I'm running in circles with this issue. I have a polygon that I am buffering 55m. Originally I buffered it in a UTM projection with the planar algorithm. My map document was in the same projection. A co-worker viewed the buffered file in a map with a geographic coordinate system, and it was in a different place and overlapped an important feature that it does not overlap when the map document is in the UTM projection.
This polygon covers 2 counties and is pretty long. So I realized I should try doing Geodesic buffering. I projected the polygon to a geographic coordinate system, and buffered it with the geodesic algorithm. In a map in a geographic coordinate system, both the planar and geodesic buffers exactly overlap and display as such, and they overlap the point of interest. However, when the map is in the UTM projection the geodesic buffer stays the same but the UTM buffer, as mentioned, jumps.
My question is: why does the projected planar buffer jump when moved between a geographic and projected map coordinate system, or why doesn't the geodesic buffer jump? If the geodesic and planar buffers overlap exactly when the map is in a geographic coordinate system projection don't they share lat/lon values, and then shouldn't they skew the same when the map coordinate system changes?