题主代码有误无法用SQL统计出 百分比Rusult， 请专家解答。

if object_id('[tempdb]..#tb') is not null drop table #tb;
go
create table #tb (id int,[R1]int,[R2] int ,[R3] int ,[R4]INT ,[R5] int ,[R6]int,NOTEXT VARCHAR(20));
go
insert into #tb
 select '1','1','3','18','19','26','29','01 03 18 19 26 29'
union all select '2','1','3','5','18','22','23','01 03 05 18 22 23'
union all select '3','1','7','8','10','12','24','01 07 08 10 12 24'
union all select '4','6','8','22','24','25','26','06 08 22 24 25 26'
union all select '5','4','11','13','22','25','32','04 11 13 22 25 32'
union all select '6','19','20','23','27','28','31','19 20 23 27 28 31'
union all select '7','4','20','22','24','26','33','04 20 22 24 26 33'
union all select '8','4','15','17','22','29','32','04 15 17 22 29 32'
union all select '9','6','14','15','19','29','31','06 14 15 19 29 31'
union all select '10','1','9','15','16','19','21','01 09 15 16 19 21'
union all select '11','4','5','12','16','22','30','04 05 12 16 22 30'
go
;with t as(
select *,
(R6-R5+R4-R3+R2-R1+0) AS AC,
(R6-R5+R4-R3+R2-R1+1) AS AC1,
(R6-R5+R4-R3+R2-R1+2) AS AC2,
(R6-R5+R4-R3+R2-R1+3) AS AC3
 from #tb
 )
select a.id,a.notext,a.R1,a.R2,a.R3,a.R4,a.R5,a.R6
,isnull(tt.query('<e> { for $i in e/r  return data($i)} </e>').value('.','varchar(20)'),' ') as same
,isnull(tt.value('count(e/*)','int'),0) as cnt 
from t a left join t b on a.id-1=b.id
outer apply (select xmlcode =cast('<n>'+REPLACE(a.notext,' ','</n><n>')+'</n>'
                                 +'<r>'+REPLACE(b.AC3,' ','</r><r>')+'</r>' as xml)) c1
outer apply(select tt=xmlcode.query('
                                <e> {    for $i in /r
                                        where data($i) =data(/n)
                                        return $i
                                     }    
                                </e>    

                                '))c3
order by id

希望解决的问题：
①、用公式R6-R5+R4-R3+R2-R1+n取得的AC数字 与 上一行R1,R2,R3,R4,R5,R6分别比对，相同的次数累计数为 Val如何取得？
②、统计的百分比Rusult =Val / max(RID)如何批量取得？
③、用SQL统计出当n=？时统计的百分比Rusult为最大。

找出了N值列表，最大的是第一个：
其他的需求参考这个思路看看能否解决

IF OBJECT_ID('[tempdb]..#tb') IS NOT NULL
    DROP TABLE #tb;
GO
CREATE TABLE #tb
(
 id INT, [R1] INT, [R2] INT, [R3] INT, [R4] INT, [R5] INT, [R6] INT, NOTEXT VARCHAR(20)
);
GO
INSERT #tb
    (id, R1, R2, R3, R4, R5, R6, NOTEXT)
VALUES
(1, 1, 3, 18, 19, 26, 29, '01 03 18 19 26 29'    ),
(2, 1, 3, 5, 18, 22, 23, '01 03 05 18 22 23'    ),
(3, 1, 7, 8, 10, 12, 24, '01 07 08 10 12 24'    ),
(4, 6, 8, 22, 24, 25, 26, '06 08 22 24 25 26'    ),
(5, 4, 11, 13, 22, 25, 32, '04 11 13 22 25 32'    ),
(6, 19, 20, 23, 27, 28, 31, '19 20 23 27 28 31'    ),
(7, 4, 20, 22, 24, 26, 33, '04 20 22 24 26 33'    ),
(8, 4, 15, 17, 22, 29, 32, '04 15 17 22 29 32'    ),
(9, 6, 14, 15, 19, 29, 31, '06 14 15 19 29 31'    ),
(10, 1, 9, 15, 16, 19, 21, '01 09 15 16 19 21'    ),
(11, 4, 5, 12, 16, 22, 30, '04 05 12 16 22 30'    )
GO
SELECT      up.Nums AS Number, COUNT(Nums) Times, MaxID, CAST(COUNT(Nums) * 100.0 / MaxID AS NUMERIC(10, 2)) [Percent]
FROM        (   SELECT  *,
                        LAG(R1) OVER (ORDER BY t.id) - N N1,
                        LAG(R2) OVER (ORDER BY t.id) - N N2,
                        LAG(R3) OVER (ORDER BY t.id) - N N3,
                        LAG(R4) OVER (ORDER BY t.id) - N N4,
                        LAG(R5) OVER (ORDER BY t.id) - N N5,
                        LAG(R6) OVER (ORDER BY t.id) - N N6,
                        MAX(t.id) OVER() MaxID
                FROM    (SELECT *, R6 - R5 + R4 - R3 + R2 - R1 AS N FROM #tb) t ) tt
    UNPIVOT (   Nums
                FOR NN IN (N1, N2, N3, N4, N5, N6)) up
GROUP BY    up.Nums,up.MaxID
ORDER BY    COUNT(Nums) DESC;

