Well, the avoidance of docking was primarily because this was a thought experiment, not quite optimizing for a global optima. We wanted to see if a Mars missio

Well, the avoidance of docking was primarily because this was a thought experiment, not quite optimizing for a global optima. We wanted to see if a Mars mission could be mounted within those tight constraints, and set about with those constraints as unwavering. Why avoiding docking was one of them, was a requirement from Mars Direct. (Flipping through A Case For Mars for a minute wasn't enough to find the reasoning...) If you wanted to do a mission with minimal assembly but with docking... I'm not sure you could get a ship with a dry mass of <25T to launch to LEO. Say, with a sizeable upper stage for orbit insertion, habitat module and life support storage space (even with the food / water / fuel all brought up later). Even a fairly large Bigelow module, electrical equipment and pressurized space for food is going to be more than a Delta IV-H can take to LEO... And then an empty upper stage big enough needs a lot of volume. Maybe two flights for assembly, and then launch supplies up?