 | +3.30.703 | -3.40.713 |
| Breakable half fence | -15.00.110 | +2.60.774 | -2.90.748 | 42.9%56 | +8.70.400 | -2.60.783 | -14.30.263 | -17.10.071 | +0.50.958 | -6.20.516 |
| Low fence | -23.70.024 | -6.10.537 | -11.50.247 | -8.70.400 | 34.2%38 | -11.20.278 | -22.90.088 | -25.80.014 | -8.20.421 | -14.80.158 |
| Single wall | -12.40.188 | +5.20.567 | -0.30.977 | +2.60.783 | +11.20.278 | 45.5%55 | -11.70.362 | -14.60.127 | +3.10.740 | -3.60.708 |
| Fence and open | -0.80.952 | +16.80.175 | +11.40.358 | +14.30.263 | +22.90.088 | +11.70.362 | 57.1%21 | -2.90.821 | +14.80.244 | +8.10.530 |
| Open | +2.10.821 | +19.70.030 | +14.30.113 | +17.10.071 | +25.80.014 | +14.60.127 | +2.90.821 | 60.0%55 | +17.60.060 | +10.90.253 |
| Rectangle | -15.50.095 | +2.10.813 | -3.30.703 | -0.50.958 | +8.20.421 | -3.10.740 | -14.80.244 | -17.60.060 | 42.4%59 | -6.70.478 |
| Octagon | -8.80.353 | +8.80.337 | +3.40.713 | +6.20.516 | +14.80.158 | +3.60.708 | -8.10.530 | -10.90.253 | +6.70.478 | 49.1%53 |
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 counted92
Win rate38.0%
Widest swing-18.0 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Sarah's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Rectangle | 54.5% | +16.5 | 0.230 | 11 · | |
| Breakable half fence | 44.4% | +6.4 | 0.677 | 9 · | |
| Breakable full fence | 42.9% | +4.8 | 0.785 | 7 · | |
| Low fence | 42.9% | +4.8 | 0.785 | 7 · | |
| Single wall | 40.0% | +2.0 | 0.893 | 10 · | |
| Fence and open | 33.3% | -4.7 | 0.864 | 3 · | |
| Octagon | 33.3% | -4.7 | 0.864 | 3 · | |
| Full fence | 27.3% | -10.8 | 0.433 | 11 · | |
| Half fence | 20.0% | -18.0 | 0.213 | 10 · | |
| Open | 20.0% | -18.0 | 0.116 | 15 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Sarah's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 5.47 on 9 df, p 0.792
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 27.3%11 | -15.60.494 | +7.30.696 | -17.20.423 | -15.60.494 | -12.70.537 | -6.10.837 | +7.30.664 | -27.30.193 | -6.10.837 |
| Breakable full fence | +15.60.494 | 42.9%7 | +22.90.309 | -1.60.949 | +0.01.000 | +2.90.906 | +9.50.778 | +22.90.262 | -11.70.629 | +9.50.778 |
| Half fence | -7.30.696 | -22.90.309 | 20.0%10 | -24.40.252 | -22.90.309 | -20.00.329 | -13.30.631 | +0.01.000 | -34.50.104 | -13.30.631 |
| Breakable half fence | +17.20.423 | +1.60.949 | +24.40.252 | 44.4%9 | +1.60.949 | +4.40.845 | +11.10.735 | +24.40.202 | -10.10.653 | +11.10.735 |
| Low fence | +15.60.494 | -0.01.000 | +22.90.309 | -1.60.949 | 42.9%7 | +2.90.906 | +9.50.778 | +22.90.262 | -11.70.629 | +9.50.778 |
| Single wall | +12.70.537 | -2.90.906 | +20.00.329 | -4.40.845 | -2.90.906 | 40.0%10 | +6.70.835 | +20.00.275 | -14.60.505 | +6.70.835 |
| Fence and open | +6.10.837 | -9.50.778 | +13.30.631 | -11.10.735 | -9.50.778 | -6.70.835 | 33.3%3 | +13.30.612 | -21.20.515 | +0.01.000 |
| Open | -7.30.664 | -22.90.262 | -0.01.000 | -24.40.202 | -22.90.262 | -20.00.275 | -13.30.612 | 20.0%15 | -34.50.067 | -13.30.612 |
| Rectangle | +27.30.193 | +11.70.629 | +34.50.104 | +10.10.653 | +11.70.629 | +14.60.505 | +21.20.515 | +34.50.067 | 54.5%11 | +21.20.515 |
| Octagon | +6.10.837 | -9.50.778 | +13.30.631 | -11.10.735 | -9.50.778 | -6.70.835 | -0.01.000 | +13.30.612 | -21.20.515 | 33.3%3 |
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.