what causes runout results like this?

[[Ma...]]

Howdy, was wondering if anyone knew of a reason as to why i get these weird curly results for runouts. the other side looks great. 
scan direction for feature is the same for datum cylinder. is this a projection issue? 

[[Ma...]]

Are you concerned about value or graphics?

Graphical issue is only magnification of deviations - can be changed in total runout characteristic.

[[Ma...]]

The value and graphics, I increased the error scaling for the post. "E side" looks great and looking at the manufactures report they seem to get a better result than i do (0.074) graphics also looks better. 
I would like to understand what the graphics are telling me when i see this loop. tends to happen often. Is this just normal or is something happening within?
Both bores are 200mm and 30mm depth same measurement strategy, using the same datum bore. 

Edited Friday at 08:30 PM

[[To...]]

Please sign in to view this username.

 I don't know if this is your problem but I see this happen whenever a probe shanks out at one point on the diameter.  However, I wouldn't necessarily see equal amounts on the and bottom. I would expect to see a greater deviation at the bottom.  I call that the "uvula" effect.  You won't see the uvula until you increase the magnification, like you did.

[[No...]]

The graph looks as if the actual geometry is shifted in the -X direction. The "loop" you marked is a natural result of the increased magnification. Only deviations are magnified, meaning that the deviation vectors are elongated. Beyond a certain mag. level the inner vectors begin to cross each other, reulting in this peculiar loop.