Reverse percent of Column total

[kkalyanrr]

Hello,

Can we get a reverse percent of Column total in the below table..I know its a little weird requirement 😞

Output should be 

Name	OverallTest	Percent column of total
A	53	0.00%
B	53	1.68%
C	32	10.06%
D	20	11.17%
E	18	17.88%
F	3	29.61%
G	0	29.61%
Total	179	100%

[AlexisOlson]

This is annoyingly tricky since you need to tie-break the rankings.

My basic approach to this would be to define a ReverseName measure that returns the name from a reverse ranking and then use that in any measure that you want to calculate the "reverse" for.

ReverseName = 
VAR _Names = ALLSELECTED ( Table1[Name] )
VAR _NameCount = COUNTROWS ( _Names )
VAR _CurrName = SELECTEDVALUE ( Table1[Name] )
VAR _Ranks =
    ADDCOLUMNS (
        _Names,
        "@RankName", RANKX ( _Names, [Name],, ASC ),
        "@RankVal", RANKX ( _Names, CALCULATE ( SUM ( Table1[OverallTest] ) ) )
    )
VAR _AscDesc =
    ADDCOLUMNS (
        _Ranks,
        "@ASC",  RANKX ( _Ranks, _NameCount * [@RankVal] + [@RankName],, ASC ),
        "@DESC", RANKX ( _Ranks, _NameCount * [@RankVal] + [@RankName],, DESC )
    )
VAR _AscRank =
    MAXX ( FILTER ( _AscDesc, [Name] = _CurrName ), [@ASC] )
RETURN
    MAXX ( FILTER ( _AscDesc, [@DESC] = _AscRank ), [Name] )

Then PercentReverse can be defined as

PercentInverse =
VAR _Name = [ReverseName]
RETURN
    CALCULATE ( [Percent], Table1[Name] = _Name )

This should work for whatever measure you want to use instead of [Percent].

[AlexisOlson]

It should be possible but it would help if you can explain why you're trying to do this.

[kkalyanrr · Accepted solution]

@AlexisOlson  - I have scenario where in I need to distribute a value across these 7 people..where in person with highest "Overalltest"  should get small chunk and lowest "Overalltest" should get big chunk.

Thanks.