I’m not really following, playing telephone and going
Markdown → HTML → text
is nowhere near clean, as HTML doesn’t convert to text properly, there are zillions of edge cases in that, plus “telephone” adds multiple layers of errors in the conversion.
I still think it makes more sense to surgically convert BBCode to plain text, because the Markdown is guaranteed to look OK as plain text. It’s literally in the definition of the Markdown project:
Readability, however, is emphasized above all else. A Markdown-formatted document should be publishable as-is, as plain text, without looking like it’s been marked up with tags or formatting instructions.
The problem is the BBCode, so I think it makes the most sense to focus on the problem rather than going for 3 levels of conversions on everything, even Markdown that does not need to be converted to “plain text”.