Filter context of table viz

[CNENFRNL]

Dear gurus,

Once again, I can't wrap head around a seemingly silly simple question: what is the filter context of a table viz?

I even came across a funny scene in this regard.

Here's a very simple dataset,

Date	Cases
2020-01-01	1
2020-01-02	2
2020-01-03	3
2020-01-04	4
2020-01-05	5

I'd retrieve the previous number of each day, i.e. 

Date	Cases	Previous
2020-01-01	1
2020-01-02	2	1
2020-01-03	3	2
2020-01-04	4	3
2020-01-05	5	4

here's the measure

Measure = 
VAR __prev = MAX ( DS[Date] ) - 1
RETURN CALCULATE ( MAX ( DS[Cases] ), DS[Date] = __prev )

When I drag columns into a table viz by order of Cases - Date - Measure, I got ONLY BLANK TABLE

whereas I got an expected result if I drag columns into a table viz by order of Date - Cases - Measure.

No any other viz, slicers or filters exist in my scenario. Therefore, I suspect that such a funny scene results from the filter context, which influence the calculation of the measure. I googled but it turned out (perhaps I didn't find) any explicit clarification of such a subject: what on earth is the filter context of a table viz? what's even more, could it be possible that order of columns in table viz produce different filter context?

Your help would be highly appreciated! TIA

(I attached the file for your trial)

[amitchandak]

@CNENFRNL , Try a new column like

Column = maxx(FILTER('Table (2)', 'Table (2)'[Date] <EARLIER('Table (2)'[Date])),LASTNONBLANKVALUE('Table (2)'[Date],'Table (2)'[Cases]))

[wdx223_Daniel]

@CNENFRNL that did not happen in my computer, whatever the order of draging columns into Table viz, the result is normal and correct, depending on the field of Cases is Whole Number.

but ,when the Cases is in Text type, the result is blank like yours, even change the order to drag the columns, it is still blank.

then i adjusted the code to add a filter statement, the normal result came back.

Measure = 
VAR __prev = MAX ( DS[Date] ) - 1
RETURN CALCULATE ( MAX ( DS[Cases] ), DS[Date] = __prev ,ALL(DS[Cases]))