Churn: Start Date within X days of End Dates

[huffd]

This is a seemingly simple question where I want to know how many IDs have start dates in the current filter context that are within X days (parameter) of any previous endDate.  Sample data are shown below with some annotations.  Both startDate and endDate are connected to the same date table, with startDate being the active relationship.

Assuming the X days parameter is set at 200 days, for each set of rows for a given ID, I would like to identify rows where a StartDate begins within 200 days of a previous EndDate.  These cells are highlighted below:

Those highlighted cells have a StartDate that begins within 200 days of a previous EndDate.  

With that, I would be able to count the number of returns within X days per ID:

or the number of returns within X days by date: 

To approach this, I began with a table of Dates[Year] and 'Sample'[ID], so the filter context is the year of startDate.  Then I need to compare a) a list of IDs with startDates in the current filter context to b) a list of IDs where any previous endDate is within X days of the current startDate. Specifically, I want to filter the list in a to only include the IDs that are in b.

I wasn't exactly sure how to adjust the filter context for b above (the IDs that have endDates within X days of the current startDate). I obviously need to activate the inactive relationship between 'Dates'[Date] and 'Sample'[EndDate], i.e., 

USERELATIONSHIP(Dates[Date],'Sample'[EndDate]). But I'm not clear how to do it.  I tried doing this inside the measure: 

COUNTROWS(

    CALCULATETABLE(

        VALUES( 'Sample'[ID] )

        , FILTER(

            ALL( Dates )

            , Dates[Date] -

                CALCULATE(

                    MIN(Dates[Date])

                    , USERELATIONSHIP(Dates[Date],'Sample'[EndDate])  

                )  

                < __X

        )

    )

)

but that doesnt work.  

How should i approach this instead? 

'Sample':

ID	StartDate	EndDate
1	1/1/2018	6/6/2019
1	1/1/2020	6/6/2021
1	1/1/2022	12/1/2022
2	12/1/2017	3/15/2018
2	8/15/2018	3/15/2019
2	1/1/2020	6/1/2020
3	4/15/2017	12/1/2018
3	8/15/2019	1/1/2020
3	1/1/2021	6/1/2021
4	1/1/2021	12/1/2022
5	3/15/2018	7/20/2018
5	1/1/2019	7/1/2019
5	1/1/2021	1/1/2022
6	6/1/2017	6/1/2018
6	1/1/2022	6/1/2022
7	2/1/2019	10/1/2019
7	6/1/2022	11/1/2022
8	10/1/2019
9	11/1/2022	12/1/2022
10	12/1/2022

[huffd]

@amitchandak I made some clarifications in the original post. hopefull it is more clear now. thank you for your reply.

[amitchandak]

@huffd , not very clear, You can get the last end date with new columns

maxx(filter(Table, [ID] =earlier([ID])  && [end date] < earlier([end Date]) ), [end date])

this can be used to get diff

Power BI DAX- Earlier, I should have known Earlier: https://youtu.be/CVW6YwvHHi8