Count Column by Group and filter Part 2.

Ah, I understand what you mean now.

This requires rather different logic we need to incorporate the concept of adjacent rows using the index.

Try this:

New Many =
VAR _CurrPID   = 'Table'[Project ID]
VAR _CurrIndex = 'Table'[Index]
VAR _CurrName  = 'Table'[Stage_Name]
VAR _SubTable_ =
    FILTER (
        'Table',
        'Table'[Project ID] = _CurrPID
            && 'Table'[Index] >= _CurrIndex
            && 'Table'[Stage_Name] = "Tech Review"
    )
RETURN
    COUNTROWS (
        FILTER (
            _SubTable_,
            VAR _NextName =
                LOOKUPVALUE (
                    'Table'[Stage_Name],
                    'Table'[Project ID], _CurrPID,
                    'Table'[Index], 'Table'[Index] + 1
                )
            RETURN
                _CurrName = "Tech Review" && _NextName <> "Tech Review"
        )
    )