Yeah, I mean, a lot of socionics theory is indeed based in observations. That's where the descriptions of types and the IEs working together as a whole take place. So if you observe ppl and then have enough and good understanding of theory to fit your observations correctly into a type, the chances of typing correctly increases. Actually a lot of the associations I mention above are wrong theoretically and are more like
assumptions (like the own understanding of certain socionics concepts) which remove the original meaning and redefine it so things just twist.
That happens with a lot of fields, not just socionics.
Btw, werent you typing as IEE?
(or it was just my imagination?)