Your example is a special case, and it really is more about how you look at it. When you select an edge, then hit the “loop” command, the edgeloop works itself around the object until it hits the pole. The problem with this example, is that the pole is right at the end and the beginning of the loop. It’s an illusion. It is terminating at the pole, but because it started there, it will find the entire loop. I’m finding it difficult to explain, so please forgive the vagarities.
Try making another pole at the other end of that edge, and you will see that the loop won’t make it all the way around, as it will encounter a pole before it terminates at it’s origin. With this example, the loop does not begin and end at a pole.
Hope some of that made sense.
And for the record, a pole is a vertex with 3, or 5 or more edges. A vertex with 2 edges is an orphan. There is also one acception to the “pole stops edgeloop selection” rule. If the pole or poles exist along an open area, ie, around a hole, poles can sometimes be selected. Peter(Urgaffel) can fill you in more on that one.
Anyway, I hope that answered some of your questions. Welcome to the thread, BTW.