Virtua Fighter 5 REVO · v1.10 · Character Matchup Winrates
Goh vs Taka
Matches counted716
Win rate44.4%
Widest swing-9.6 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 |
|---|---|---|---|---|---|
| Open | 50.9% | +6.5 | 0.306 | 57 · | |
| Rectangle | 48.5% | +4.1 | 0.473 | 68 · | |
| Low fence | 46.1% | +1.7 | 0.767 | 65 · | |
| Half fence | 45.5% | +1.0 | 0.822 | 99 | |
| Fence and open | 45.0% | +0.6 | 0.939 | 40 · | |
| Octagon | 44.9% | +0.5 | 0.928 | 69 · | |
| Breakable full fence | 43.0% | -1.4 | 0.782 | 86 | |
| Breakable half fence | 41.4% | -3.0 | 0.597 | 70 · | |
| Full fence | 40.3% | -4.1 | 0.498 | 62 · | |
| Single wall | 34.8% | -9.6 | 0.090 | 69 · |
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 4.85 on 9 df, p 0.847
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 40.3%62 | -2.70.742 | -5.10.523 | -1.10.897 | -5.80.507 | +5.50.513 | -4.70.640 | -10.60.248 | -8.20.347 | -4.60.595 |
| Breakable full fence | +2.70.742 | 43.0%86 | -2.40.740 | +1.60.841 | -3.10.701 | +8.20.297 | -2.00.835 | -7.90.356 | -5.50.496 | -1.90.812 |
| Half fence | +5.10.523 | +2.40.740 | 45.5%99 | +4.00.603 | -0.70.930 | +10.70.167 | +0.50.961 | -5.40.514 | -3.10.696 | +0.50.946 |
| Breakable half fence | +1.10.897 | -1.60.841 | -4.00.603 | 41.4%70 | -4.70.580 | +6.70.420 | -3.60.716 | -9.40.288 | -7.10.402 | -3.50.677 |
| Low fence | +5.80.507 | +3.10.701 | +0.70.930 | +4.70.580 | 46.1%65 | +11.40.180 | +1.10.908 | -4.70.602 | -2.40.784 | +1.20.887 |
| Single wall | -5.50.513 | -8.20.297 | -10.70.167 | -6.70.420 | -11.40.180 | 34.8%69 | -10.20.291 | -16.10.069 | -13.80.103 | -10.20.224 |
| Fence and open | +4.70.640 | +2.00.835 | -0.50.961 | +3.60.716 | -1.10.908 | +10.20.291 | 45.0%40 | -5.90.569 | -3.50.723 | +0.10.994 |
| Open | +10.60.248 | +7.90.356 | +5.40.514 | +9.40.288 | +4.70.602 | +16.10.069 | +5.90.569 | 50.9%57 | +2.40.794 | +6.00.506 |
| Rectangle | +8.20.347 | +5.50.496 | +3.10.696 | +7.10.402 | +2.40.784 | +13.80.103 | +3.50.723 | -2.40.794 | 48.5%68 | +3.60.673 |
| Octagon | +4.60.595 | +1.90.812 | -0.50.946 | +3.50.677 | -1.20.887 | +10.20.224 | -0.10.994 | -6.00.506 | -3.60.673 | 44.9%69 |
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 counted636
Win rate44.3%
Widest swing-12.1 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 |
|---|---|---|---|---|---|
| Open | 50.0% | +5.7 | 0.412 | 48 · | |
| Rectangle | 49.2% | +4.9 | 0.392 | 67 · | |
| Low fence | 47.4% | +3.0 | 0.630 | 57 · | |
| Half fence | 46.5% | +2.2 | 0.663 | 86 | |
| Full fence | 46.1% | +1.8 | 0.783 | 52 · | |
| Octagon | 45.3% | +1.0 | 0.869 | 64 · | |
| Breakable full fence | 41.3% | -3.0 | 0.577 | 75 · | |
| Fence and open | 40.0% | -4.3 | 0.595 | 35 · | |
| Breakable half fence | 39.7% | -4.7 | 0.433 | 63 · | |
| Single wall | 32.2% | -12.1 | 0.049 | 59 · |
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 6.33 on 9 df, p 0.706
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 46.1%52 | +4.80.590 | -0.40.967 | +6.50.485 | -1.20.899 | +13.90.132 | +6.20.570 | -3.90.701 | -3.10.737 | +0.80.928 |
| Breakable full fence | -4.80.590 | 41.3%75 | -5.20.509 | +1.60.844 | -6.00.489 | +9.10.278 | +1.30.895 | -8.70.346 | -7.90.344 | -4.00.637 |
| Half fence | +0.40.967 | +5.20.509 | 46.5%86 | +6.80.406 | -0.90.920 | +14.30.085 | +6.50.514 | -3.50.698 | -2.70.736 | +1.20.884 |
