Find highest and corresponding values with duplicates

[momcaro]

Hello,
I have a list of tasks (column A) with risk score associated to them (column B), in random order.
For an executive dashboard, I want to list those tasks in order of highest risk score. I have that set up and formulas down but I hit a glitch when I have duplicate risk scores. (I use LARGE which returns the highest score, then I look up the row number, and INDEX gives me the task associated with that row).
But when I have dup risk scores, my formulas return the same task twice. How do I get it to grab the next task with the same risk score?
I need to update the formula that looks up the row number for the nth risk score, but I'm not sure how to.
I have attached a simplified file of what I mean.
Thank you.

 

[eibi]

Does it matter what order the duplicates appear in?

If not, you can use the following array formula in cell E2 and drag down.

=INDEX($A$2:$A$8,IF(1,N(LARGE((D2=$B$2:$B$8)*(ROW($A$2:$A$8)-1),COUNTIF($D$2:D2,D2)))))

*remember, arrays must be confirmed with Ctrl+Shift+Enter

(see attached)

 

[momcaro]

That is exactly what I wanted -no, it doesn't matter the order of the dups.
Thank you (and wow)!!

 

[momcaro]

I get #REF errors if I add rows above. I assume it has to do with the ROW function? How do I fix? Thank you.

 

[momcaro]

I got it... change the -1 to -(whatever number of extra rows above).