Virtua Fighter 5 REVO · v1.10 · Every character

Aoi

Mid and high combined30,805 matchesMid ranks27,678 matchesHigh ranks3,127 matches

The same figures as a 1080×1080 card, for posting. Click it to copy the image. Full size ↗

Win rate49.9%

Standing#10 of 19 · tier B

Pick rate4.1%

Most picked#14 of 19

Matches counted30,805

Unique players197

Period2026/03/26 – 2026/09/23

Against the roster

Win rate in each matchup, over decided non-mirror matches

Opponent	Win %	Wins	Losses	Matches
Brad	54.02%	954	812	1,766
Jacky	52.39%	1,556	1,414	2,970
Lei Fei	51.73%	687	641	1,328
Sarah	51.46%	898	847	1,745
Jean	51.28%	782	743	1,525
Vanessa	51.22%	629	599	1,228
Wolf	51.11%	970	928	1,898
Goh	50.32%	855	844	1,699
Pai	49.94%	771	773	1,544
Akira	49.75%	1,683	1,700	3,383
Dural	49.38%	119	122	241
Kage	49.32%	945	971	1,916
Shun	48.97%	808	842	1,650
El Blaze	48.85%	553	579	1,132
Lau	48.02%	1,008	1,091	2,099
Eileen	47.12%	270	303	573
Jeffry	46.33%	890	1,031	1,921
Lion	45.80%	780	923	1,703
Taka	41.32%	200	284	484

By shapeWalls or no wallsBy shape · roundsWalls or no walls · rounds

Stage Impact To Win Probability

Change from this character's own win rate, in points

The same figures as a 1080×1080 card, for posting. Click it to copy the image. Full size ↗

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.

Each matchup, stage by stage

Win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Brad54.0%	54	57	52	50	55	59	56	51	51	57
Jacky52.4%	52	52	49	52	50	51	56	52	59	52
Lei Fei51.7%	48	52	52	54	46	54	43	57	55	54
Sarah51.5%	52	48	47	48	55	56	48	51	57	45
Jean51.3%	53	57	44	49	46	51	50	54	57	54
Vanessa51.2%	44	51	50	51	51	54	47	53	51	54
Wolf51.1%	52	49	51	50	47	53	58	54	55	48
Goh50.3%	52	48	49	46	52	48	53	49	56	51
Pai49.9%	46	52	46	53	51	46	57	44	48	55
Akira49.8%	42	50	53	49	50	50	56	47	46	53
Dural49.4%	55	48	65	27	56	48	17	33	58	67
Kage49.3%	48	47	47	48	53	48	50	53	47	48
Shun49.0%	58	53	53	47	46	43	49	39	50	52
El Blaze48.9%	40	59	62	60	44	48	43	47	47	42
Lau48.0%	46	35	46	55	53	50	41	52	47	52
Eileen47.1%	47	37	42	40	40	35	59	56	51	60
Jeffry46.3%	49	47	45	46	40	49	41	51	37	54
Lion45.8%	46	43	44	47	42	33	51	50	51	48
Taka41.3%	29	56	41	50	37	45	50	37	38	37

Worse on this stageNo changeBetter on this stage

Walls Impact To Win Probability

Against this character's own rate, in points

Walls or no walls-0.2 pts behind walls (95% interval -1.6 to +1.2); within chance across the 20 characters tested, q=0.843

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.

Walls Impact To Win Probability

Win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Brad54.0%	53	57
Jacky52.4%	53	51
Lei Fei51.7%	53	51
Sarah51.5%	51	52
Jean51.3%	52	50
Vanessa51.2%	50	53
Wolf51.1%	52	49
Goh50.3%	50	50
Pai49.9%	49	50
Akira49.8%	48	51
Dural49.4%	48	55
Kage49.3%	48	50
Shun49.0%	50	47
El Blaze48.9%	52	44
Lau48.0%	47	51
Eileen47.1%	46	46
Jeffry46.3%	46	48
Lion45.8%	47	42
Taka41.3%	42	40

Worse on this stageNo changeBetter on this stage

Stage Impact To Round Win Probability

Change from this character's own round win rate, in points

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.

