OK, I've been lazy. I just had a good look at the LTD Symmetrex arms as included in the Mk II versions of the WTL tables, and it appears that the differences are 1) the gasket has been changed to rubber from metal, and 2) the golf ball now has a metal cap that the filament threads through. These are the retrofitted arms, not the standalone, all-metal arms.
Assuming this is what rob400 has, my guess that the rubber gasket might improve azimuth drift might be actually be at play in rob400's experience of better consistency of azimuth (maybe). I'm not sure I can get my head around the VTA thing, but still pondering. This would have to mean that the golf ball lifts over 4.6mm (0.1833 inches) over the 10.5" arm to effect a VTA change of 1 degree. That's a pretty significant lift, one that I'm sure Mr. Firebaugh would have noticed and engineered out. Of course, even minute differences in VTA on the order of tenths of a degree are perceivable IME, but I think it would have to be a sudden change in VTA for it to be noticed (by me at least). Not ruling it out, however.
But back to the new LTD tech: I wonder whether the metal cap also raises the effective mass of the arm significantly - does anyone know whether this is true? If so, does the increased mass allow it to seat in the silicone better, increasing the effectiveness of the damping, lowering resonances further? Guess I can shoot Mr. P an email and get off my lazy a**.
These questions aside from the "dynamic tracking" feature...