I have an FSR setup, but are using the oldstyle original arms, skates an effector mounts (no ball joints or magnets or anything). Probing works fine, an I get repeatable probes if I run them after eachother. Yesterday I was probing in the same conditions as the day before, but I get a difference of up to 0.1 mm that varies greatly from tower to tower. I used the same temp for bed and hotend. That's half a layer and thus far too much. So it looks like cooling down and heating up again makes a difference. Also I think I see some bed bulging after heating up (center seems a little bit higher after warming up.) To get all even I have to do endstopscrew tweaking again. Mind that I did not dismount the bed in the meantime. Is that strange or something to get used to? Now while probing I can measure these thing much better then before
Also I see some tilt when the nozzle hits the bed. There is play in the hinges that connect the arms to the effector, and probably the skates too. When the arm closest to the tower seems to have more play, and the effector is visibly tilting a bit. That can't be good! Will the newer ball cups arms better this? If so, I will have something to add to my birthday-present list.
One more question: Which option is better to countermeasure the probing flex? There is a "Z_PROBE_HEIGHT" and a "Z_PROBE_SURFACE_THICKNESS" (the last one I am not sure out of the top of my head, I am not at home...).
I am DeltaCon, I have a delta, my name is Con, I am definitely PRO delta!

Rostock V2 / E3D Volcano / FSR kit / Duet 0.6
PS.: Sorry for the avatar, that's my other hobby!