Virtua Fighter 5 REVO · v1.10 · Rank 21–46

Stage Impact To Match WP

WP = win probability: here, the chance of winning the match.

Rank bandLow, mid and high combined699,157 matchesLow ranks327,096 matchesMid ranks318,500 matchesHigh ranks53,561 matches

StagesBy shape10 typesWalls or no walls2 groups

5.2

Widest swing

V1.10
STAGE IMPACT TO MATCH WP

Rank 21–46

699,157 matches · 5,300 players

2026/03/26 – 2026/09/23

Matches counted699,157

Stage types10

Most affectedTaka

Widest swing5.2 pts

Period2026/03/26 – 2026/09/23

Who the stage matters to

Best stage against worst, in points

Every character, every stage

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

Character	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Taka	-2.3	+0.2	-0.1	+0.3	+0.9	+3.0	+1.3	-1.2	-0.2	-0.4
Eileen	+0.5	+2.3	-1.4	+0.5	-0.3	+0.4	-1.4	-2.7	-0.7	+1.8
Lion	-0.1	-0.6	-0.3	+0.7	-1.2	+1.6	-0.8	-0.6	-0.7	+1.9
Jeffry	+0.5	+0.6	-0.6	+0.1	+0.1	-1.3	+1.2	+0.9	+1.1	-1.9
Vanessa	+0.2	-0.7	+0.3	-0.2	+1.8	+0.3	-0.1	-1.1	+0.6	-0.8
Wolf	-1.0	-0.1	-0.1	-1.1	+0.8	+0.3	-0.7	-0.7	+0.5	+1.6
Lau	-0.2	+1.0	+0.1	-1.1	-1.4	-0.3	+1.0	+1.1	+0.8	-0.8
Goh	+0.0	-0.1	-0.6	-1.2	-0.0	+0.8	-0.1	-0.5	-0.2	+1.3
Kage	+0.1	+0.2	+0.1	+0.4	+0.5	-0.1	+1.6	-0.6	-0.4	-0.6
Sarah	-0.7	+0.5	+0.8	+0.7	-0.1	-0.5	-0.5	+0.8	-1.2	-0.1
Jacky	+0.1	+0.0	+0.8	+0.5	+0.4	-0.6	-0.8	+0.7	-0.3	-1.1
Aoi	-0.5	-0.8	-0.5	-0.6	-0.3	+0.3	+0.7	-0.1	+1.0	+0.6
Brad	+0.4	-0.8	-0.9	+0.3	+0.0	+0.7	+0.9	+0.9	-0.8	-0.1
El Blaze	+0.4	+0.5	-0.7	-0.8	-0.2	+1.0	+0.6	-0.3	+0.9	-0.3
Pai	-0.6	-0.2	+0.1	+0.6	+0.1	-0.4	-1.0	+0.4	-0.2	+0.8
Shun	+0.4	-0.2	-0.5	+0.3	-0.1	+0.4	+0.6	+0.3	-1.1	+0.1
Jean	+0.9	+0.1	+0.4	+0.1	+0.2	-0.7	-0.6	-0.6	-0.7	-0.3
Akira	+0.5	-0.1	+0.6	+0.2	-0.8	-0.7	-0.1	+0.1	+0.6	-0.2
Lei Fei	+0.3	+0.1	-0.2	+0.1	+1.0	-0.4	-0.2	-0.4	+0.1	-0.4
Dural	-0.2	-1.1	+2.1	+1.4	+1.6	-0.0	-5.0	-1.2	+1.1	-0.5

Worse on this stageNo changeBetter on this stage

The matchups the stage decides

One matchup on one stage, against the same matchup everywhere

Matchup	Stage	Overall	On this stage	Change	Matches
Lau vs Aoi	Breakable full fence	52.2%	60.3%	+8.2	421
Pai vs Brad	Half fence	49.6%	57.3%	+7.8	300
Jean vs Lau	Full fence	47.3%	55.0%	+7.6	393
Wolf vs Lau	Low fence	47.4%	55.0%	+7.6	518
Lei Fei vs Wolf	Breakable half fence	49.6%	57.0%	+7.3	316
Lion vs Lau	Open	49.7%	57.0%	+7.3	365
Aoi vs Jacky	Fence and open	48.2%	55.4%	+7.2	352
Jean vs Kage	Breakable full fence	50.8%	57.9%	+7.1	361
Vanessa vs Jacky	Low fence	50.2%	57.2%	+7.0	470
Akira vs Taka	Full fence	44.7%	51.7%	+7.0	352
Sarah vs Shun	Octagon	51.0%	57.3%	+6.3	461
Wolf vs Sarah	Single wall	50.1%	56.4%	+6.2	550
Brad vs Goh	Full fence	51.2%	57.0%	+5.9	319
Aoi vs Lau	Breakable half fence	47.8%	53.7%	+5.8	369
Wolf vs Brad	Open	46.5%	52.3%	+5.8	365
Pai vs Shun	Open	50.0%	55.8%	+5.8	423
Aoi vs El Blaze	Half fence	45.4%	51.1%	+5.8	311
Lion vs Jeffry	Open	47.6%	53.4%	+5.7	401
Vanessa vs Wolf	Breakable half fence	52.1%	57.8%	+5.7	384
Jeffry vs Lion	Breakable half fence	52.4%	58.0%	+5.6	357
Jean vs Pai	Low fence	49.9%	55.6%	+5.6	351
Jean vs Lau	Low fence	47.3%	52.9%	+5.6	395
Wolf vs Jean	Open	45.8%	51.2%	+5.4	473
Kage vs Goh	Breakable half fence	51.3%	56.7%	+5.4	411

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Full fence	70,608	10.10%
Breakable full fence	70,356	10.06%
Half fence	70,092	10.03%
Breakable half fence	70,147	10.03%
Low fence	69,365	9.92%
Single wall	69,182	9.90%
Fence and open	34,758	4.97%
Octagon	69,911	10.00%
Rectangle	70,294	10.05%
Open	69,674	9.97%
Not recorded	34,770	4.97%

Significantly better with wallsSignificantly better without wallsDimmed: could be chance

Two-proportion z, Benjamini–Hochberg at 5% FDR

—

Widest walls gap

V1.10
STAGE IMPACT TO MATCH WP
Walls or no walls

Rank 21–46

699,157 matches · 5,300 players

2026/03/26 – 2026/09/23

Matches counted699,157

Stage groups2

Most affected—

Widest walls gap—

Period2026/03/26 – 2026/09/23

Who the walls suit

Walled stages against open ones, in points

Better in the openBetter behind wallsHolds up after correctionWithin chance95% interval

Is any of this chance?

