This is the explanation closest to what I was thinking beforehand. The problem seems like one of the difference between {the difficulty of predicting an event} and {the likelihood of correctly reporting an observed event}. I think Dagon's argument about Map vs. Territory is a good one too.

Question for you, though... please define "ghost"? I have a feeling your definition is different than mine because I find events such as

certain environmental factors like (low-level poisoning from radon, carbon monoxide, et al; certain acoustic effects; certain architectural events such as uneven expansion due to temperature changes; &c.) cause minor hallucinations or illusions resulting in supersocial minds like those in humans perceiving "people" where there are none

very much more likely than

something of a person that exists independent of the usual corporeal form and (typically) despite the loss of that form is detectable by an uninformed and objective observer.

EDIT: formatted for better readability