Firstly I generally think of information provided voluntarily as being more suspect than that which is elicited, coerced or happenstance. Think of the criminal who under questioning volunteers to "help" the cops by putting them on someone else's trail with a false accusation. (Made to seem offhand, of course.)

This applies to the stalking example too - in terms of possibly being able to frame and spin something by mentioning it first, when you know some version of it is going to come out regardless.

I agree with your point about written records, at least to the extent that writing behooves the writer to consider things carefully. Although it still doesn't preclude someone's being inept or unwise about it.

Just speaking generally, not necessarily about this case.

The story contains various propositions that can basically be divided into:

1. propositions that can likely be independently corroborated by large numbers of people:

--- the conference really happened (we are not told which conference where, but if we knew that, numerous people could be found to attest to it, if it had been real).

--- that the organizer was loudly accosted by the woman, accusing him of stalking: this is a public incident that supposedly happened, in front of witnesses.

--- the non grata couple is real; people don't like them for some reason and warn conference organizers

--- the non grata couple are known for harassing behavior

2. claims made by the organizer, like:

--- he denied participation in the conference to the couple

--- he was harassed by the woman with repeated contact attempts

--- the woman's allegations were false

The claims under (1) are verifiable true or false. If they are true, they lend overwhelming credibility to (2). Basically if everything under (1) is true, it's almost inconceivable that the (2) claims aren't.