The whole cut, tested at once

Walled stages dealt evenly across the rosterχ²=22.2 on 19 df, p=0.275

Matches sharing one stage drawabout 2.0, so a match is not an independent draw

Allowance made for thatevery interval and p-value here widened by ×1.30

Characters whose gap holds up0 of 20 — chance alone would give about 1.0

Matchups whose gap holds up0 of 124 — chance alone would give about 6.2

The roster taken togetherχ²=210.1 on 124 df, p=2e−6

Every character, tested

The gap, the range it could be, and the odds of it being nothing

Character	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches	Players
Jeffry	51.5%	50.1%	+1.5	+0.4 … +2.5	0.005	0.054	71,066	930
Lau	51.5%	50.4%	+1.1	+0.1 … +2.1	0.040	0.158	66,238	789
Akira	48.6%	47.8%	+0.9	+0.1 … +1.6	0.019	0.129	138,067	1,132
Jacky	50.2%	49.5%	+0.8	-0.1 … +1.6	0.069	0.172	112,742	1,148
Sarah	50.0%	49.6%	+0.4	-0.6 … +1.4	0.452	0.753	73,779	729
Jean	50.9%	50.6%	+0.3	-0.8 … +1.4	0.620	0.852	57,804	707
Kage	51.0%	51.0%	+0.0	-1.0 … +1.1	0.933	0.984	68,109	829
Dural	52.8%	52.8%	-0.0	-2.6 … +2.6	0.994	0.994	11,103	164
Lei Fei	50.1%	50.1%	-0.1	-1.4 … +1.2	0.935	0.984	43,664	496
Pai	49.9%	50.0%	-0.1	-1.2 … +0.9	0.809	0.952	69,458	787
El Blaze	51.5%	51.7%	-0.2	-1.4 … +1.1	0.758	0.948	47,079	558
Shun	49.5%	49.7%	-0.2	-1.3 … +0.8	0.639	0.852	67,009	608
Brad	50.5%	50.8%	-0.3	-1.5 … +0.8	0.559	0.852	53,796	643
Aoi	47.6%	48.0%	-0.5	-1.5 … +0.6	0.390	0.709	70,619	621
Vanessa	49.4%	50.0%	-0.6	-1.9 … +0.7	0.357	0.709	45,145	526
Eileen	51.4%	52.3%	-0.9	-2.7 … +0.9	0.321	0.709	23,618	395
Lion	50.3%	51.3%	-1.0	-2.1 … +0.1	0.062	0.172	62,492	783
Goh	48.7%	49.8%	-1.1	-2.1 … -0.1	0.035	0.158	68,442	806
Wolf	49.0%	50.3%	-1.3	-2.2 … -0.4	0.005	0.054	86,048	905
Taka	52.7%	54.4%	-1.7	-3.5 … +0.1	0.058	0.172	22,980	471

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.

Every character, every stage

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

Character	Walls	No walls
Jeffry	+0.4	-1.0
Lau	+0.3	-0.8
Akira	+0.3	-0.6
Jacky	+0.3	-0.4
Sarah	+0.1	-0.2
Jean	+0.0	-0.3
Kage	-0.1	-0.1
Dural	+0.4	+0.4
Lei Fei	-0.0	+0.0
Pai	+0.0	+0.1
El Blaze	+0.0	+0.2
Shun	-0.1	+0.1
Brad	-0.1	+0.2
Aoi	-0.2	+0.2
Vanessa	-0.1	+0.5
Eileen	-0.2	+0.6
Lion	-0.3	+0.8
Goh	-0.4	+0.7
Wolf	-0.4	+0.9
Taka	-0.6	+1.2

Worse on this stageNo changeBetter on this stage

The widest matchup gaps

One matchup behind walls against the same matchup in the open, widest first — none of them survives correction

Matchup	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches
Lau vs Wolf	54.2%	48.5%	+5.7	+1.6 … +9.8	0.006	0.734	4,370
Eileen vs Akira	54.0%	48.9%	+5.1	-0.0 … +10.1	0.051	0.744	2,805
El Blaze vs Lion	51.5%	46.5%	+5.0	-0.5 … +10.6	0.076	0.744	2,341
Sarah vs Vanessa	52.6%	47.7%	+4.8	-0.3 … +9.9	0.065	0.744	2,832
Jeffry vs Vanessa	49.3%	44.8%	+4.5	-0.7 … +9.6	0.091	0.744	2,781
Lei Fei vs Wolf	51.5%	47.0%	+4.5	-0.4 … +9.4	0.073	0.744	3,008
Pai vs Brad	51.5%	47.1%	+4.4	-0.6 … +9.3	0.087	0.744	2,893
Sarah vs Wolf	51.1%	46.9%	+4.2	+0.5 … +7.9	0.025	0.744	5,484
Jeffry vs Shun	50.7%	46.7%	+3.9	-0.3 … +8.2	0.071	0.744	4,009
Shun vs Lion	47.2%	43.2%	+3.9	-0.7 … +8.4	0.092	0.744	3,495
Lau vs Aoi	53.5%	49.6%	+3.9	-0.5 … +8.3	0.085	0.744	3,610
Jacky vs Vanessa	50.9%	47.1%	+3.8	-0.3 … +7.9	0.067	0.744	4,487
Kage vs Vanessa	54.1%	50.3%	+3.8	-1.4 … +9.1	0.154	0.903	2,605
Jeffry vs Lei Fei	55.3%	51.7%	+3.6	-1.8 … +9.0	0.188	0.903	2,460
Kage vs Lion	51.1%	47.5%	+3.6	-0.7 … +7.9	0.098	0.744	3,881
Jeffry vs Jacky	51.5%	48.0%	+3.6	+0.3 … +6.8	0.030	0.744	7,009
Shun vs Vanessa	51.9%	48.4%	+3.5	-1.9 … +8.9	0.209	0.903	2,478
Akira vs Wolf	48.7%	45.4%	+3.4	+0.8 … +6.0	0.012	0.744	10,816
Brad vs Aoi	49.4%	46.2%	+3.2	-1.7 … +8.1	0.198	0.903	3,082
Sarah vs Kage	49.4%	46.2%	+3.2	-1.1 … +7.5	0.147	0.903	3,983
Jeffry vs Goh	51.9%	48.9%	+3.0	-1.3 … +7.2	0.169	0.903	4,056
Shun vs El Blaze	51.0%	48.0%	+3.0	-2.3 … +8.2	0.271	0.903	2,656
Sarah vs Lei Fei	52.1%	49.2%	+2.9	-2.3 … +8.1	0.270	0.903	2,734
El Blaze vs Goh	53.6%	50.7%	+2.9	-2.3 … +8.1	0.275	0.903	2,778

