RANKX Circular Reference Help

[DaveAl]

Hi All,

I feel like this one should be really simple, but I can't seem to work it out. I have a very basic, static table with a persons name and a score. I would like to rank those score within the individual. Example:

Dataset:

Person	Score
John	50
John	75
John	50
John	100
Tom	50
Tom	75
Tom	100
Tom	100
Dave	50
Dave	75
Sally	50

Desired Result:

Person	Score	Rank
Dave	50	2
Dave	75	1
John	50	3
John	50	3
John	75	2
John	100	1
Sally	50	1
Tom	50	3
Tom	75	2
Tom	100	1
Tom	100	1

If I use Rank = Rankx('Table','Table'[Score],,DESC,DENSE) it will rank all of the scores, not bound to an individual. 

If I use Rank = Rankx(ALLEXCEPT('Table','Table'[Person]),'Table'[Score],,DESC,DENSE) then I get a circular dependancy error. 

Am I missing something obvious here? I feel like I am...

Thanks again for all your help!

[AlexisOlson · Accepted solution]

The ALLEXCEPT function is a bit weird when used as a table function. I you're trying to write a calculated column, I'd suggest this version instead.

RANKX (
    FILTER ( 'Table', 'Table'[Person] = EARLIER ( 'Table'[Person] ) ),
    'Table'[Score],, DESC, DENSE
)

[DaveAl]

Hi AlexisOlson,

Works like a charm! Thank you for the assist.