[hallmark]

If they're faced with a low probability landing scenario, such as the fastest re-entry velocity to date, and they need to "test" their control algorithms, which first stage would they choose? A near end-of-life used booster or a late block booster from which they would hope to get another several good launches? - assuming all others characteristics support the current launch parameters.

They're amazingly deliberate in how they mingle bleeding edge R&D and commercial activity. Something that fascinates me. And rockets.