Each matchup, stage by stage, in rounds

Round win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Brad51.7%	52	54	51	49	53	54	53	49	50	53
Jacky51.6%	51	51	50	51	51	51	53	52	54	51
Vanessa51.0%	47	50	50	50	51	53	50	52	52	53
Lei Fei50.9%	49	51	52	52	49	51	48	54	51	52
Jean50.9%	53	53	47	51	48	51	50	51	53	52
Sarah50.9%	51	48	49	48	52	54	50	51	54	48
Wolf50.4%	50	49	50	50	48	51	53	52	52	49
Dural50.2%	52	50	59	37	54	51	26	39	54	59
Shun50.1%	55	52	52	49	47	48	51	45	51	52
Goh49.9%	51	49	49	47	50	49	52	49	52	51
Akira49.9%	46	49	52	49	50	51	53	49	48	52
Kage49.9%	49	48	48	50	52	49	50	52	49	49
Pai49.8%	47	51	49	53	51	47	54	49	48	51
El Blaze49.3%	45	54	56	54	46	50	45	48	49	47
Lau49.1%	48	44	47	53	51	49	47	51	49	51
Eileen48.5%	48	43	44	46	45	44	55	52	54	54
Lion48.0%	48	47	47	49	47	43	50	50	49	49
Jeffry47.5%	49	48	47	48	45	48	45	48	43	51
Taka45.6%	39	53	47	51	43	49	48	41	44	44

Worse on this stageNo changeBetter on this stage

Walls Impact To Round Win Probability

Against this character's own round rate, in points

Walls or no walls-0.3 pts behind walls (95% interval -1.0 to +0.4); within chance across the 20 characters tested, q=0.478

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.

Walls Impact To Round Win Probability

Round win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Brad51.7%	51	53
Jacky51.6%	52	51
Vanessa51.0%	50	52
Lei Fei50.9%	52	50
Jean50.9%	51	50
Sarah50.9%	50	51
Wolf50.4%	51	49
Dural50.2%	49	54
Shun50.1%	51	49
Goh49.9%	50	50
Akira49.9%	49	51
Kage49.9%	49	50
Pai49.8%	49	50
El Blaze49.3%	51	48
Lau49.1%	49	51
Eileen48.5%	48	48
Lion48.0%	48	46
Jeffry47.5%	47	48
Taka45.6%	46	45

Worse on this stageNo changeBetter on this stage

Where they are played

Share of that region's own picks, and where that ranks

Region	Share	Standing	Appearances
Asia	3.74%	16	20,203
South America	4.70%	11	5,276
North America	4.67%	10	3,539
Europe	3.72%	10	648
Middle East	0.00%	9	0
Oceania	0.00%	9	0
Not recorded	6.38%	4	2,323

The same figures as a 1080×1080 card, for posting. Click it to copy the image. Full size ↗

Win rate49.8%

Standing#12 of 19 · tier B

Pick rate4.3%

Most picked#14 of 19

Matches counted27,678

Unique players187

Period2026/03/26 – 2026/09/23

Against the roster

Win rate in each matchup, over decided non-mirror matches

Opponent	Win %	Wins	Losses	Matches
Brad	55.22%	852	691	1,543
Jacky	52.57%	1,430	1,290	2,720
Lei Fei	52.52%	614	555	1,169
Wolf	51.09%	939	899	1,838
Sarah	51.07%	815	781	1,596
Jean	50.92%	748	721	1,469
Vanessa	50.13%	560	557	1,117
Goh	50.12%	810	806	1,616
Lau	49.82%	838	844	1,682
Pai	49.70%	670	678	1,348
Akira	49.52%	1,499	1,528	3,027
Kage	48.98%	839	874	1,713
Shun	48.98%	693	722	1,415
El Blaze	47.73%	474	519	993
Dural	46.19%	103	120	223
Eileen	46.17%	223	260	483
Jeffry	46.01%	836	981	1,817
Lion	45.94%	685	806	1,491
Taka	39.95%	167	251	418

By shapeWalls or no wallsBy shape · roundsWalls or no walls · rounds

Stage Impact To Win Probability

Change from this character's own win rate, in points

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.

