Drilldown Changing Underlying Data Values

Hello,

I am unsure of what is causing the issue I'm experiencing.  I am trying to plot the year over year changes of attendence at regional franchise leadership development events.  I want to drill it down like this , x-axis with fiscal year, drill down from (total attendance, regoinal attendance, franchise attendance).

In order to get YoY change I use the standerd YTD and PYTD measuers:

YTDSummitAttendees = TOTALYTD(COUNT('Leadership Summit Attendees'[Reference ID]), dateDim[Date], "06/30")

PYTDSummitAttendees = CALCULATE([YTDSummitAttendees],  SAMEPERIODLASTYEAR(dateDim[Date]),'dateDim'[isPast] = True)

%YoY = DIVIDE([YTDSummitAttendees], [PYTDSummitAttendees], 0) -1 

When I plot YoY% with respect to fiscal year I get the Correct information:

But when I add a drilldown feature for specific regional information, the values change:

However, the summit drilldown field is correct - but the top drilldown field (fiscal year and YoY%) is no longer correct.  Why would the underlying field in the drilldown be right but the top field not be correct?  When I add another field to the drilldown (YoY% Change of # of franchise attendees at each event) it further changes the top drilldown field, makes the second drilldown field (individual summits) also change.

So again, what is the mechanism that is causing the lowest level of the drilldown to be correct, but the preceding levels to not be correct? Is it a formula issue, or the way drilldown aggregates the descending levels?  Is my data model causing issues here?

The underlying data isn't changing, just the axis.