Creating a column based on Date in the past

[LMSReportsHelp]

I have 2 tables, I am trying to create a new column based on Table1.ColumnA (text) and Table2.ColumnB (date).  The tables are set with a  One:Many relationship (the person is assigned 1 class, but might register for multiple class times).

It's pretty simple in a lot of ways, but I must be doing something wrong because I keep checking the 1 person I know (right now) who meets the criteria and they aren't getting my text in their column.  The part in bold (the date) seems to be where I'm hitting the snag.  I've tried several different functions and MAX is the only one I'm not getting an error when I complete my formula, but I'm pretty sure it isn't doing what I want. 

Missed Class =

    var PastClass = IF(TODAY()>MAX(Table2[ColumnB]),1,0)

    var CompletStat = IF(RELATED('Table1'[ColumnA])="Incomplete",1,0)

    RETURN

    CALCULATE(IF(PastClass+CompletStat=2,"Missed Class",""))

Is there a better way to write "if the class is the past and the student shows incomplete, they missed class"? I suspect what I need is something simple and I'm just being a dummy.

[LMSReportsHelp · Accepted solution]

I apparently was over-complicating my formula.....

Missed Class  =

var PastClass = IF(TODAY()>Table2[ColumnB].[Date],1,0)

var CompStat = IF(RELATED(Table1[ColumnA])="Incomplete",1,0)

RETURN

CALCULATE(IF(PastClass+CompStat=2,"Missed Class",""))

[ray_aramburo]

I would tackle this with Power Query but would need a sample of your data removing any sensitive information to provide you a solution example.

[LMSReportsHelp]

Not to sound dumber than I already do, but how would Power Query help me create this column? I've already done some transformation of my data, but that still means I'd need to get the DAX formula right, wouldn't it?

[ray_aramburo]

The functionality of ifs, merging an comparing past dates can be done faster and in my opinion way easier with Power Query than with DAX. At least for this scenario. However, I could still provide you a DAX solution but would need to take a peek to a sample data.