First I'd like to express my appreciation for the analysis that John, Doug, Ben, Ron and the others have put into this. It is clear to me that this changes how the load is transmitted though the hitch, but it's not clear to me that the change is sufficient to reject the design. The latch appears to only be required to carry the load back to a pin or ramp that transfers the load out to the body of the coupler. As long as the latch mechanism is in compression and the pin that carries the load out to the coupler are reasonably strong, the design should work.
From my (limited) point of view, in conventional designs the latch mechanism already has to carry the force of the trailer when backing up (static load), and it already has to carry any impact that occurs when the truck decelerates faster than the trailer (hitting large pothole with the truck), and because existing latch designs seem to be doing this without catastrophic failure, the existing latch designs must be reasonably sound.
One application I can see for this is for the minivan + 3500# trailer combination. On the light weight end of the towing spectrum, traditional WD hitches add significantly to the tongue weight, and this design is probably 100# lighter than other designs, so perhaps this WD design is appropriate even though there may be limitations that make it unsuitable for heavy trailers?
--Mike