Each matchup, stage by stage

Win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Brad55.2%	56	58	53	52	55	62	55	51	52	59
Jacky52.6%	53	53	48	52	50	50	58	53	58	53
Lei Fei52.5%	50	52	52	54	48	56	39	58	55	56
Wolf51.1%	52	49	51	50	46	54	57	54	55	48
Sarah51.1%	52	49	47	48	55	54	48	51	56	44
Jean50.9%	54	57	43	50	45	49	50	54	57	52
Vanessa50.1%	43	50	50	49	47	53	45	51	51	55
Goh50.1%	51	48	49	44	53	48	51	49	56	52
Lau49.8%	49	35	46	58	56	52	46	52	48	52
Pai49.7%	46	52	47	54	49	45	57	41	50	54
Akira49.5%	41	50	53	50	50	50	56	45	46	54
Kage49.0%	46	46	47	47	53	47	49	54	47	47
Shun49.0%	58	53	51	47	47	44	50	39	51	50
El Blaze47.7%	40	58	58	57	42	45	46	49	46	41
Dural46.2%	58	43	52	21	56	45	17	33	58	65
Eileen46.2%	49	37	39	40	36	36	58	57	47	62
Jeffry46.0%	48	45	46	46	41	48	41	50	37	54
Lion45.9%	47	42	42	48	41	33	54	50	52	50
Taka40.0%	32	53	43	47	38	44	47	35	39	29

Worse on this stageNo changeBetter on this stage

Walls Impact To Win Probability

Against this character's own rate, in points

Walls or no walls-0.3 pts behind walls (95% interval -1.8 to +1.2); within chance across the 20 characters tested, q=0.799

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.

Walls Impact To Win Probability

Win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Brad55.2%	54	59
Jacky52.6%	53	51
Lei Fei52.5%	54	53
Wolf51.1%	52	49
Sarah51.1%	51	51
Jean50.9%	52	49
Vanessa50.1%	49	52
Goh50.1%	50	51
Lau49.8%	48	54
Pai49.7%	49	49
Akira49.5%	47	51
Kage49.0%	48	50
Shun49.0%	50	47
El Blaze47.7%	51	43
Dural46.2%	43	53
Eileen46.2%	45	45
Jeffry46.0%	45	48
Lion45.9%	47	41
Taka40.0%	41	37

Worse on this stageNo changeBetter on this stage

Stage Impact To Round Win Probability

Change from this character's own round win rate, in points

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.

Each matchup, stage by stage, in rounds

Round win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Brad52.5%	53	54	52	51	54	55	53	50	50	54
Jacky51.7%	51	51	50	51	52	51	54	52	54	51
Lei Fei51.2%	49	50	53	52	50	52	45	55	52	52
Sarah50.6%	51	49	49	48	52	53	50	50	54	47
Jean50.6%	53	53	47	51	48	50	50	51	53	51
Vanessa50.4%	47	49	50	49	50	52	50	51	52	54
Wolf50.4%	50	49	51	50	48	51	52	52	52	50
Shun50.1%	55	52	52	48	48	48	52	46	51	50
Lau50.0%	49	44	47	54	53	51	49	51	49	51
Goh49.8%	51	49	50	46	51	49	50	49	52	51
Pai49.8%	47	50	49	53	51	47	53	48	49	51
Kage49.8%	48	48	48	49	52	50	50	52	49	48
Akira49.7%	46	48	51	50	50	51	53	48	48	52
Dural48.9%	53	47	54	36	54	50	26	39	54	59
El Blaze48.3%	44	53	54	51	45	48	44	48	48	47
Lion47.9%	49	46	47	48	46	42	51	50	49	50
Eileen47.9%	49	43	43	46	43	45	53	52	51	54
Jeffry47.2%	49	47	47	48	45	47	45	48	43	51
Taka44.9%	40	52	48	50	43	47	46	41	44	40

Worse on this stageNo changeBetter on this stage

Walls Impact To Round Win Probability

Against this character's own round rate, in points

Walls or no walls-0.4 pts behind walls (95% interval -1.1 to +0.3); within chance across the 20 characters tested, q=0.600

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.