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Walls	421,408	60.27%
No walls	208,221	29.78%
Not recorded	34,770	4.97%

StagesBy shape10 typesWalls or no walls2 groups

5.1

Widest swing

V1.10
STAGE IMPACT TO MATCH WP
Low ranks

Rank 21–29

327,096 matches · 4,487 players

2026/03/26 – 2026/09/23

Matches counted327,096

Stage types10

Most affectedLau

Widest swing5.1 pts

Period2026/03/26 – 2026/09/23

Who the stage matters to

Best stage against worst, in points

Every character, every stage

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

Character	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Lau	-1.0	-0.4	+0.5	-1.7	-0.9	-1.0	+1.5	+2.9	+1.5	-2.2
Lion	+0.4	+0.3	+0.2	+1.8	-2.6	+0.8	-0.4	-1.0	-1.2	+2.2
Eileen	+1.4	+2.0	-0.8	+1.3	-1.4	+0.5	-1.3	-2.5	-1.3	+1.9
Wolf	-1.8	-0.6	-0.2	-1.1	+0.9	+1.1	-0.3	-0.5	+0.6	+2.2
Jeffry	+0.0	+0.5	-0.7	+0.7	+0.2	-1.6	+1.8	+1.6	+0.5	-2.1
Vanessa	-1.4	-1.5	+0.9	+2.3	+1.0	-0.6	-2.1	+0.9	+0.7	-0.4
Goh	+1.0	+0.5	-0.6	-2.2	-0.3	+1.3	-0.9	+0.2	-0.6	+0.7
Lei Fei	+1.5	+0.5	-0.7	-0.2	+2.0	-0.7	+0.1	+0.1	-1.1	+0.0
Jean	+0.9	+1.3	+0.1	-0.7	+1.0	-1.6	-0.4	-0.9	-0.9	-0.7
Brad	+1.4	-0.7	-0.2	-0.3	+1.3	+0.2	+1.1	+0.4	-1.3	-0.3
Kage	+0.6	+0.7	-0.7	+0.3	+0.1	-0.5	+1.2	-1.4	+0.2	-0.1
Shun	+1.2	+1.3	-0.4	+0.0	-1.3	+0.8	+0.6	-0.3	-1.2	+0.3
El Blaze	+1.2	+0.7	+0.0	-0.6	-0.2	+0.4	+0.5	-0.5	+0.5	-1.3
Pai	-1.0	-0.8	-0.3	+0.5	+0.8	-0.1	-0.1	-0.1	-0.8	+1.5
Sarah	-1.4	+0.7	+0.9	+0.1	+0.9	-0.7	-1.2	-0.0	-0.2	-0.1
Aoi	-0.1	-1.1	-0.3	-0.7	+0.1	+0.8	+0.9	-0.3	+1.1	+0.1
Taka	-1.4	+0.5	+0.2	+0.5	-1.6	+2.3	-0.8	-0.4	+0.7	+1.1
Jacky	+0.2	+0.5	-0.1	+0.1	-0.2	+0.1	-1.0	+0.8	+0.2	-1.2
Akira	+0.5	-0.8	+0.6	+0.9	-0.6	-0.7	+0.6	-0.1	+0.4	-0.7
Dural	-2.6	-3.3	+4.3	+2.8	+4.5	+1.4	-8.3	-3.8	+0.4	+2.9

Worse on this stageNo changeBetter on this stage

The matchups the stage decides

One matchup on one stage, against the same matchup everywhere

Matchup	Stage	Overall	On this stage	Change	Matches
Sarah vs Lei Fei	Rectangle	50.6%	62.8%	+12.2	121
Shun vs Lion	Octagon	45.5%	57.5%	+12.0	160
Vanessa vs Sarah	Open	49.4%	61.3%	+11.9	142
Lei Fei vs Shun	Single wall	55.4%	66.7%	+11.2	135
Lei Fei vs Wolf	Breakable half fence	48.5%	59.1%	+10.6	127
Shun vs Aoi	Full fence	53.2%	63.0%	+9.8	243
Aoi vs Jacky	Fence and open	45.5%	55.2%	+9.7	230
Lion vs Pai	Full fence	50.3%	59.9%	+9.6	172
Wolf vs Lau	Open	47.1%	56.5%	+9.4	216
Pai vs Vanessa	Open	51.1%	60.1%	+9.0	138
Sarah vs El Blaze	Low fence	49.0%	57.9%	+8.9	164
Lau vs Jeffry	Open	48.0%	56.9%	+8.8	153
Jeffry vs Vanessa	Breakable full fence	47.4%	56.1%	+8.8	139
Pai vs Brad	Full fence	50.0%	58.7%	+8.7	121
Lau vs Wolf	Rectangle	52.9%	61.4%	+8.5	236
Akira vs Eileen	Single wall	46.0%	54.5%	+8.5	189
El Blaze vs Jean	Half fence	47.9%	56.4%	+8.5	126
Jeffry vs Jean	Low fence	46.3%	54.8%	+8.5	166
Sarah vs Shun	Octagon	52.4%	60.9%	+8.5	248
Goh vs Lau	Half fence	48.4%	56.8%	+8.4	125
Pai vs Brad	Breakable half fence	50.0%	58.3%	+8.4	144
Akira vs Taka	Full fence	45.1%	53.5%	+8.4	187
Jeffry vs Kage	Full fence	50.7%	59.0%	+8.3	173
Akira vs Taka	Half fence	45.1%	53.4%	+8.2	178

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Full fence	32,737	10.01%
Breakable full fence	32,985	10.08%
Half fence	33,146	10.13%
Breakable half fence	32,958	10.08%
Low fence	32,406	9.91%
Single wall	32,276	9.87%
Fence and open	16,354	5.00%
Octagon	32,306	9.88%
Rectangle	32,920	10.06%
Open	32,637	9.98%
Not recorded	16,371	5.00%

Significantly better with wallsSignificantly better without wallsDimmed: could be chance

Two-proportion z, Benjamini–Hochberg at 5% FDR

2.0

Widest walls gap

V1.10
STAGE IMPACT TO MATCH WP
Low ranks · Walls or no walls

Rank 21–29

327,096 matches · 4,487 players

2026/03/26 – 2026/09/23

Matches counted327,096

Stage groups2

Most affectedWolf

Widest walls gap2.0 pts

Period2026/03/26 – 2026/09/23

Who the walls suit

Walled stages against open ones, in points

Better in the openBetter behind wallsHolds up after correctionWithin chance95% interval

Is any of this chance?

