What is warranted is that as long as one has not performed a detailed identification of the more proper solicitations, the simplified ones won't be the best technical estimate available when studying more deeply the problem.
Respect the risk of undersizing the foundation with the pinned scheme, at least in buildings it must not be a very significant problem, since you have a notional scheme able to sustain all the loads above and pass such loads to the foundation; your columns, baseplates, and foundations will be of necessity sized to sustain such loads at the limited deformations required at both service level and limit load level, and experiencie proves the scheme has been used successfully in many occasions without problems -a result on which the generous sizing of concrete footings and conservative estimate of the working bearing loads at the foundations may have been having a say.
You may be right in that a more deeper identification of the solicitations may be necessary, particularly in cases where middle to high cycle dynamical actions are present, and then improper considerations of some actions actually present, typically tensile action on the bolts from moments, may cause unforeseen failure. This has happened lots of times with road signal structuras and of course foundations of machines.
