Lateral deflection of the binding when the ski is on-edge reduces its angulation, and some bindings deflect more readily than others. But how significant is this in actual skiing?
If you click your boot (not while wearing it) into a binding, and then anchor the ski so it doesn't move (say, put the ski on the kitchen floor and straddle the binding), you'll notice two things: (1) it's relatively easy to rock the boot laterally, by at least a few degrees; and (2) some bindings clearly resist this rocking better than others. I don't know how this translates into the kind of forces a ski experiences when it's up on edge, but it seems plausible this lateral compliance could result in a few degrees of angle loss (equivalent to canting in the soft direction) at the apex. This seems like it would be significant.
As a practical matter, I just acquired a pair of Supershape KERS and am trying to decide between an FF14 and an FF16. A contact at Head reports that the FF16 shows less lateral deflection than the 14 because of a stiffer housing (I've not confirmed this personally), but that the FF14 has the potential safety advantage of the diagonal heel and full-diagonal toe.