VR definitely needs a reinvented wheel in this case. Being able to move your head in virtual space but not your hands is clearly suboptimal, and the motion sensors in that generation of controllers are not up to the task.
My take: this a bandaid, so they can get the headset out in Q1, ahead of the proper controllers in Q2. Not what they wanted to do, but Vive forced their hand on timing.
I'm curious as to why you think that freedom of hand motion is essential for VR. Certainly I can think of many VR experiences where it would be useful, but I can think of just as many where having discreet inputs (buttons) would be preferable.
For example, a spaceship piloting game that allows the players head to rotate and look out the windows could very easily be controlled with a traditional gamepad.
I definitely agree with you that the "cockpit" games (racing, space combat, et al.), and things like Lucky's Tale are well served by the current controller, but I don't think it's going serve VR well outside of those niches. Which isn't what you want from the default, bundled controller.
Even just a few years ago most people were thinking that game genre popularity would just map into VR - i.e. Witchers and CoDs would be king. What's becoming more and more apparent is that VR is such a different medium that we need a whole new set game genres and mechanics.
For instance: locomotion with the left controller stick, head movement with the right (which the current controllers are brilliantly adapted for) is such a core controller "idiom" right now, but when applied to VR it's just horrible - a recipe for nausea and disorientation.
If you look at the kinds of experiences that are being developed around Valve's controllers, they're all huge departures from the games we currently play, and the reason is that Valve, as game designers, realized from an early point just how far back to the drawing board we need to go. VR is just too different, with different objectives (immersion), possible modes of interaction, and human weaknesses to address.
Because the entire VR experience is a summation of its immersive elements. Immersion in one domain (control, sound, display) can make up for immersive shortcomings in another domain.
My take: this a bandaid, so they can get the headset out in Q1, ahead of the proper controllers in Q2. Not what they wanted to do, but Vive forced their hand on timing.