Agreed. This is analogous to saying that if you say a guy named Jesus walked the Levant in the first century preaching, one can accept that that very well may have happened, or even that it probably did, but when you throw in resurrection, well then no, that almost certainly didn't happen. When you say take her word for it, I think you mean assign a high probability for truth. The fraction of people who claim to have eaten a ham sandwich and did nothing like it or were mistaken about what they did do must be vanishingly small.If you claimed to eat a ham sandwich for lunch we might take your word for it. Ham sandwiches actually exist and sandwiches are a common thing to eat at lunch. But when the claim is something not known to exist or be true, like gods and messengers for gods, then that is an extraordinary claim and we require extraordinary evidence.
But unlike faith-based belief, we don't conclude that it (definitely) happened when we say we believe it. We believe it tentatively, always ready to revise that estimate if new evidence surfaces justifying we do that, such as the knowledge that the claimant was a pathological liar. Now, the likelihood falls to a lower number. Likewise, if we witness a bona fide resurrection from the dead, the odds on Christ's resurrection improve considerably from extremely unlikely to reasonably likely.
This is obvious and trivial to the critical thinker, but seems to elude many of our apologist friends, even though it comports perfectly with the definition of relevant evidence being anything that increases or decreases the likelihood of a proposition being correct or incorrect.