Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Goh vs Sarah
Matches counted2,204
Win rate49.9%
Widest swing-5.3 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Low fence | 54.1% | +4.2 | 0.219 | 194 | |
| Open | 53.1% | +3.2 | 0.331 | 209 | |
| Rectangle | 52.2% | +2.3 | 0.418 | 272 | |
| Half fence | 50.2% | +0.3 | 0.921 | 225 | |
| Breakable full fence | 50.0% | +0.1 | 0.977 | 222 | |
| Full fence | 49.6% | -0.3 | 0.912 | 228 | |
| Octagon | 48.0% | -1.9 | 0.551 | 231 | |
| Breakable half fence | 47.8% | -2.1 | 0.501 | 226 | |
| Single wall | 47.6% | -2.4 | 0.456 | 225 | |
| Fence and open | 44.6% | -5.3 | 0.322 | 83 |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 4.99 on 9 df, p 0.836
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 49.6%228 | -0.40.926 | -0.70.888 | +1.80.705 | -4.60.350 | +2.00.669 | +5.00.437 | -3.50.458 | -2.60.556 | +1.50.746 |
| Breakable full fence | +0.40.926 | 50.0%222 | -0.20.963 | +2.20.640 | -4.10.401 | +2.40.605 | +5.40.399 | -3.10.518 | -2.20.626 | +1.90.678 |
| Half fence | +0.70.888 | +0.20.963 | 50.2%225 | +2.40.605 | -3.90.425 | +2.70.572 | +5.60.379 | -2.90.547 | -2.00.660 | +2.20.643 |
| Breakable half fence | -1.80.705 | -2.20.640 | -2.40.605 | 47.8%226 | -6.30.195 | +0.20.961 | +3.20.616 | -5.30.267 | -4.40.326 | -0.30.955 |
| Low fence | +4.60.350 | +4.10.401 | +3.90.425 | +6.30.195 | 54.1%194 | +6.60.180 | +9.50.145 | +1.00.838 | +1.90.683 | +6.10.212 |
| Single wall | -2.00.669 | -2.40.605 | -2.70.572 | -0.20.961 | -6.60.180 | 47.6%225 | +3.00.642 | -5.50.248 | -4.70.302 | -0.50.916 |
| Fence and open | -5.00.437 | -5.40.399 | -5.60.379 | -3.20.616 | -9.50.145 | -3.00.642 | 44.6%83 | -8.50.188 | -7.60.224 | -3.50.587 |
| Open | +3.50.458 | +3.10.518 | +2.90.547 | +5.30.267 | -1.00.838 | +5.50.248 | +8.50.188 | 53.1%209 | +0.90.844 | +5.10.289 |
| Rectangle | +2.60.556 | +2.20.626 | +2.00.660 | +4.40.326 | -1.90.683 | +4.70.302 | +7.60.224 | -0.90.844 | 52.2%272 | +4.20.353 |
| Octagon | -1.50.746 | -1.90.678 | -2.20.643 | +0.30.955 | -6.10.212 | +0.50.916 | +3.50.587 | -5.10.289 | -4.20.353 | 48.0%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 counted2,024
Win rate48.9%
Widest swing-7.2 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Low fence | 53.6% | +4.7 | 0.186 | 179 | |
| Open | 52.4% | +3.5 | 0.317 | 189 | |
| Rectangle | 51.1% | +2.2 | 0.439 | 260 | |
| Half fence | 49.3% | +0.4 | 0.915 | 205 | |
| Full fence | 48.8% | -0.1 | 0.968 | 205 | |
| Breakable full fence | 48.8% | -0.1 | 0.965 | 203 | |
| Octagon | 47.8% | -1.2 | 0.713 | 222 | |
| Breakable half fence | 46.5% | -2.4 | 0.472 | 200 | |
| Single wall | 45.2% | -3.7 | 0.260 | 210 | |
| Fence and open | 41.7% | -7.2 | 0.210 | 72 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 6.27 on 9 df, p 0.712
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 48.8%205 | +0.00.998 | -0.50.921 | +2.30.646 | -4.80.343 | +3.50.470 | +7.10.298 | -3.60.475 | -2.40.611 | +1.00.831 |
| Breakable full fence | -0.00.998 | 48.8%203 | -0.50.920 | +2.30.648 | -4.90.343 | +3.50.472 | +7.10.300 | -3.60.475 | -2.40.610 | +1.00.833 |
| Half fence | +0.50.921 | +0.50.920 | 49.3%205 | +2.80.577 | -4.40.393 | +4.00.411 | +7.60.266 | -3.10.537 | -1.90.686 | +1.50.753 |
| Breakable half fence | -2.30.646 | -2.30.648 | -2.80.577 | 46.5%200 | -7.10.166 | +1.30.798 | +4.80.480 | -5.90.246 | -4.70.322 | -1.20.798 |