The whole cut, tested at once

Walled stages dealt evenly across the rosterχ²=20.5 on 19 df, p=0.363

Matches sharing one stage drawabout 2.0, so a match is not an independent draw

Allowance made for thatevery interval and p-value here widened by ×1.27

Characters whose gap holds up1 of 20 — chance alone would give about 1.0

Matchups whose gap holds up0 of 122 — chance alone would give about 6.1

The roster taken togetherχ²=196.7 on 122 df, p=2e−5

Every character, tested

The gap, the range it could be, and the odds of it being nothing

Character	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches	Players
Lau	51.0%	49.4%	+1.7	-0.1 … +3.4	0.057	0.333	23,878	562
Jeffry	52.0%	50.4%	+1.6	+0.2 … +3.1	0.029	0.294	33,489	712
Akira	47.7%	46.8%	+0.9	-0.1 … +1.9	0.067	0.333	71,298	868
Jacky	51.0%	50.2%	+0.7	-0.5 … +1.9	0.227	0.647	50,622	838
El Blaze	51.5%	50.9%	+0.6	-1.2 … +2.4	0.513	0.944	21,379	392
Jean	53.2%	52.8%	+0.4	-1.3 … +2.1	0.655	0.944	23,064	530
Vanessa	51.0%	50.7%	+0.3	-1.6 … +2.3	0.739	0.944	18,784	396
Shun	50.3%	50.1%	+0.2	-1.3 … +1.6	0.806	0.944	32,985	460
Kage	50.3%	50.2%	+0.1	-1.5 … +1.8	0.885	0.944	24,924	563
Sarah	50.0%	50.0%	-0.1	-1.4 … +1.3	0.944	0.944	39,171	565
Lion	50.1%	50.2%	-0.1	-1.7 … +1.6	0.920	0.944	25,620	573
Eileen	50.7%	51.1%	-0.3	-2.7 … +2.1	0.779	0.944	12,159	288
Lei Fei	50.2%	50.7%	-0.4	-2.5 … +1.6	0.676	0.944	16,243	363
Brad	50.8%	51.4%	-0.5	-2.4 … +1.4	0.588	0.944	19,269	464
Aoi	46.2%	46.9%	-0.6	-1.9 … +0.7	0.352	0.782	42,920	493
Taka	52.9%	53.5%	-0.6	-3.2 … +1.9	0.628	0.944	10,704	336
Goh	49.2%	50.1%	-0.8	-2.3 … +0.6	0.259	0.647	31,931	631
Pai	49.5%	50.7%	-1.2	-2.5 … +0.2	0.089	0.358	38,684	608
Wolf	49.9%	51.9%	-2.0	-3.2 … -0.7	0.002	0.032	47,458	708
Dural	55.7%	59.0%	-3.3	-7.3 … +0.8	0.115	0.383	4,160	108

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.

Every character, every stage

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

Character	Walls	No walls
Lau	+0.3	-1.4
Jeffry	+0.4	-1.2
Akira	+0.2	-0.7
Jacky	+0.3	-0.5
El Blaze	+0.2	-0.4
Jean	-0.0	-0.4
Vanessa	+0.3	-0.0
Shun	+0.1	-0.1
Kage	-0.0	-0.2
Sarah	+0.0	+0.1
Lion	+0.1	+0.1
Eileen	+0.0	+0.4
Lei Fei	+0.0	+0.4
Brad	-0.1	+0.4
Aoi	-0.3	+0.3
Taka	+0.0	+0.6
Goh	-0.3	+0.5
Pai	-0.4	+0.8
Wolf	-0.6	+1.4
Dural	-0.4	+2.9

Worse on this stageNo changeBetter on this stage

The widest matchup gaps

One matchup behind walls against the same matchup in the open, widest first — none of them survives correction

Matchup	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches
Eileen vs Akira	57.2%	48.7%	+8.5	+2.0 … +14.9	0.010	0.492	1,655
Jeffry vs Vanessa	50.2%	42.0%	+8.2	+0.5 … +15.7	0.036	0.775	1,231
Lau vs Wolf	54.9%	47.3%	+7.6	+1.6 … +13.4	0.012	0.492	1,971
Jacky vs Eileen	54.4%	47.0%	+7.4	-0.7 … +15.4	0.074	0.775	1,068
Pai vs Brad	52.8%	45.8%	+7.0	-0.3 … +14.2	0.060	0.775	1,301
El Blaze vs Goh	53.0%	46.1%	+6.9	-0.8 … +14.4	0.079	0.775	1,205
Jacky vs Lei Fei	52.4%	46.2%	+6.1	-1.1 … +13.4	0.099	0.775	1,306
Lion vs Pai	52.4%	46.4%	+6.0	-0.2 … +12.2	0.060	0.775	1,744
Vanessa vs Goh	54.5%	48.7%	+5.8	-2.8 … +14.4	0.188	0.947	912
Akira vs Wolf	48.0%	42.4%	+5.5	+2.2 … +8.8	0.001	0.132	6,357
Brad vs Goh	52.8%	47.4%	+5.4	-2.2 … +12.9	0.164	0.947	1,159
Jean vs Aoi	55.0%	49.6%	+5.4	-1.2 … +11.9	0.108	0.775	1,693
Jeffry vs Jacky	51.7%	46.3%	+5.4	+0.7 … +9.9	0.024	0.717	3,216
Kage vs Lau	52.6%	47.4%	+5.3	-2.9 … +13.3	0.206	0.947	1,062
Lau vs Jean	51.1%	45.9%	+5.2	-3.3 … +13.7	0.232	0.947	916
Lau vs Sarah	49.3%	44.1%	+5.2	-1.1 … +11.3	0.105	0.775	1,827
Lei Fei vs Wolf	50.3%	45.2%	+5.1	-1.9 … +12.1	0.156	0.947	1,387
Shun vs Pai	52.1%	47.0%	+5.1	-0.4 … +10.5	0.068	0.775	2,334
Akira vs Brad	49.7%	44.7%	+5.0	+0.0 … +10.0	0.049	0.775	2,775
El Blaze vs Sarah	52.3%	47.5%	+4.8	-1.7 … +11.3	0.150	0.947	1,577
El Blaze vs Lion	51.3%	46.5%	+4.7	-3.6 … +13.0	0.267	0.947	1,000
Brad vs Shun	51.9%	47.4%	+4.5	-3.2 … +12.2	0.251	0.947	1,110
Taka vs Aoi	58.9%	54.7%	+4.2	-4.6 … +13.1	0.351	0.947	836
Lei Fei vs Goh	55.4%	51.3%	+4.1	-4.5 … +12.6	0.349	0.947	889

