I need to rank a list based on two columns. So, first I have to rank all the data based on column1 if value is above certain no. but if cells have same values then rank those particular cells based on column2 and once the value in column 1 is below that threshhold rank based on column2 but if cells have same values then rank those particular cells based on column1.


If column A value is greater than 1000 then First rank by columnA, if conflict in ranking (same value in column A) then rank by columnB, else rank by columnB, if conflict in ranking (same value in columnB) then rank by columnA again.



Here is an example
ColumnA....ColumnB.....Rank
4999....2.56......1 <<colA is greater than 1000, but rank based on colB
4999....1.59......2<<colA is greater than 1000, but rank based on colB
3149....3.59......3<<rank based on columnA
2482....0.00......4<<rank based on columnA
1712....0.00......5<<rank based on columnA
1422....0.73......6<<rank based on columnA
184......4.73......7<<colA smaller than1000, so rank based on colB
554......0.00......8<<colB has same values, rank using 4colA values only
377......0.00......9
298......0.00......10
196......0.00......11

and so on.....

I tried this but obviously it gives me same rank number where column A and column B has same values...
=IF(A2>=1000,RANK(A2,$A$2:$A$42,0),RANK(B2,$B$2:$B$42,0)+COUNTIF($A$2:$A$42,">=1000"))

from this I get rank as following...
1
1
3
4
5
6
7
8
8
8
8

I hope this is clear enough...
Thanks, Jay