If the piezo elements are designed so the casing is used as the ground contact the surely it would make more sense that they should be a nice snug fit to ensure contact is maintained?
The strings do hold the elements in place but it doesn't take a lot of lateral pressure on the strings to make the elerments move from side to side. Anyone who hits the strings quite hard will be exposed to the risk of them moving and causing a "click" as they hit the side of the casing.
The other problem is that the elements are not high enough above the casing to ensure the strings are clear of the edge of the casing and they also run quite close to the edge of the actual bridge itself. There are grooves cut into the casing but these are clearly not wide enough to allow for a standard low E width of 46. Most guitar saddles ensure plenty of clearance from the top of the saddle to any fixings in front and avoid any potential contact to the rest of the hardware on the bridge. IMO this is poor design and the source of many of the "plink", "plonk", "clang", "clank", digital artefact issues we see reported in the forum. It would seem that the only solutions are those carried out by katiekerry which are to use a needle file to widen the grooves in the front edge of the piezo element casings, and possibly to also file down the edge of the bridge below the low E to ensure no contact can be made even if strings are hit hard. And for the movement of the piezo's it would seem that you have to install ground wires and then wrap the element to prevent movement which is not an easy job and which may be beyond the abilities of many.
Thanks everyone for your input to this discussion.