How to eliminate outliers.

  I'd like to remove the top and bottom N values. I think I'm close.

Here's what I'm working with where I fudge the rownum values to make the query work
as I know there are 220 rows. So I want all values without the top and bottom N.

How can I at run time do something like

rownum <= count(*) -N

For rnum it would be easy enough as I could just do....

rnum >= +N


select * from
    (
           select p.*, rownum rnum
                   from (
                select PRCSNAME,datediff('mi',enddttm,begindttm )*-1
                from PROCESS_SCHEDULER_HISTORY where prcsname=<job name>
                order by datediff('mi',enddttm,begindttm )*-1 desc