--sql2005
select *,DENSE_RANK() over( order by grade) from tb
--sql 2000
select *,排名=(select count(distict grade) from tb where grade>=a.grade)
from tb a

------解决方案--------------------
SQL code

declare @t table (id int,name int,grade int)
insert into @t
select 1,2,80 union all
select 2,2,90 union all
select 3,3,80 union all
select 4,4,88

select *,dense_rank() over (order by grade desc) as rank from @t
/*
id          name        grade       rank
----------- ----------- ----------- --------------------
2           2           90          1
4           4           88          2
1           2           80          3
3           3           80          3
*/

------解决方案--------------------
SQL code
  create table tb (id int,name int,grade int)
  insert into tb
  select 1,2,80 union
  select 2,2,90 union
  select 3,3,80 union
  select 4,4,88
  
  --sql2005
select *,DENSE_RANK() over( order by grade desc) from tb
--sql 2000
select *,排名=(select count(distinct grade) from tb where grade>=a.grade)
from tb a
id          name        grade       
----------- ----------- ----------- --------------------
2           2           90          1
4           4           88          2
1           2           80          3
3           3           80          3

(4 行受影响)

id          name        grade       排名
----------- ----------- ----------- -----------
1           2           80          3
2           2           90          1
3           3           80          3
4           4           88          2

(4 行受影响)