发帖子时，以下文字发送不了，所以发的是图片，特借此补充说明如下
按照公式 ：R6-R5+R4-R3+R2-R1+n

手动取值结果：n=0 时 R6-R5+R4-R3+R2-R1+0得到 AC
n=1 时 R6-R5+R4-R3+R2-R1+1得到 AC1
n=2 时 R6-R5+R4-R3+R2-R1+2得到 AC2
n=3 时 R6-R5+R4-R3+R2-R1+3得到 AC3
......
......
其中 n 必须满足 1<R6-R5+R4-R3+R2-R1+n<=33

手动统计结果：统计的百分比Rusult = Val / max(RID)*100

                  n=0 时 Rusult =0
                  n=1 时 Rusult =27.27%
                  n=2 时 Rusult =54.55%
                  n=3 时 Rusult =90.91%

......
......Rusult =90.91%为最大时结果：n=3

想要解决的问题：
①、用公式R6-R5+R4-R3+R2-R1+n取得的AC数字 与 上一行R1,R2,R3,R4,R5,R6分别比对，相同的次数累计数为 Val如何批量得到？
②、统计的百分比Rusult =Val / max(RID) 如何批量得到？
③、用SQL统计 当百分比Rusult为最大时n 的值是多少？

点赞！感谢帮助，学习 收藏了
修学习代码后，添加了条件，但报错，继续请教
①、大师找出了N值列表，最大的是第一个，非常好。题主手动取n=0时 percent=0，这个手动算出来的结果在N值列表里没有出现，希望这个结果不会成为漏网之鱼。
②、感谢大师找出了N值列表，在N值列表里，Number<0是可以的，但要求1<R6-R5+R4-R3+R2-R1+Number<=33
③、如何添加条件 1<R6-R5+R4-R3+R2-R1+Number<=33

SELECT      up.Nums AS Number, COUNT(Nums) Times, MaxID, CAST(COUNT(Nums) * 100.0 / MaxID AS NUMERIC(10, 2)) [Percent]
FROM        (   SELECT  *,
                        LAG(R1) OVER (ORDER BY t.id) - N N1,
                        LAG(R2) OVER (ORDER BY t.id) - N N2,
                        LAG(R3) OVER (ORDER BY t.id) - N N3,
                        LAG(R4) OVER (ORDER BY t.id) - N N4,
                        LAG(R5) OVER (ORDER BY t.id) - N N5,
                        LAG(R6) OVER (ORDER BY t.id) - N N6,
                        MAX(t.id) OVER() MaxID
                FROM    (SELECT *, R6 - R5 + R4 - R3 + R2 - R1 AS N FROM #tb
                where (R6-R5+R4-R3+R2-R1+Number) between 1 and 33  -- 如何添加条件 1<（R6-R5+R4-R3+R2-R1+Number)<=33,如何同时允许 Nums=0 可以使查询成立
                ) t ) tt
    UNPIVOT (   Nums
                FOR NN IN (N1, N2, N3, N4, N5, N6)) up
GROUP BY    up.Nums,up.MaxID
ORDER BY    COUNT(Nums) DESC;

