Data modelling advice / filtering a table based on unselected filter measure

[J-ace]

I am trying to discover the best way possible to approach a problem, the data looks something similar to table A:

I need to allow the user to filter out (deselect from select all) a country, from this I want to display Table B:

However, I actually need to filter out the categories that are not present in 'Spain', therefore returning something similar to Table C, the average totals for the categories that exist in Spain AND the other countries (if present), but excluding the spain totals:

One approach I have tried it to create an independent table: 'Distinct Country':

Distinct Country = SUMMARIZE('Table A','Table A'[Country],'Table A'[Category])

With an unselected country measure used to filter the table:

 Unselected Country Measure =

var _1 = except(All('Table A'[Country]), allselected('Table A'[Country]))

return

calculate(countrows('Distinct Country'), filter('Distinct Country', 'Distinct Country'[Country] in _1))

 Distinct Country Table With Measure Added:

This returns the filtered result, but I am struggling to filter the original table based on these results, any suggestions would be appreciated or alternative methods to achieving the same result

Many thanks in advance!

[lbendlin]

There are multiple red flags here.  What is the desired behavior when both SPAIN and MEX are deselected?

[J-ace]

Thanks for the reply, for now there is no requirement for multi-select, only single-select filters.

[lbendlin · Accepted solution]

There is no way to "select all but one" elements in a slicer.