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Walls	197,052	60.24%
No walls	97,319	29.75%
Not recorded	16,371	5.00%

StagesBy shape10 typesWalls or no walls2 groups

7.7

Widest swing

V1.10
STAGE IMPACT TO MATCH WP
Mid ranks

~

318,500 matches · 2,216 players

2026/03/26 – 2026/09/23

Matches counted318,500

Stage types10

Most affectedTaka

Widest swing7.7 pts

Period2026/03/26 – 2026/09/23

Who the stage matters to

Best stage against worst, in points

Every character, every stage

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

Character	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Taka	-2.9	-0.2	-0.9	+0.7	+4.0	+3.5	+2.3	-3.7	-0.0	-1.1
Vanessa	+2.1	+0.1	+0.2	-2.5	+3.1	+0.5	+0.6	-1.9	+0.1	-1.2
Lion	-0.3	-1.4	-0.0	-0.3	+0.0	+2.8	-0.9	-0.6	-0.7	+1.1
El Blaze	+0.1	+0.4	-1.6	-0.4	-1.1	+2.4	+0.4	+0.5	+1.1	+0.2
Jacky	+0.5	-0.2	+1.9	+1.0	+0.8	-1.9	-1.4	+0.6	-0.4	-1.3
Sarah	-0.3	+0.6	+0.1	+1.3	-0.9	-0.0	+0.3	+1.4	-2.2	+0.4
Pai	+0.2	+0.9	+0.7	+0.8	-0.8	-1.6	-2.2	+1.2	+0.3	-0.0
Jeffry	+0.8	+0.7	-1.0	+0.1	-0.3	-0.6	+0.5	+0.7	+1.6	-1.8
Lau	-0.0	+1.7	-0.4	-1.1	-1.4	-0.1	+1.0	-0.7	+1.1	+0.7
Kage	+0.0	-0.4	+0.4	+1.2	+0.5	+0.6	+1.4	-0.6	-1.5	-0.7
Aoi	-0.9	-0.6	-1.1	-0.3	-1.1	-0.5	+0.7	+0.1	+1.1	+1.6
Brad	-0.5	-0.8	-1.0	+0.7	-1.0	+0.9	+0.7	+1.5	-0.3	-0.4
Shun	-0.4	-1.3	-0.2	+0.1	+0.6	+0.7	+0.9	+0.8	-1.5	-0.4
Wolf	-0.3	+0.7	-0.0	-1.4	+1.1	-1.0	-0.6	-0.7	+0.5	+0.7
Goh	-0.8	-0.5	-0.7	-0.4	+0.0	+0.5	+1.4	-0.5	-0.1	+1.6
Akira	+0.6	+0.6	+0.5	-0.8	-0.8	-0.6	-1.0	+0.2	+1.0	+0.5
Lei Fei	-1.0	-0.4	-0.2	+0.3	+0.4	-0.6	+0.3	-0.8	+0.8	+0.0
Jean	+0.7	-0.9	+0.7	+0.7	-0.1	-0.5	-0.6	-0.2	-0.9	+0.1
Dural	+1.0	+0.2	+1.3	+0.2	+0.0	-0.2	-1.5	+0.2	+0.9	-3.1
Eileen	-1.1	+1.5	-1.9	-1.3	+1.8	+1.5	-2.0	-2.6	+0.0	+1.9

Worse on this stageNo changeBetter on this stage

The matchups the stage decides

One matchup on one stage, against the same matchup everywhere

Matchup	Stage	Overall	On this stage	Change	Matches
Lau vs Aoi	Breakable full fence	50.2%	65.2%	+15.0	161
Lion vs Shun	Low fence	54.9%	67.1%	+12.2	155
Wolf vs Lau	Low fence	47.6%	57.6%	+10.0	255
Brad vs Lion	Octagon	49.8%	59.6%	+9.8	166
Kage vs Pai	Single wall	50.0%	59.6%	+9.7	166
Aoi vs Shun	Full fence	49.0%	58.3%	+9.3	163
Wolf vs Sarah	Single wall	47.7%	57.0%	+9.3	186
Jeffry vs Aoi	Rectangle	54.0%	63.2%	+9.2	174
Jean vs Kage	Breakable full fence	48.9%	58.1%	+9.2	241
Vanessa vs Jacky	Low fence	51.4%	60.4%	+9.0	245
Jeffry vs Lei Fei	Breakable full fence	55.8%	64.5%	+8.7	152
Brad vs Jeffry	Breakable half fence	52.0%	60.6%	+8.6	203
Akira vs Aoi	Full fence	50.5%	59.0%	+8.5	322
Vanessa vs Kage	Low fence	49.0%	57.5%	+8.5	146
Brad vs Akira	Breakable half fence	50.2%	58.6%	+8.4	348
Lau vs Jacky	Single wall	50.9%	59.3%	+8.4	356
Shun vs Wolf	Open	50.7%	59.0%	+8.3	161
Akira vs Pai	Fence and open	51.4%	59.4%	+8.1	143
Pai vs Sarah	Rectangle	51.0%	59.0%	+8.0	200
Aoi vs Lau	Breakable half fence	49.8%	57.8%	+8.0	147
Jacky vs El Blaze	Low fence	46.6%	54.5%	+8.0	242
Kage vs Vanessa	Half fence	51.0%	58.9%	+7.9	151
Aoi vs Jeffry	Open	46.0%	53.9%	+7.9	180
Sarah vs Pai	Single wall	49.0%	56.9%	+7.9	167

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Full fence	32,531	10.21%
Breakable full fence	31,993	10.04%
Half fence	31,534	9.90%
Breakable half fence	31,798	9.98%
Low fence	31,706	9.95%
Single wall	31,515	9.89%
Fence and open	15,745	4.94%
Octagon	32,209	10.11%
Rectangle	31,893	10.01%
Open	31,765	9.97%
Not recorded	15,811	4.96%

Significantly better with wallsSignificantly better without wallsDimmed: could be chance

Two-proportion z, Benjamini–Hochberg at 5% FDR

—

Widest walls gap

V1.10
STAGE IMPACT TO MATCH WP
Mid ranks · Walls or no walls

~

318,500 matches · 2,216 players

2026/03/26 – 2026/09/23

Matches counted318,500

Stage groups2

Most affected—

Widest walls gap—

Period2026/03/26 – 2026/09/23

Who the walls suit

Walled stages against open ones, in points

Better in the openBetter behind wallsHolds up after correctionWithin chance95% interval

Is any of this chance?

