mmdulken

https://www.youtube.com/watch?v=eKIryzmF-VM Top gear analogy that conveys the same point you're making

https://www.skf-mtrx.de/en/product-details SKF already offers these as a direct drop in for headsets

https://www.vitalmtb.com/photos/features/PIT-BITS-New-Products-and-Prototypes-at-Crankworx-Whistler,9274/Ryan-Howards-Custom-Trek-Session-Park,95761/bturman,109

No, you are wrong. They don't have ZERO drag or instant engagement.

It is intentional to have them all load on the same side because the amount it deflects can be tightly controlled and that is the direction the freehub is flexing. Most would think the freehub would be pushed 180° from the single pawl that is engaged, but its actually rotated into itself. You have to imagine that the tip of the pawl engaged with the drive ring is your fulcrum point. If you apply torque to the system with that as a fulcrum, it is going to inherently rotate back and up. therefore, that is where you want the location of your proceeding pawl to engage. A few things to keep in mind is that the amount of allowable deflection is comparable to most manufacturing tolerances. Every material will flex, you can't think of these metallic components as being 'rigid'. Everything acts as fluid when enough force is applied, but you don't want to exceed the materials yield strength. Bearings come in different allowable clearances.

https://www.youtube.com/watch?v=mB1PIcbgGok I don't think you know what you think you know about points of engagement.