Spigots for me every time (except poles, where deflection curve is very much less important than stiffness, weight and butt diameter). I have yet to see a rod with overfit joints that didn't exhibit a slight change in curve at the junction when under load, and some cheaper ones can be really strange, with more deflection below the ferrule than above.
Overfits make no sense to me, mechanically or in production, save one factor I'll come to later. Having the upper section fit over the lower one means you have two different diameters of tube to marry up, so you have to increase the stiffness/diameter ratio of the lower section to compensate. This means using different grades of carbon, or more layers, for each section - a much more complex process. Conversely a spigot jointed rod can be made as a one piece blank, then cut in half (or thirds) and spigots inserted. A proper spigot should fit together so there's a 1/4" gap between the adjacent parts of the blank, to allow for wear. When the gap closes and the joint becomes loose, you can cut off 1/4" of the female and put a new reinforcing whipping on, so the rod's lifespan is increased.
The only reason I can see why manufacturers use overfits (with all the extra work involved at the blank rolling stage) is that the tolerance need not be so precise. It matters not a jot whether one section slides 2" down the corresponding male, and another only 1.75". You can mass-produce a rod safe in the knowledge that any one of the top sections will fit any of the butt sections, meaning there's no need to keep tips and butts paired up through the finishing stages, and it's easy to supply a replacement joint.