My view of the tube, from Jon's post below, is that it was a classic tube bend on the fore-aft axis. The thin walled tube is elliptical axis -- compressed by the bend.

Your point about stabilizing the dingy so that you minimize the dynamic loads on a set of davits is -- first by designing the structure to resist lateral forces and SECONDLY, by minimizing those by restraining the motion. In the case of any load, the dynamic loads which involve resisting deaccelerations are usually the failure issue. (That includes the roll and pitch of the boat as well as the oscillations of the dingy.)

My view of that structure is that some either neglected any dynamic loading; and, or that person really never really calculated the static loads property. Those fittings and thin walled tubes are not designed for bending -- they are intended to work more like a simple truss where the members are in much more straight forward compression or tension.

Anyway, AS your post above notes, torsion and lateral loads need to be taken into account and protected against. In a seaway, a boat and anything attached to it is going to be moving around a lot in odd ways.