Sets in slicer - analysis of behaviour

Chris Webb recently blogged about a strange phenomenon regarding sets in where clause and auto-exists. Today it occured to me that I had experienced a similar issue while doing some testing a year from now. A quick search on the right words (Tomislav Darren set slicer) and the exact thread popped up first in SSAS forum search engine.

After I've read it, I was very much intrigued by Chris' post, so I've read it and tested it too, and then decided to spend few hours in order to find out what's going on in the examples he provided, hoping to discover a general rule of this "weird" behavior and, most of all, the answer "why".

Whether I succeeded or not, I'll leave up to you. The explaination follows.