Matrix visual behavior difference ...

[jp222]

I'm trying to understand a difference in how a matrix is behaving. I think the easiest way is to show with a small example.

Here is a sample table that I am using. Row/Sensor combinations with values for each combo:

I create a measure for average value on this table by filtering using the sensor column:

This produces the following matrix:

This was what I was expecting, I am seeing every row/sensor combination in the matrix but the value is restricted to only averaging those rows in the table with a sensor value of 1.

However, I changed the measure to the following:

This change resulted in the following changes to the matrix visual:

I was expecting that even though the measure is using a different column to filter on that I would still see every row/sensor combination from the original table displayed in the matrix. However, those row/sensor combinations that don't meet the filter criteria in the measure are now not being displayed in the matrix. There are no other filters applied to the matrix visual.

Why is the matrix not showing all of the row/sensor combinations when the second version of the measure is being used? I understand that the value being calculated would not necesarrily be meaningful data for all of the combinations, but shouldn't every combination still appear in the visual (similar to version 1 of the measure)?

Thanks!

[AllisonKennedy]

@jp222  I believe this is due to the filter context and the fact that you're using the Value column both in the Average and in the filter. 

Try: 

Sensor 3 = CALCULATE ( CALCULATE( AVERAGE[ Table[Value]) ), Table[Value] = 10) 

[jp222]

It looked like the average function was missing an opening parenthesis in the modified formula. I added that in but it did not change the visual. The matrix still shows only a subset of the row/sensor combinations as if the measure itself is filtering what is shown on the visual.

Thanks.