This argument isn't really outside any boxes (and nor is it especially original), and it's also not really a “non-evidential” argument as claimed. The bottom line is that we have good grounds to accept reported experiences only when the probability that the report is false is not significantly greater than the prior probability of the experience; this rule arises from Bayes' theorem, but it's also the same principle given in informal terms by Hume:
No testimony is sufficient to establish a miracle unless it is of such a kind that its falsehood would be more miraculous than the fact that it tries to establish.
Dillon also misses another key point: we have the testimony in (probably) his own words of one very important early Christian—i.e. Paul—who reports seeing Jesus after his death, and it is clearly the description of a vision. We don't have any good reason to believe that this wasn't typical of early Christians. Even the author of Mark didn't see any need to include any physical post-resurrection experiences.