Walls Impact To Round Win Probability

Round win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Brad52.5%	52	54
Jacky51.7%	52	51
Lei Fei51.2%	52	51
Sarah50.6%	50	51
Jean50.6%	51	49
Vanessa50.4%	50	52
Wolf50.4%	51	50
Shun50.1%	51	49
Lau50.0%	49	52
Goh49.8%	50	50
Pai49.8%	49	49
Kage49.8%	49	50
Akira49.7%	49	51
Dural48.9%	47	54
El Blaze48.3%	50	47
Lion47.9%	48	46
Eileen47.9%	47	48
Jeffry47.2%	47	48
Taka44.9%	45	44

Worse on this stageNo changeBetter on this stage

Where they are played

Share of that region's own picks, and where that ranks

Region	Share	Standing	Appearances
Asia	3.94%	15	17,248
South America	4.72%	11	5,268
North America	4.89%	9	3,502
Europe	3.50%	12	513
Middle East	0.00%	9	0
Oceania	0.00%	9	0
Not recorded	6.92%	3	2,295

The same figures as a 1080×1080 card, for posting. Click it to copy the image. Full size ↗

Win rate50.0%

Standing#8 of 19 · tier B

Pick rate2.8%

Most picked#17 of 19

Matches counted3,127

Unique players21

Period2026/03/26 – 2026/09/23

Against the roster

Win rate in each matchup, over decided non-mirror matches

Opponent	Win %	Wins	Losses	Matches
Dural	88.89%	16	2	18
Vanessa	62.16%	69	42	111
Jean	60.71%	34	22	56
El Blaze	56.83%	79	60	139
Sarah	55.70%	83	66	149
Goh	54.22%	45	38	83
Eileen	52.22%	47	43	90
Kage	52.22%	106	97	203
Jeffry	51.92%	54	50	104
Akira	51.69%	184	172	356
Wolf	51.67%	31	29	60
Pai	51.53%	101	95	196
Jacky	50.40%	126	124	250
Taka	50.00%	33	33	66
Shun	48.94%	115	120	235
Lei Fei	45.91%	73	86	159
Brad	45.74%	102	121	223
Lion	44.81%	95	117	212
Lau	40.77%	170	247	417

By shapeWalls or no wallsBy shape · roundsWalls or no walls · rounds

Stage Impact To Win Probability

Change from this character's own win rate, in points

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.

Each matchup, stage by stage

Win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Dural88.9%	33	100	100	100	·	100	·	·	·	100
Vanessa62.2%	75	60	45	67	77	64	67	75	44	46
Jean60.7%	38	40	67	33	67	78	·	·	75	100
El Blaze56.8%	42	62	79	90	50	59	33	20	75	60
Sarah55.7%	50	39	48	50	56	75	50	56	88	53
Goh54.2%	60	55	44	73	25	40	71	54	57	33
Eileen52.2%	25	50	50	42	62	29	67	56	62	50
Kage52.2%	62	64	48	59	47	48	53	46	48	54
Jeffry51.9%	57	64	33	33	33	67	50	64	33	50
Akira51.7%	52	53	53	30	52	50	56	62	50	50
Wolf51.7%	55	50	25	50	83	40	100	62	·	25
Pai51.5%	47	53	38	45	58	57	67	60	41	61
Jacky50.4%	50	52	58	52	45	54	17	48	64	30
Taka50.0%	0	71	33	80	33	57	100	50	33	59
Shun48.9%	56	58	60	48	38	38	38	33	42	60
Lei Fei45.9%	33	56	48	57	35	41	62	47	50	38
Brad45.7%	42	50	42	42	53	43	60	46	44	38
Lion44.8%	38	48	67	42	56	40	31	50	46	36
Lau40.8%	30	35	44	46	39	39	18	46	43	49

Worse on this stageNo changeBetter on this stage

Walls Impact To Win Probability

Against this character's own rate, in points

Walls or no walls+0.5 pts behind walls (95% interval -3.7 to +4.8); within chance across the 20 characters tested, q=0.887

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.

