Seriously, why use a strap on plastic guard with the toughest hubs? they're pretty much the first hubs deigned around serious freestyle bmx abuse with huge 17-20mm axles with slip on collars, it only makes sense to make the collars replaceable with a nice slip on guard..
The plastic makes sense because you're avoiding a "rock-hit-rock" approach to protection. when you attach the guard onto the spokes, it spins WITH the wheel. So when you smack your bike into a grind, the plastic just cushions the sharp edge of the obstacle from spiking into the spokes while the impact is spread to most/all the spokes (since the guard leans onto all of them). the spinning of the wheel helps preventing obstacles from constantly hitting one specific spot on your guard every grind, and being able to spin also reduces grind friction.
as for making it into a slip on collar, that'd actually make it work on a COMPLETELY different principle:
you'll never be able to make one guard that fits on every wheel in a way that the guard rests on the spokes evenly and snugly, let alone have it spin with the wheel. this means the guard has to have a lot more rigidity to it to withstand impact and stop itself bending into the spokes. this is now back to a rock-hit-rock approach, which almost seems primitive in this day and age given how much BMX engineerers have learnt over the years.
it would also create an immense point of leverage on the axle/bearing race. imagine this slip-on collar rigid guard as the horizontal blue bar in this:
the red arrow is the impact applied to the guard when you smack your bike onto a rail for a grind, whereas the triangle pivot is the contact area that this collar sits on the axle, against the inner bearing race. the resultant force applied to the bike (since its also where the collar rests against the inner race of the bearing) is the green arrow, which is effectively on the same spot as the triangle pivot in this case. the impact from the grind is, therefore, largely amplified thru leverage. this means you're creating a mechanism that's prone to deformation from a rock hit rock approach and long leverage, and the consequence of when that happens is a possible combination of fucked bearing/bearing race/hub wobble/axle damage.
based on this above i hope you can see that while a collar slip on guard seems like a good idea, it would requie an axle that's hard enough (so maybe steel is needed), a guard itself with very strong rigidity and possibly extra beefiness at the bearing race contact area, all of which are NOT REQUIRED with a guard that strap onto the spokes. it even loses out on how it spreads out impact.
what's really desirable is, therefore, (probably) for the newest gland to be thick enough so that once its attached, it ALMOST fills the gap between the hub flange and the inside of the dropout yet without touching. this way, when you hop onto a grind and land on the guard, it effectively slides it back out to the bottom of the frame dropout and the peg, without putting pressure on the collar whatsoever.