I don't remember which study it was, but as I recall, when they examined the Attleboro (and Mansfield?) alternatives, the simulator literally crashed trying to fit all of the rush hour trains. Specifically, the problem was that those alternatives prevented the reuse of the Stoughton Line's slots, and instead forced the simulator to fit in additional slots for South Coast trains in addition to the Stoughton trains. If I remember correctly, the AM rush had an OTP of, like 60% (as opposed to >90% for the Stoughton alts), and the PM rush just made the simulator crash.
- Go via Attleboro. Either you build a new 3-mile bypass (as proposed in the 1990s), or have a reverse move at the station. Either way, you have to deal with creating NEC congestion and the Taunton grade crossings. Rather indirect route.
- Go via Mansfield. The original ROW through downtown is redeveloped, so you'd have to build a new ROW along 140 and 495. Deals with NEC congestion (slightly less than Attleboro) and the Taunton grade crossings. Fairly direct route.
I realize "the simulator crashed" probably isn't as dire as it sounds, but it cracked me up when I first read the study and it cracks me up now.