Walls Impact To Win Probability

Win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Dural88.9%	87	100
Vanessa62.2%	61	62
Jean60.7%	47	80
El Blaze56.8%	61	55
Sarah55.7%	52	62
Goh54.2%	57	33
Eileen52.2%	50	48
Kage52.2%	53	49
Jeffry51.9%	52	50
Akira51.7%	52	51
Wolf51.7%	52	48
Pai51.5%	47	59
Jacky50.4%	54	46
Taka50.0%	47	52
Shun48.9%	51	47
Lei Fei45.9%	48	38
Brad45.7%	45	43
Lion44.8%	47	43
Lau40.8%	40	43

Worse on this stageNo changeBetter on this stage

Stage Impact To Round Win Probability

Change from this character's own round win rate, in points

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.

Each matchup, stage by stage, in rounds

Round win rate in that matchup on that stage

Opponent	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Dural67.1%	43	86	73	67	·	67	·	·	·	75
Vanessa57.3%	65	56	49	61	64	61	60	65	52	46
Jean57.2%	48	52	61	46	57	65	·	·	62	79
El Blaze56.2%	45	56	65	78	54	57	49	43	77	57
Sarah53.2%	56	44	46	40	52	62	53	56	64	53
Eileen51.9%	44	50	47	46	56	39	73	53	61	52
Goh51.3%	55	55	47	57	33	42	66	49	50	42
Akira51.2%	49	53	54	41	54	51	50	57	49	49
Jeffry51.2%	54	58	43	43	43	54	46	55	43	50
Jacky50.7%	50	52	53	49	48	54	32	49	59	44
Kage50.6%	55	53	47	57	48	48	50	50	49	53
Shun50.6%	57	53	54	53	41	46	43	41	50	58
Pai50.2%	46	54	45	48	50	55	68	55	46	51
Taka49.8%	29	60	44	63	39	58	100	46	41	53
Wolf49.8%	55	44	41	44	65	41	60	58	·	44
Lei Fei49.1%	49	53	50	52	43	40	60	52	48	50
Lion48.4%	47	48	56	49	57	48	42	47	49	44
Brad46.2%	47	51	40	43	50	46	51	47	47	41
Lau45.8%	44	43	46	48	45	44	34	49	48	50

Worse on this stageNo changeBetter on this stage

Walls Impact To Round Win Probability

Against this character's own round rate, in points

Walls or no walls+0.6 pts behind walls (95% interval -1.6 to +2.8); within chance across the 20 characters tested, q=0.729

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.

Walls Impact To Round Win Probability

Round win rate in that matchup with a wall, and without

Opponent	Walls	No walls
Dural67.1%	67	69
Vanessa57.3%	57	57
Jean57.2%	52	65
El Blaze56.2%	58	55
Sarah53.2%	50	56
Eileen51.9%	50	49
Goh51.3%	52	40
Akira51.2%	51	52
Jeffry51.2%	51	49
Jacky50.7%	52	50
Kage50.6%	51	49
Shun50.6%	51	49
Pai50.2%	49	51
Taka49.8%	48	51
Wolf49.8%	51	47
Lei Fei49.1%	51	44
Lion48.4%	49	49
Brad46.2%	46	45
Lau45.8%	46	47

Worse on this stageNo changeBetter on this stage

Where they are played

Share of that region's own picks, and where that ranks

Region	Share	Standing	Appearances
Asia	2.90%	16	2,955
North America	0.88%	18	37
Europe	4.90%	8	135
South America	1.41%	7	8
Not recorded	0.86%	17	28