ya when you put it like that it is scary, though in most (keyword "most") driving situations i think it is more of a shear force, as the wheel is pushing directly upwards.
Having said that, sliding sideways on pavement is something that can happen i.e., winter driving and sliding off ice to dry pavement etc. and would probally be a good example of when youll find out exactly how well a spacer will hold up to "cantilevered loading". Which is probally the exact moment when whoever is driving on a set of spacers will wonder, did I pay the extra few bucks for the high end alloy spacers, and how much are they worth to me right now?
It starts out as a Shear force in supporting the weight, but with bending, tension, and compression forces added when cornering. With the offset stud spacers installed they turn it into a bigger complex of forces. The original stresses are still there, but now both at the original studs and then again at the new studs. Since the spacer moves the wheel mounting surface outward it introduces some more bending into the situation. That bending translates into tension and compression plus bending within the spacer. The required shape of the spacer can not help but have some nasty stress risers built into it.
If the wheel back-space combined with the spacer's thickness puts the center of the tire in the stock position then those loads are probably reasonable, though still present.
If the back-space combined with the spacer moves the tire center outwards then the loads and stresses can get ugly pretty fast.
I wouldn't and don't run this type of spacer. I do have a problem with others running them, but not from the "Oh my gawd, yer gonna kill everybody and their grandmother in Minnesota!" Chicken Little point of view. More from the standpoint that the decision was made with information about the downsides lacking. In science all it takes is one exception to ruin a perfectly good theory. I see this the same way. Most of those running them say they're fine (self delusion?), but who wants to be the one who proves that they are not? I'd like to see those making them post the results from an Engineering Analysis, preferably an FEA picture, from an independent & expert source.
With the thin type of spacer that fits between the wheel and the mounting flange the only real problem comes from running alloy wheels that are thick where the wheel stud holes are. Those can reduce the amount of thread engagement to less than the minimum acceptable.
FWIW search out these that are made from steel or are not cast aluminum. Like cast aluminum pinion wedges they will break. Steel or aluminum sheet and plate stock are what is best. They also tend to be both flatter and have their side be more parallel than the castings.
There is a guy local to me who can cut out such spacers on his small CNC plasma table. He has a $60 minimum and I'll bet that you could have several different thickness' made for that $60 Look around, I'll bet that such a service is available in most of the U.S. They may be water-jet cut too.