With a 15 degree tilt (top of the scale leaned towards me, with the standard axis zeroed)...
I assume you mean the scale was zero'd end-to-end (as normally done) but the whole apparatus tilted sideways toward you?
I admit that I'd expect the axis bearing surface to be a bit "scruffy and draggy", but I'm surprised that it introduced an actual bias. Why do you think that happened?
For myself, I use the RCBS Chargemaster 98% of the time -- left "ON" and calibrated with check weights before every session. (No appreciable drift so far -- remarkable). Whenever I get paranoid I pull out my Texan Balance