The whole cut, tested at once

Walled stages dealt evenly across the rosterχ²=12.3 on 19 df, p=0.871

Matches sharing one stage drawabout 1.9, so a match is not an independent draw

Allowance made for thatevery interval and p-value here widened by ×1.12

Characters whose gap holds up0 of 20 — chance alone would give about 1.0

Matchups whose gap holds up0 of 131 — chance alone would give about 6.6

The roster taken togetherχ²=162.9 on 131 df, p=0.031

Every character, tested

The gap, the range it could be, and the odds of it being nothing

Character	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches	Players
Dural	51.0%	49.3%	+1.7	-1.2 … +4.5	0.257	0.505	6,471	77
Pai	50.6%	49.1%	+1.5	+0.1 … +2.9	0.040	0.142	26,851	263
Jeffry	51.5%	50.1%	+1.4	+0.1 … +2.6	0.029	0.142	34,549	354
Jacky	49.6%	48.3%	+1.3	+0.3 … +2.3	0.009	0.058	53,734	457
Akira	49.3%	48.6%	+0.7	-0.3 … +1.6	0.171	0.380	58,156	422
Sarah	50.4%	50.1%	+0.3	-1.0 … +1.7	0.613	0.799	30,259	238
Lau	50.9%	50.6%	+0.3	-0.9 … +1.6	0.601	0.799	33,025	308
Jean	49.5%	49.3%	+0.2	-1.1 … +1.5	0.757	0.841	32,516	276
Brad	50.5%	50.5%	+0.1	-1.3 … +1.4	0.900	0.900	29,359	253
Lei Fei	49.0%	49.1%	-0.1	-1.7 … +1.4	0.851	0.896	22,767	185
Kage	51.7%	52.0%	-0.3	-1.5 … +0.9	0.646	0.799	35,592	356
Aoi	49.6%	49.9%	-0.3	-1.8 … +1.2	0.679	0.799	24,874	187
Wolf	48.0%	48.4%	-0.5	-1.7 … +0.8	0.443	0.738	35,150	301
El Blaze	51.3%	51.8%	-0.5	-2.1 … +1.1	0.547	0.799	20,839	231
Shun	48.8%	49.5%	-0.8	-2.1 … +0.6	0.278	0.505	28,406	233
Vanessa	49.5%	50.7%	-1.2	-2.8 … +0.4	0.153	0.380	20,677	187
Goh	48.4%	49.6%	-1.2	-2.5 … +0.1	0.068	0.194	33,145	272
Lion	50.9%	52.7%	-1.9	-3.2 … -0.5	0.006	0.058	29,361	296
Eileen	52.2%	54.9%	-2.6	-5.2 … -0.1	0.043	0.142	8,278	151
Taka	52.2%	55.4%	-3.2	-5.5 … -0.9	0.007	0.058	9,879	168

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.

Every character, every stage

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

Character	Walls	No walls
Dural	+0.6	-1.0
Pai	+0.7	-0.8
Jeffry	+0.5	-0.9
Jacky	+0.6	-0.8
Akira	+0.3	-0.3
Sarah	+0.1	-0.2
Lau	+0.1	-0.2
Jean	+0.0	-0.2
Brad	-0.1	-0.1
Lei Fei	-0.2	-0.1
Kage	-0.1	+0.1
Aoi	-0.3	+0.0
Wolf	-0.2	+0.3
El Blaze	+0.0	+0.5
Shun	-0.4	+0.3
Vanessa	-0.3	+0.8
Goh	-0.5	+0.7
Lion	-0.6	+1.3
Eileen	-0.9	+1.8
Taka	-1.2	+2.0

Worse on this stageNo changeBetter on this stage

The widest matchup gaps

One matchup behind walls against the same matchup in the open, widest first — none of them survives correction

Matchup	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches
Shun vs Vanessa	53.3%	44.5%	+8.8	+1.6 … +15.8	0.017	0.813	1,040
Jacky vs Vanessa	50.6%	43.4%	+7.2	+2.0 … +12.3	0.007	0.813	2,053
Lei Fei vs Lion	49.9%	42.9%	+7.0	+0.1 … +13.8	0.047	0.813	1,193
Pai vs Sarah	52.6%	45.9%	+6.7	+0.4 … +13.0	0.037	0.813	1,377
Sarah vs Vanessa	53.1%	46.9%	+6.2	-0.5 … +12.8	0.068	0.813	1,205
Lau vs Aoi	52.0%	46.5%	+5.6	-0.4 … +11.4	0.066	0.813	1,517
Shun vs Lion	46.5%	41.0%	+5.5	-0.2 … +11.1	0.059	0.813	1,603
Aoi vs Lion	47.0%	41.5%	+5.5	-1.0 … +11.8	0.097	0.841	1,344
Sarah vs Wolf	53.6%	48.3%	+5.3	+0.1 … +10.5	0.046	0.813	1,996
Jeffry vs Shun	51.6%	46.4%	+5.3	+0.1 … +10.4	0.045	0.813	1,999
Shun vs El Blaze	51.9%	46.8%	+5.1	-2.0 … +12.2	0.160	0.856	1,072
Brad vs Aoi	46.3%	41.3%	+5.0	-1.1 … +11.0	0.107	0.841	1,398
Sarah vs Kage	48.8%	44.1%	+4.7	-0.6 … +9.9	0.082	0.841	1,943
Lei Fei vs Wolf	52.6%	47.9%	+4.7	-1.2 … +10.6	0.122	0.841	1,494
Kage vs Vanessa	52.7%	48.0%	+4.7	-1.7 … +11.0	0.151	0.856	1,295
Kage vs Lion	51.8%	47.2%	+4.6	-0.3 … +9.5	0.068	0.813	2,177
Wolf vs Goh	51.1%	46.6%	+4.5	-0.7 … +9.7	0.087	0.841	1,989
Jeffry vs Lei Fei	57.4%	53.1%	+4.3	-1.8 … +10.3	0.166	0.856	1,417
Jacky vs Shun	52.1%	47.9%	+4.2	-0.1 … +8.3	0.053	0.813	3,034
Jeffry vs Sarah	54.2%	50.2%	+4.0	-1.4 … +9.4	0.147	0.856	1,818
Lion vs Vanessa	55.2%	51.2%	+4.0	-2.7 … +10.7	0.243	0.856	1,134
Lau vs Wolf	53.5%	49.5%	+4.0	-1.1 … +9.0	0.122	0.841	2,075
Akira vs Aoi	52.6%	48.7%	+3.9	-0.6 … +8.3	0.092	0.841	2,737
Pai vs Jeffry	49.1%	45.3%	+3.9	-2.0 … +9.6	0.196	0.856	1,568

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Walls	191,958	60.27%
No walls	94,986	29.82%
Not recorded	15,811	4.96%