(11 行受影响)
消息 207，级别 16，状态 1，第 35 行
列名 'nNums' 无效。
消息 207，级别 16，状态 1，第 35 行
列名 'nNums' 无效。
SQL Server 分析和编译时间:
CPU 时间 = 0 毫秒，占用时间 = 0 毫秒。

这段代码的逻辑就是找出匹配上一行的数字，如果列表里的数字都在1到33之间，那么不用加任何条件。否则去掉上一行的不在区间的数字即可。比如case when LAG(R1) between 1 and 33 then ...... else null end
另外，如果匹配完后没有0，直接再union all一个0的上去就完事了。

可能理解不正确，结果和你贴出来的不一样：

IF OBJECT_ID('[tempdb]..#tb') IS NOT NULL
    DROP TABLE #tb;
GO
CREATE TABLE #tb
(
 id INT, [R1] INT, [R2] INT, [R3] INT, [R4] INT, [R5] INT, [R6] INT, NOTEXT VARCHAR(20)
);
GO
INSERT #tb
    (id, R1, R2, R3, R4, R5, R6, NOTEXT)
VALUES
(1, 1, 3, 18, 19, 26, 29, '01 03 18 19 26 29'),
(2, 1, 3, 5, 18, 22, 23, '01 03 05 18 22 23'),
(3, 1, 7, 8, 10, 12, 24, '01 07 08 10 12 24'),
(4, 6, 8, 22, 24, 25, 26, '06 08 22 24 25 26'),
(5, 4, 11, 13, 22, 25, 32, '04 11 13 22 25 32'),
(6, 19, 20, 23, 27, 28, 31, '19 20 23 27 28 31'),
(7, 4, 20, 22, 24, 26, 33, '04 20 22 24 26 33'),
(8, 4, 15, 17, 22, 29, 32, '04 15 17 22 29 32'),
(9, 6, 14, 15, 19, 29, 31, '06 14 15 19 29 31'),
(10, 1, 9, 15, 16, 19, 21, '01 09 15 16 19 21'),
(11, 4, 5, 12, 16, 22, 30, '04 05 12 16 22 30')
GO
DROP TABLE IF EXISTS #result
SELECT     up.Nums AS Number, COUNT(Nums) Times, CAST(COUNT(Nums) * 100.0 / MaxID AS NUMERIC(10, 2)) [Percent]
INTO #result
FROM        (   SELECT  *,
                        case when LAG(R1) OVER (ORDER BY t.id) between 2 and 33 then LAG(R1) OVER (ORDER BY t.id) - N else null end N1,
                        case when LAG(R2) OVER (ORDER BY t.id) between 2 and 33 then LAG(R2) OVER (ORDER BY t.id) - N else null end N2,
                        case when LAG(R3) OVER (ORDER BY t.id) between 2 and 33 then LAG(R3) OVER (ORDER BY t.id) - N else null end N3,
                        case when LAG(R4) OVER (ORDER BY t.id) between 2 and 33 then LAG(R4) OVER (ORDER BY t.id) - N else null end N4,
                        case when LAG(R5) OVER (ORDER BY t.id) between 2 and 33 then LAG(R5) OVER (ORDER BY t.id) - N else null end N5,
                        case when LAG(R6) OVER (ORDER BY t.id) between 2 and 33 then LAG(R6) OVER (ORDER BY t.id) - N else null end N6,
                        MAX(t.id) OVER() MaxID
                FROM    (SELECT *, R6 - R5 + R4 - R3 + R2 - R1 AS N FROM #tb) t ) tt
    UNPIVOT (   Nums
                FOR NN IN (N1, N2, N3, N4, N5, N6)) up
GROUP BY    up.Nums,up.MaxID
ORDER BY    COUNT(Nums) DESC;

SELECT * FROM #result r
WHERE NOT EXISTS (SELECT * FROM #tb t WHERE R6 - R5 + R4 - R3 + R2 - R1 + r.Number NOT BETWEEN 1 AND 33)

原理弄明白了就好了，具体的需求可以多琢磨，慢慢想办法解决。