| Breakable half fence | -6.50.485 | -1.60.844 | -6.80.406 | 39.7%63 | -7.70.396 | +7.50.390 | -0.30.975 | -10.30.278 | -9.60.273 | -5.60.521 |
| Low fence | +1.20.899 | +6.00.489 | +0.90.920 | +7.70.396 | 47.4%57 | +15.20.095 | +7.40.490 | -2.60.788 | -1.90.834 | +2.10.821 |
| Single wall | -13.90.132 | -9.10.278 | -14.30.085 | -7.50.390 | -15.20.095 | 32.2%59 | -7.80.444 | -17.80.062 | -17.10.052 | -13.10.136 |
| Fence and open | -6.20.570 | -1.30.895 | -6.50.514 | +0.30.975 | -7.40.490 | +7.80.444 | 40.0%35 | -10.00.367 | -9.20.373 | -5.30.610 |
| Open | +3.90.701 | +8.70.346 | +3.50.698 | +10.30.278 | +2.60.788 | +17.80.062 | +10.00.367 | 50.0%48 | +0.80.937 | +4.70.623 |
| Rectangle | +3.10.737 | +7.90.344 | +2.70.736 | +9.60.273 | +1.90.834 | +17.10.052 | +9.20.373 | -0.80.937 | 49.2%67 | +3.90.652 |
| Octagon | -0.80.928 | +4.00.637 | -1.20.884 | +5.60.521 | -2.10.821 | +13.10.136 | +5.30.610 | -4.70.623 | -3.90.652 | 45.3%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.
Matches counted80
Win rate45.0%
Widest swing-45.0 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 |
|---|---|---|---|---|---|
| Fence and open | 80.0% | +35.0 | 0.104 | 5 · | |
| Breakable half fence | 57.1% | +12.1 | 0.499 | 7 · | |
| Open | 55.6% | +10.6 | 0.499 | 9 · | |
| Breakable full fence | 54.5% | +9.6 | 0.493 | 11 · | |
| Single wall | 50.0% | +5.0 | 0.734 | 10 · | |
| Octagon | 40.0% | -5.0 | 0.816 | 5 · | |
| Half fence | 38.5% | -6.5 | 0.605 | 13 · | |
| Low fence | 37.5% | -7.5 | 0.653 | 8 · | |
| Full fence | 10.0% | -35.0 | 0.017 | 10 · | |
| Rectangle | 0.0% | -45.0 | 0.363 | 1 · |
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 10.04 on 9 df, p 0.347
| Stage | Full | Break. full | Half | Break. half | Low | Wall | Mixed | Open | Rect | Oct |
|---|---|---|---|---|---|---|---|---|---|---|
| Full fence | 10.0%10 | -44.50.031 | -28.50.123 | -47.10.036 | -27.50.163 | -40.00.051 | -70.00.007 | -45.60.033 | +10.00.740 | -30.00.171 |
| Breakable full fence | +44.50.031 | 54.5%11 | +16.10.431 | -2.60.914 | +17.10.463 | +4.50.835 | -25.40.330 | -1.00.964 | +54.50.296 | +14.60.590 |
| Half fence | +28.50.123 | -16.10.431 | 38.5%13 | -18.70.423 | +1.00.965 | -11.50.580 | -41.50.114 | -17.10.429 | +38.50.439 | -1.50.952 |
| Breakable half fence | +47.10.036 | +2.60.914 | +18.70.423 | 57.1%7 | +19.60.447 | +7.10.772 | -22.90.408 | +1.60.949 | +57.10.285 | +17.10.558 |
| Low fence | +27.50.163 | -17.10.463 | -1.00.965 | -19.60.447 | 37.5%8 | -12.50.596 | -42.50.135 | -18.10.457 | +37.50.453 | -2.50.928 |
| Single wall | +40.00.051 | -4.50.835 | +11.50.580 | -7.10.772 | +12.50.596 | 50.0%10 | -30.00.264 | -5.60.809 | +50.00.338 | +10.00.714 |
| Fence and open | +70.00.007 | +25.40.330 | +41.50.114 | +22.90.408 | +42.50.135 | +30.00.264 | 80.0%5 | +24.40.360 | +80.00.121 | +40.00.197 |
| Open | +45.60.033 | +1.00.964 | +17.10.429 | -1.60.949 | +18.10.457 | +5.60.809 | -24.40.360 | 55.6%9 | +55.60.292 | +15.60.577 |
| Rectangle | -10.00.740 | -54.50.296 | -38.50.439 | -57.10.285 | -37.50.453 | -50.00.338 | -80.00.121 | -55.60.292 | 0.0%1 | -40.00.439 |
| Octagon | +30.00.171 | -14.60.590 | +1.50.952 | -17.10.558 | +2.50.928 | -10.00.714 | -40.00.197 | -15.60.577 | +40.00.439 | 40.0%5 |
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.