StagesBy shape10 typesWalls or no walls2 groups

12.3

Widest swing

V1.10
STAGE IMPACT TO MATCH WP
High ranks

~

53,561 matches · 433 players

2026/03/26 – 2026/09/23

Matches counted53,561

Stage types10

Most affectedGoh

Widest swing12.3 pts

Period2026/03/26 – 2026/09/23

Who the stage matters to

Best stage against worst, in points

Every character, every stage

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

Character	Full	Break. full	Half	Break. half	Low	Wall	Mixed	Oct	Rect	Open
Goh	-0.3	-1.6	+0.1	+2.1	+2.5	-0.3	-8.4	-5.5	+1.5	+4.0
Taka	-3.2	+0.4	+1.7	-2.2	+0.6	+3.6	+6.1	+6.2	-5.7	-4.1
Jeffry	+0.9	+0.0	+4.0	-5.7	+4.1	-6.8	+3.6	-3.5	+3.2	+0.2
Jean	+2.7	+1.1	-0.7	-0.2	-1.6	+5.5	-7.5	-4.4	+4.5	-0.8
Eileen	+1.7	+5.7	-2.5	+1.8	-2.2	-3.2	+0.1	-3.7	-0.1	+1.5
Wolf	+2.5	-0.4	-0.5	+0.7	-2.6	+2.0	-6.3	-3.2	-1.1	+2.0
Vanessa	-1.2	-0.9	-1.4	+0.2	-1.0	+3.2	+4.0	-4.1	+2.3	-0.8
Sarah	+4.0	-1.4	+4.4	+2.1	-3.1	-2.9	+0.4	+3.3	-3.1	-3.1
El Blaze	-1.3	-0.2	+0.2	-3.2	+3.9	-2.4	+1.6	-2.6	+1.5	+2.7
Lion	-0.5	-0.9	-2.9	+0.8	-0.9	-0.5	-1.9	+1.1	+1.2	+4.2
Shun	+0.1	-3.5	-2.0	+2.9	+3.6	-3.4	-0.9	+1.2	+1.2	+0.6
Lei Fei	+2.6	+0.8	+1.6	+0.1	-0.3	+1.4	-4.2	-0.2	+0.6	-4.2
Aoi	-4.2	+1.6	+2.6	-0.6	-0.8	-0.3	-2.1	+1.9	-1.0	+0.0
Pai	-2.2	-1.4	-0.5	+0.2	-1.1	+3.7	-3.0	+0.9	+1.1	-0.8
Jacky	-2.8	-1.0	-1.0	+0.2	+1.2	+3.6	+3.4	+0.9	-2.6	+1.0
Kage	-1.5	+1.6	+0.9	-2.7	+1.7	-2.0	+3.5	+1.3	+2.7	-2.3
Lau	+1.4	+2.1	+0.8	+0.5	-2.5	+0.8	-0.1	+3.1	-2.1	-2.1
Brad	+1.8	-0.3	-2.9	+0.5	+1.2	+1.1	+1.5	-0.8	-2.2	+2.2
Akira	+0.5	+0.4	+1.6	+1.1	-1.8	-1.0	+0.8	+0.1	-0.6	-0.5
Dural	+3.2	-0.6	-4.0	+3.5	+1.5	-9.4	-37.7	+3.9	+11.3	+5.1

Worse on this stageNo changeBetter on this stage

The matchups the stage decides

One matchup on one stage, against the same matchup everywhere

Matchup	Stage	Overall	On this stage	Change	Matches
Sarah vs Shun	Single wall	45.4%	78.6%	+33.2	28
Pai vs Lei Fei	Open	45.1%	75.0%	+29.9	28
Brad vs Jeffry	Octagon	49.3%	76.0%	+26.7	25
Lau vs Eileen	Half fence	48.7%	75.0%	+26.3	24
Lau vs Wolf	Breakable full fence	51.5%	77.3%	+25.7	22
Jean vs Jacky	Single wall	46.5%	72.0%	+25.5	25
Jacky vs Wolf	Breakable half fence	50.6%	76.0%	+25.4	25
Lion vs Wolf	Rectangle	44.1%	69.2%	+25.1	26
Sarah vs Shun	Half fence	45.4%	69.2%	+23.9	26
Jacky vs Shun	Fence and open	55.5%	79.2%	+23.6	24
Jean vs Akira	Rectangle	48.6%	72.0%	+23.4	25
Kage vs Goh	Octagon	53.4%	76.7%	+23.3	30
Lei Fei vs Brad	Half fence	54.3%	76.9%	+22.6	26
Lei Fei vs Shun	Rectangle	53.5%	75.9%	+22.4	29
Lion vs Shun	Single wall	47.3%	69.0%	+21.8	42
Eileen vs Jacky	Fence and open	46.6%	68.0%	+21.4	25
Eileen vs Shun	Open	50.0%	71.4%	+21.4	21
Lei Fei vs Eileen	Full fence	59.5%	81.0%	+21.4	21
Sarah vs El Blaze	Full fence	41.6%	63.0%	+21.4	27
Akira vs Aoi	Breakable half fence	48.3%	69.6%	+21.3	23
Lion vs Lei Fei	Octagon	45.8%	66.7%	+20.9	39
Kage vs Jeffry	Breakable full fence	52.3%	73.1%	+20.8	26
Sarah vs Akira	Fence and open	46.3%	66.7%	+20.4	24
Taka vs Jean	Breakable full fence	36.0%	56.0%	+20.0	25

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Full fence	5,340	9.97%
Breakable full fence	5,378	10.04%
Half fence	5,412	10.10%
Breakable half fence	5,391	10.07%
Low fence	5,253	9.81%
Single wall	5,391	10.07%
Fence and open	2,659	4.96%
Octagon	5,396	10.07%
Rectangle	5,481	10.23%
Open	5,272	9.84%
Not recorded	2,588	4.83%

Significantly better with wallsSignificantly better without wallsDimmed: could be chance

Two-proportion z, Benjamini–Hochberg at 5% FDR

—

Widest walls gap

V1.10
STAGE IMPACT TO MATCH WP
High ranks · Walls or no walls

~

53,561 matches · 433 players

2026/03/26 – 2026/09/23

Matches counted53,561

Stage groups2

Most affected—

Widest walls gap—

Period2026/03/26 – 2026/09/23

Who the walls suit

Walled stages against open ones, in points

Better in the openBetter behind wallsHolds up after correctionWithin chance95% interval

Is any of this chance?