| Low fence | +4.80.343 | +4.90.343 | +4.40.393 | +7.10.166 | 53.6%179 | +8.40.099 | +12.00.086 | +1.20.810 | +2.50.610 | +5.90.241 |
| Single wall | -3.50.470 | -3.50.472 | -4.00.411 | -1.30.798 | -8.40.099 | 45.2%210 | +3.60.599 | -7.10.154 | -5.90.202 | -2.50.601 |
| Fence and open | -7.10.298 | -7.10.300 | -7.60.266 | -4.80.480 | -12.00.086 | -3.60.599 | 41.7%72 | -10.70.122 | -9.50.154 | -6.10.368 |
| Open | +3.60.475 | +3.60.475 | +3.10.537 | +5.90.246 | -1.20.810 | +7.10.154 | +10.70.122 | 52.4%189 | +1.20.797 | +4.60.349 |
| Rectangle | +2.40.611 | +2.40.610 | +1.90.686 | +4.70.322 | -2.50.610 | +5.90.202 | +9.50.154 | -1.20.797 | 51.1%260 | +3.40.456 |
| Octagon | -1.00.831 | -1.00.833 | -1.50.753 | +1.20.798 | -5.90.241 | +2.50.601 | +6.10.368 | -4.60.349 | -3.40.456 | 47.8%222 |
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 counted180
Win rate61.1%
Widest swing18.9 pts
Survive correction0
Period2026/03/26 – 2026/09/23
Where it was fought
Goh's win rate on each stage, against this matchup everywhere| Stage | Stage effect | Win % | Change | p | Matches |
|---|---|---|---|---|---|
| Single wall | 80.0% | +18.9 | 0.117 | 15 · | |
| Rectangle | 75.0% | +13.9 | 0.307 | 12 · | |
| Fence and open | 63.6% | +2.5 | 0.859 | 11 · | |
| Breakable full fence | 63.2% | +2.0 | 0.847 | 19 · | |
| Half fence | 60.0% | -1.1 | 0.914 | 20 · | |
| Low fence | 60.0% | -1.1 | 0.927 | 15 · | |
| Open | 60.0% | -1.1 | 0.914 | 20 · | |
| Breakable half fence | 57.7% | -3.4 | 0.699 | 26 · | |
| Full fence | 56.5% | -4.6 | 0.629 | 23 · | |
| Octagon | 55.6% | -5.5 | 0.726 | 9 · |
p < 0.05 on its ownStill there after correcting for how many were tested
Stage against stage
Goh's win rate on the row's stage, minus the same on the column'sAll ten shapes at once: chi-square 3.70 on 9 df, p 0.930
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 56.5%23 | -6.60.663 | -3.50.818 | -1.20.934 | -3.50.832 | -23.50.136 | -7.10.693 | -3.50.818 | -18.50.283 | +1.00.960 |
| Breakable full fence | +6.60.663 | 63.2%19 | +3.20.839 | +5.50.712 | +3.20.851 | -16.80.285 | -0.50.979 | +3.20.839 | -11.80.492 | +7.60.700 |
| Half fence | +3.50.818 | -3.20.839 | 60.0%20 | +2.30.875 | +0.01.000 | -20.00.207 | -3.60.842 | +0.01.000 | -15.00.387 | +4.40.822 |
| Breakable half fence | +1.20.934 | -5.50.712 | -2.30.875 | 57.7%26 | -2.30.885 | -22.30.147 | -6.00.736 | -2.30.875 | -17.30.304 | +2.10.911 |
| Low fence | +3.50.832 | -3.20.851 | -0.01.000 | +2.30.885 | 60.0%15 | -20.00.232 | -3.60.851 | +0.01.000 | -15.00.411 | +4.40.831 |
| Single wall | +23.50.136 | +16.80.285 | +20.00.207 | +22.30.147 | +20.00.232 | 80.0%15 | +16.40.353 | +20.00.207 | +5.00.756 | +24.40.202 |
| Fence and open | +7.10.693 | +0.50.979 | +3.60.842 | +6.00.736 | +3.60.851 | -16.40.353 | 63.6%11 | +3.60.842 | -11.40.554 | +8.10.714 |
| Open | +3.50.818 | -3.20.839 | -0.01.000 | +2.30.875 | -0.01.000 | -20.00.207 | -3.60.842 | 60.0%20 | -15.00.387 | +4.40.822 |
| Rectangle | +18.50.283 | +11.80.492 | +15.00.387 | +17.30.304 | +15.00.411 | -5.00.756 | +11.40.554 | +15.00.387 | 75.0%12 | +19.40.350 |
| Octagon | -1.00.960 | -7.60.700 | -4.40.822 | -2.10.911 | -4.40.831 | -24.40.202 | -8.10.714 | -4.40.822 | -19.40.350 | 55.6%9 |
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.