The whole cut, tested at once

Walled stages dealt evenly across the rosterχ²=19.6 on 19 df, p=0.419

Matches sharing one stage drawabout 2.0, so a match is not an independent draw

Allowance made for thatevery interval and p-value here widened by ×1.09

Characters whose gap holds up0 of 20 — chance alone would give about 1.0

Matchups whose gap holds up0 of 114 — chance alone would give about 5.7

The roster taken togetherχ²=136.1 on 114 df, p=0.077

Every character, tested

The gap, the range it could be, and the odds of it being nothing

Character	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches	Players
Sarah	47.1%	42.6%	+4.5	+1.0 … +7.9	0.011	0.157	4,349	23
Dural	52.2%	48.0%	+4.2	-6.3 … +14.5	0.439	0.711	472	13
Lau	54.9%	52.6%	+2.3	-0.1 … +4.6	0.060	0.399	9,335	59
Lei Fei	54.9%	52.9%	+1.9	-1.4 … +5.2	0.250	0.625	4,654	29
Akira	52.0%	50.4%	+1.6	-0.9 … +4.1	0.199	0.569	8,613	61
Eileen	51.6%	50.0%	+1.6	-2.5 … +5.6	0.456	0.711	3,181	33
Kage	50.3%	49.0%	+1.3	-1.3 … +3.9	0.324	0.711	7,593	67
Jeffry	46.9%	46.3%	+0.6	-3.5 … +4.7	0.783	0.887	3,028	47
Aoi	50.2%	49.6%	+0.5	-3.7 … +4.8	0.808	0.887	2,825	21
Shun	48.2%	48.4%	-0.2	-3.2 … +2.9	0.918	0.918	5,618	32
Jean	46.8%	47.3%	-0.5	-5.2 … +4.3	0.843	0.887	2,224	41
Taka	54.0%	54.6%	-0.6	-5.1 … +4.0	0.799	0.887	2,397	27
Pai	48.1%	49.0%	-0.9	-4.5 … +2.8	0.642	0.856	3,923	33
Wolf	47.9%	48.9%	-0.9	-4.8 … +3.0	0.638	0.856	3,440	33
Lion	48.6%	49.5%	-1.0	-3.6 … +1.6	0.462	0.711	7,511	56
Vanessa	43.6%	45.0%	-1.4	-4.4 … +1.6	0.364	0.711	5,684	30
Brad	48.6%	50.8%	-2.2	-5.3 … +1.0	0.173	0.569	5,168	44
El Blaze	52.4%	54.8%	-2.4	-5.7 … +0.8	0.145	0.569	4,861	40
Goh	47.4%	50.2%	-2.8	-6.8 … +1.1	0.158	0.569	3,366	20
Jacky	50.2%	53.2%	-3.1	-5.6 … -0.6	0.016	0.157	8,386	76

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.

Every character, every stage

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

Character	Walls	No walls
Sarah	+1.4	-3.0
Dural	+2.8	-1.4
Lau	+0.9	-1.3
Lei Fei	+0.9	-1.0
Akira	+0.5	-1.1
Eileen	+0.4	-1.1
Kage	+0.4	-0.9
Jeffry	-0.2	-0.8
Aoi	+0.2	-0.3
Shun	+0.1	+0.2
Jean	+0.4	+0.9
Taka	-0.5	+0.1
Pai	-0.3	+0.5
Wolf	-0.3	+0.6
Lion	-0.2	+0.8
Vanessa	-0.8	+0.6
Brad	-0.7	+1.5
El Blaze	-0.9	+1.5
Goh	-0.7	+2.2
Jacky	-1.1	+2.0

Worse on this stageNo changeBetter on this stage

The widest matchup gaps

One matchup behind walls against the same matchup in the open, widest first — none of them survives correction

Matchup	Behind walls	In the open	Gap	95% interval	p	q (corrected)	Matches
Jeffry vs Shun	53.1%	33.3%	+19.7	+2.8 … +35.0	0.023	0.481	164
Wolf vs Vanessa	59.1%	39.7%	+19.4	+3.3 … +34.1	0.018	0.481	195
Sarah vs El Blaze	48.5%	31.6%	+16.9	+4.6 … +28.2	0.007	0.412	317
Sarah vs Jeffry	47.8%	31.1%	+16.7	-0.0 … +31.6	0.051	0.793	176
Jacky vs Jean	57.6%	42.0%	+15.6	-0.6 … +30.8	0.059	0.793	187
Jacky vs Taka	52.3%	38.6%	+13.7	-4.4 … +30.4	0.140	0.862	145
Lau vs Goh	60.1%	47.2%	+12.9	+1.6 … +23.9	0.025	0.481	418
Sarah vs Jacky	56.6%	43.9%	+12.7	+1.8 … +23.1	0.022	0.481	429
Lau vs Jacky	55.2%	42.6%	+12.6	+4.9 … +20.0	0.001	0.140	933
Pai vs Aoi	53.0%	41.1%	+11.9	-4.0 … +27.0	0.144	0.862	188
Kage vs Jeffry	56.5%	45.4%	+11.2	-2.9 … +24.8	0.122	0.862	254
Jeffry vs Wolf	48.8%	37.7%	+11.1	-6.9 … +27.9	0.231	0.862	141
Lei Fei vs Shun	55.2%	44.4%	+10.7	-3.1 … +24.0	0.130	0.862	304
Taka vs Lau	49.6%	39.1%	+10.6	-3.9 … +24.3	0.153	0.862	230
Sarah vs Brad	49.1%	39.4%	+9.7	-6.8 … +25.2	0.250	0.862	178
Lei Fei vs Pai	58.0%	48.8%	+9.2	-4.9 … +22.9	0.204	0.862	255
El Blaze vs Wolf	62.8%	53.9%	+9.0	-7.2 … +24.9	0.282	0.862	178
Lau vs Jean	68.3%	59.8%	+8.6	-4.9 … +22.2	0.216	0.862	248
Goh vs Lion	50.3%	42.4%	+7.9	-6.2 … +21.4	0.272	0.862	260
Vanessa vs Shun	52.1%	44.4%	+7.8	-4.1 … +19.2	0.200	0.862	362
Eileen vs Jacky	46.6%	38.9%	+7.7	-4.8 … +19.6	0.227	0.862	319
Akira vs Vanessa	58.7%	51.2%	+7.5	-2.1 … +17.1	0.125	0.862	586
Aoi vs Jacky	53.8%	46.3%	+7.5	-9.0 … +23.4	0.381	0.862	227
Akira vs Jacky	53.6%	46.5%	+7.2	-0.4 … +14.6	0.063	0.793	987

Every stage type

How much of the data each stage type is

Stage	Matches	Share
Walls	32,398	60.49%
No walls	15,916	29.72%
Not recorded	2,588	4.83%